From 15469cfc74037cdd6b46686368da4bcd28dc9c2f Mon Sep 17 00:00:00 2001 From: Liyulingyue <852433440@qq.com> Date: Mon, 4 Aug 2025 22:39:29 +0800 Subject: [PATCH 1/7] timm2 --- .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../model.py | 2206 +++++++++ .../weight_meta.py | 2074 ++++++++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../model.py | 4246 +++++++++++++++++ .../weight_meta.py | 4090 ++++++++++++++++ 12 files changed, 12628 insertions(+) create mode 100644 samples/timm/beit3_base_patch16_224.in22k_ft_in1k/graph_hash.txt create mode 100644 samples/timm/beit3_base_patch16_224.in22k_ft_in1k/graph_net.json create mode 100644 samples/timm/beit3_base_patch16_224.in22k_ft_in1k/input_meta.py create mode 100644 samples/timm/beit3_base_patch16_224.in22k_ft_in1k/input_tensor_constraints.py create mode 100644 samples/timm/beit3_base_patch16_224.in22k_ft_in1k/model.py create mode 100644 samples/timm/beit3_base_patch16_224.in22k_ft_in1k/weight_meta.py create mode 100644 samples/timm/beit3_large_patch16_224.in22k_ft_in1k/graph_hash.txt create mode 100644 samples/timm/beit3_large_patch16_224.in22k_ft_in1k/graph_net.json create mode 100644 samples/timm/beit3_large_patch16_224.in22k_ft_in1k/input_meta.py create mode 100644 samples/timm/beit3_large_patch16_224.in22k_ft_in1k/input_tensor_constraints.py create mode 100644 samples/timm/beit3_large_patch16_224.in22k_ft_in1k/model.py create mode 100644 samples/timm/beit3_large_patch16_224.in22k_ft_in1k/weight_meta.py diff --git a/samples/timm/beit3_base_patch16_224.in22k_ft_in1k/graph_hash.txt b/samples/timm/beit3_base_patch16_224.in22k_ft_in1k/graph_hash.txt new file mode 100644 index 0000000000..88da88ab8a --- /dev/null +++ b/samples/timm/beit3_base_patch16_224.in22k_ft_in1k/graph_hash.txt @@ -0,0 +1 @@ +49efabd40a93e856c10a3cf40aef7d7ad7e6f9fcfed3fc57f883ebd0e7d3f938 \ No newline at end of file diff --git a/samples/timm/beit3_base_patch16_224.in22k_ft_in1k/graph_net.json b/samples/timm/beit3_base_patch16_224.in22k_ft_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/beit3_base_patch16_224.in22k_ft_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/beit3_base_patch16_224.in22k_ft_in1k/input_meta.py b/samples/timm/beit3_base_patch16_224.in22k_ft_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/beit3_base_patch16_224.in22k_ft_in1k/input_tensor_constraints.py b/samples/timm/beit3_base_patch16_224.in22k_ft_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/beit3_base_patch16_224.in22k_ft_in1k/model.py b/samples/timm/beit3_base_patch16_224.in22k_ft_in1k/model.py new file mode 100644 index 0000000000..77beedcb74 --- /dev/null +++ b/samples/timm/beit3_base_patch16_224.in22k_ft_in1k/model.py @@ -0,0 +1,2206 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_num_prefix_tokens: torch.SymInt, + L_self_modules_fc_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_fc_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_num_prefix_tokens = L_self_num_prefix_tokens + l_self_modules_fc_norm_parameters_weight_ = ( + L_self_modules_fc_norm_parameters_weight_ + ) + l_self_modules_fc_norm_parameters_bias_ = ( + L_self_modules_fc_norm_parameters_bias_ + ) + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + x_2 = torch.cat([expand, x_1], dim=1) + expand = x_1 = None + x_3 = x_2 + l_self_parameters_pos_embed_ + x_2 = l_self_parameters_pos_embed_ = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (768,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 197, 3, 12, 64) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 197, 768) + transpose_1 = None + x_8 = torch.nn.functional.layer_norm( + x_7, + (768,), + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_ + ) = None + x_9 = torch._C._nn.linear( + x_8, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_8 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = x_4 + x_10 + x_4 = x_10 = None + x_12 = torch.nn.functional.layer_norm( + x_11, + (768,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_13 = torch._C._nn.linear( + x_12, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_12 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_14 = torch._C._nn.gelu(x_13, approximate="none") + x_13 = None + x_15 = torch.nn.functional.dropout(x_14, 0.0, False, False) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (3072,), + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_15 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_16 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_18 = torch.nn.functional.dropout(x_17, 0.0, False, False) + x_17 = None + x_19 = x_11 + x_18 + x_11 = x_18 = None + x_20 = torch.nn.functional.layer_norm( + x_19, + (768,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 197, 3, 12, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_21 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_21.transpose(1, 2) + x_21 = None + x_22 = transpose_2.reshape(1, 197, 768) + transpose_2 = None + x_23 = torch.nn.functional.layer_norm( + x_22, + (768,), + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_22 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_ + ) = None + x_24 = torch._C._nn.linear( + x_23, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_23 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_25 = torch.nn.functional.dropout(x_24, 0.0, False, False) + x_24 = None + x_26 = x_19 + x_25 + x_19 = x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (768,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_29 = torch._C._nn.gelu(x_28, approximate="none") + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (3072,), + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_30 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_32 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + x_34 = x_26 + x_33 + x_26 = x_33 = None + x_35 = torch.nn.functional.layer_norm( + x_34, + (768,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_35, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_35 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 197, 3, 12, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_36 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_36.transpose(1, 2) + x_36 = None + x_37 = transpose_3.reshape(1, 197, 768) + transpose_3 = None + x_38 = torch.nn.functional.layer_norm( + x_37, + (768,), + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_ + ) = None + x_39 = torch._C._nn.linear( + x_38, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_38 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = x_34 + x_40 + x_34 = x_40 = None + x_42 = torch.nn.functional.layer_norm( + x_41, + (768,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_43 = torch._C._nn.linear( + x_42, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_42 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_44 = torch._C._nn.gelu(x_43, approximate="none") + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = torch.nn.functional.layer_norm( + x_45, + (3072,), + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_45 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_41 + x_48 + x_41 = x_48 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (768,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 197, 3, 12, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_51 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_51.transpose(1, 2) + x_51 = None + x_52 = transpose_4.reshape(1, 197, 768) + transpose_4 = None + x_53 = torch.nn.functional.layer_norm( + x_52, + (768,), + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_52 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_ + ) = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = x_49 + x_55 + x_49 = x_55 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (768,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_58 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_59 = torch._C._nn.gelu(x_58, approximate="none") + x_58 = None + x_60 = torch.nn.functional.dropout(x_59, 0.0, False, False) + x_59 = None + x_61 = torch.nn.functional.layer_norm( + x_60, + (3072,), + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_60 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_62 = torch._C._nn.linear( + x_61, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_61 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_63 = torch.nn.functional.dropout(x_62, 0.0, False, False) + x_62 = None + x_64 = x_56 + x_63 + x_56 = x_63 = None + x_65 = torch.nn.functional.layer_norm( + x_64, + (768,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_65, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_65 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 197, 3, 12, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_66 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_66.transpose(1, 2) + x_66 = None + x_67 = transpose_5.reshape(1, 197, 768) + transpose_5 = None + x_68 = torch.nn.functional.layer_norm( + x_67, + (768,), + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_67 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_ + ) = None + x_69 = torch._C._nn.linear( + x_68, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_68 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_71 = x_64 + x_70 + x_64 = x_70 = None + x_72 = torch.nn.functional.layer_norm( + x_71, + (768,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_74 = torch._C._nn.gelu(x_73, approximate="none") + x_73 = None + x_75 = torch.nn.functional.dropout(x_74, 0.0, False, False) + x_74 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (3072,), + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_75 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_78 = torch.nn.functional.dropout(x_77, 0.0, False, False) + x_77 = None + x_79 = x_71 + x_78 + x_71 = x_78 = None + x_80 = torch.nn.functional.layer_norm( + x_79, + (768,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_80, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_80 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 197, 3, 12, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_81 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_81.transpose(1, 2) + x_81 = None + x_82 = transpose_6.reshape(1, 197, 768) + transpose_6 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (768,), + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_82 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_ + ) = None + x_84 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_85 = torch.nn.functional.dropout(x_84, 0.0, False, False) + x_84 = None + x_86 = x_79 + x_85 + x_79 = x_85 = None + x_87 = torch.nn.functional.layer_norm( + x_86, + (768,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_88 = torch._C._nn.linear( + x_87, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_87 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_89 = torch._C._nn.gelu(x_88, approximate="none") + x_88 = None + x_90 = torch.nn.functional.dropout(x_89, 0.0, False, False) + x_89 = None + x_91 = torch.nn.functional.layer_norm( + x_90, + (3072,), + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_90 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_92 = torch._C._nn.linear( + x_91, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_91 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_93 = torch.nn.functional.dropout(x_92, 0.0, False, False) + x_92 = None + x_94 = x_86 + x_93 + x_86 = x_93 = None + x_95 = torch.nn.functional.layer_norm( + x_94, + (768,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_95, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_95 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 197, 3, 12, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_96 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_96.transpose(1, 2) + x_96 = None + x_97 = transpose_7.reshape(1, 197, 768) + transpose_7 = None + x_98 = torch.nn.functional.layer_norm( + x_97, + (768,), + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_97 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_ + ) = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + x_101 = x_94 + x_100 + x_94 = x_100 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (768,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch.nn.functional.layer_norm( + x_105, + (3072,), + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_105 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_107 = torch._C._nn.linear( + x_106, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_106 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_101 + x_108 + x_101 = x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (768,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_110, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_110 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 197, 3, 12, 64) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_111 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_111.transpose(1, 2) + x_111 = None + x_112 = transpose_8.reshape(1, 197, 768) + transpose_8 = None + x_113 = torch.nn.functional.layer_norm( + x_112, + (768,), + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_112 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_ + ) = None + x_114 = torch._C._nn.linear( + x_113, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_113 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_115 = torch.nn.functional.dropout(x_114, 0.0, False, False) + x_114 = None + x_116 = x_109 + x_115 + x_109 = x_115 = None + x_117 = torch.nn.functional.layer_norm( + x_116, + (768,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_118 = torch._C._nn.linear( + x_117, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_117 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_119 = torch._C._nn.gelu(x_118, approximate="none") + x_118 = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (3072,), + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_120 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_121 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) + x_122 = None + x_124 = x_116 + x_123 + x_116 = x_123 = None + x_125 = torch.nn.functional.layer_norm( + x_124, + (768,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_125, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_125 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 197, 3, 12, 64) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_126 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_126.transpose(1, 2) + x_126 = None + x_127 = transpose_9.reshape(1, 197, 768) + transpose_9 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (768,), + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_127 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) + x_129 = None + x_131 = x_124 + x_130 + x_124 = x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (768,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_134 = torch._C._nn.gelu(x_133, approximate="none") + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = torch.nn.functional.layer_norm( + x_135, + (3072,), + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_135 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_131 + x_138 + x_131 = x_138 = None + x_140 = torch.nn.functional.layer_norm( + x_139, + (768,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_140, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_140 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 197, 3, 12, 64) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_141 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_141.transpose(1, 2) + x_141 = None + x_142 = transpose_10.reshape(1, 197, 768) + transpose_10 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (768,), + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_142 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_ + ) = None + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_143 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_145 = torch.nn.functional.dropout(x_144, 0.0, False, False) + x_144 = None + x_146 = x_139 + x_145 + x_139 = x_145 = None + x_147 = torch.nn.functional.layer_norm( + x_146, + (768,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_147 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_149 = torch._C._nn.gelu(x_148, approximate="none") + x_148 = None + x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) + x_149 = None + x_151 = torch.nn.functional.layer_norm( + x_150, + (3072,), + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_150 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_152 = torch._C._nn.linear( + x_151, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_151 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_153 = torch.nn.functional.dropout(x_152, 0.0, False, False) + x_152 = None + x_154 = x_146 + x_153 + x_146 = x_153 = None + x_155 = torch.nn.functional.layer_norm( + x_154, + (768,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_155, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_155 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 197, 3, 12, 64) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_156 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_156.transpose(1, 2) + x_156 = None + x_157 = transpose_11.reshape(1, 197, 768) + transpose_11 = None + x_158 = torch.nn.functional.layer_norm( + x_157, + (768,), + l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_157 = l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_ + ) = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_158 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_154 + x_160 + x_154 = x_160 = None + x_162 = torch.nn.functional.layer_norm( + x_161, + (768,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_163 = torch._C._nn.linear( + x_162, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_162 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_164 = torch._C._nn.gelu(x_163, approximate="none") + x_163 = None + x_165 = torch.nn.functional.dropout(x_164, 0.0, False, False) + x_164 = None + x_166 = torch.nn.functional.layer_norm( + x_165, + (3072,), + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_165 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_167 = torch._C._nn.linear( + x_166, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_166 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_169 = x_161 + x_168 + x_161 = x_168 = None + x_170 = torch.nn.functional.layer_norm( + x_169, + (768,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_170, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_170 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 197, 3, 12, 64) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_171 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_171.transpose(1, 2) + x_171 = None + x_172 = transpose_12.reshape(1, 197, 768) + transpose_12 = None + x_173 = torch.nn.functional.layer_norm( + x_172, + (768,), + l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_172 = l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_ + ) = None + x_174 = torch._C._nn.linear( + x_173, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_173 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_175 = torch.nn.functional.dropout(x_174, 0.0, False, False) + x_174 = None + x_176 = x_169 + x_175 + x_169 = x_175 = None + x_177 = torch.nn.functional.layer_norm( + x_176, + (768,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_178 = torch._C._nn.linear( + x_177, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_177 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_179 = torch._C._nn.gelu(x_178, approximate="none") + x_178 = None + x_180 = torch.nn.functional.dropout(x_179, 0.0, False, False) + x_179 = None + x_181 = torch.nn.functional.layer_norm( + x_180, + (3072,), + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_180 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_182 = torch._C._nn.linear( + x_181, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_181 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_183 = torch.nn.functional.dropout(x_182, 0.0, False, False) + x_182 = None + x_184 = x_176 + x_183 + x_176 = x_183 = None + x_185 = x_184[ + (slice(None, None, None), slice(l_self_num_prefix_tokens, None, None)) + ] + x_184 = l_self_num_prefix_tokens = None + x_186 = x_185.mean(dim=1) + x_185 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (768,), + l_self_modules_fc_norm_parameters_weight_, + l_self_modules_fc_norm_parameters_bias_, + 1e-05, + ) + x_186 = ( + l_self_modules_fc_norm_parameters_weight_ + ) = l_self_modules_fc_norm_parameters_bias_ = None + x_188 = torch.nn.functional.dropout(x_187, 0.0, False, False) + x_187 = None + x_189 = torch._C._nn.linear( + x_188, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_188 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_189,) diff --git a/samples/timm/beit3_base_patch16_224.in22k_ft_in1k/weight_meta.py b/samples/timm/beit3_base_patch16_224.in22k_ft_in1k/weight_meta.py new file mode 100644 index 0000000000..54a373afa3 --- /dev/null +++ b/samples/timm/beit3_base_patch16_224.in22k_ft_in1k/weight_meta.py @@ -0,0 +1,2074 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.229 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [768, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 197, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_num_prefix_tokens: + name = "L_self_num_prefix_tokens" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_self_modules_fc_norm_parameters_weight_: + name = "L_self_modules_fc_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_fc_norm_parameters_bias_: + name = "L_self_modules_fc_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/beit3_large_patch16_224.in22k_ft_in1k/graph_hash.txt b/samples/timm/beit3_large_patch16_224.in22k_ft_in1k/graph_hash.txt new file mode 100644 index 0000000000..44b074954d --- /dev/null +++ b/samples/timm/beit3_large_patch16_224.in22k_ft_in1k/graph_hash.txt @@ -0,0 +1 @@ +a79caac738b040f293c01cd92e1770f312f059371fd757fa138f608c37b3cf6e \ No newline at end of file diff --git a/samples/timm/beit3_large_patch16_224.in22k_ft_in1k/graph_net.json b/samples/timm/beit3_large_patch16_224.in22k_ft_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/beit3_large_patch16_224.in22k_ft_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/beit3_large_patch16_224.in22k_ft_in1k/input_meta.py b/samples/timm/beit3_large_patch16_224.in22k_ft_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/beit3_large_patch16_224.in22k_ft_in1k/input_tensor_constraints.py b/samples/timm/beit3_large_patch16_224.in22k_ft_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/beit3_large_patch16_224.in22k_ft_in1k/model.py b/samples/timm/beit3_large_patch16_224.in22k_ft_in1k/model.py new file mode 100644 index 0000000000..c424406a3e --- /dev/null +++ b/samples/timm/beit3_large_patch16_224.in22k_ft_in1k/model.py @@ -0,0 +1,4246 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_num_prefix_tokens: torch.SymInt, + L_self_modules_fc_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_fc_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_num_prefix_tokens = L_self_num_prefix_tokens + l_self_modules_fc_norm_parameters_weight_ = ( + L_self_modules_fc_norm_parameters_weight_ + ) + l_self_modules_fc_norm_parameters_bias_ = ( + L_self_modules_fc_norm_parameters_bias_ + ) + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + x_2 = torch.cat([expand, x_1], dim=1) + expand = x_1 = None + x_3 = x_2 + l_self_parameters_pos_embed_ + x_2 = l_self_parameters_pos_embed_ = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (1024,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 197, 3, 16, 64) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 197, 1024) + transpose_1 = None + x_8 = torch.nn.functional.layer_norm( + x_7, + (1024,), + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_ + ) = None + x_9 = torch._C._nn.linear( + x_8, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_8 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = x_4 + x_10 + x_4 = x_10 = None + x_12 = torch.nn.functional.layer_norm( + x_11, + (1024,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_13 = torch._C._nn.linear( + x_12, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_12 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_14 = torch._C._nn.gelu(x_13, approximate="none") + x_13 = None + x_15 = torch.nn.functional.dropout(x_14, 0.0, False, False) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (4096,), + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_15 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_16 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_18 = torch.nn.functional.dropout(x_17, 0.0, False, False) + x_17 = None + x_19 = x_11 + x_18 + x_11 = x_18 = None + x_20 = torch.nn.functional.layer_norm( + x_19, + (1024,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 197, 3, 16, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_21 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_21.transpose(1, 2) + x_21 = None + x_22 = transpose_2.reshape(1, 197, 1024) + transpose_2 = None + x_23 = torch.nn.functional.layer_norm( + x_22, + (1024,), + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_22 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_ + ) = None + x_24 = torch._C._nn.linear( + x_23, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_23 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_25 = torch.nn.functional.dropout(x_24, 0.0, False, False) + x_24 = None + x_26 = x_19 + x_25 + x_19 = x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (1024,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_29 = torch._C._nn.gelu(x_28, approximate="none") + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (4096,), + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_30 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_32 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + x_34 = x_26 + x_33 + x_26 = x_33 = None + x_35 = torch.nn.functional.layer_norm( + x_34, + (1024,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_35, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_35 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 197, 3, 16, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_36 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_36.transpose(1, 2) + x_36 = None + x_37 = transpose_3.reshape(1, 197, 1024) + transpose_3 = None + x_38 = torch.nn.functional.layer_norm( + x_37, + (1024,), + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_ + ) = None + x_39 = torch._C._nn.linear( + x_38, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_38 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = x_34 + x_40 + x_34 = x_40 = None + x_42 = torch.nn.functional.layer_norm( + x_41, + (1024,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_43 = torch._C._nn.linear( + x_42, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_42 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_44 = torch._C._nn.gelu(x_43, approximate="none") + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = torch.nn.functional.layer_norm( + x_45, + (4096,), + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_45 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_41 + x_48 + x_41 = x_48 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (1024,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 197, 3, 16, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_51 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_51.transpose(1, 2) + x_51 = None + x_52 = transpose_4.reshape(1, 197, 1024) + transpose_4 = None + x_53 = torch.nn.functional.layer_norm( + x_52, + (1024,), + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_52 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_ + ) = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = x_49 + x_55 + x_49 = x_55 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (1024,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_58 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_59 = torch._C._nn.gelu(x_58, approximate="none") + x_58 = None + x_60 = torch.nn.functional.dropout(x_59, 0.0, False, False) + x_59 = None + x_61 = torch.nn.functional.layer_norm( + x_60, + (4096,), + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_60 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_62 = torch._C._nn.linear( + x_61, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_61 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_63 = torch.nn.functional.dropout(x_62, 0.0, False, False) + x_62 = None + x_64 = x_56 + x_63 + x_56 = x_63 = None + x_65 = torch.nn.functional.layer_norm( + x_64, + (1024,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_65, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_65 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 197, 3, 16, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_66 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_66.transpose(1, 2) + x_66 = None + x_67 = transpose_5.reshape(1, 197, 1024) + transpose_5 = None + x_68 = torch.nn.functional.layer_norm( + x_67, + (1024,), + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_67 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_ + ) = None + x_69 = torch._C._nn.linear( + x_68, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_68 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_71 = x_64 + x_70 + x_64 = x_70 = None + x_72 = torch.nn.functional.layer_norm( + x_71, + (1024,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_74 = torch._C._nn.gelu(x_73, approximate="none") + x_73 = None + x_75 = torch.nn.functional.dropout(x_74, 0.0, False, False) + x_74 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (4096,), + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_75 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_78 = torch.nn.functional.dropout(x_77, 0.0, False, False) + x_77 = None + x_79 = x_71 + x_78 + x_71 = x_78 = None + x_80 = torch.nn.functional.layer_norm( + x_79, + (1024,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_80, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_80 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 197, 3, 16, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_81 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_81.transpose(1, 2) + x_81 = None + x_82 = transpose_6.reshape(1, 197, 1024) + transpose_6 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (1024,), + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_82 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_ + ) = None + x_84 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_85 = torch.nn.functional.dropout(x_84, 0.0, False, False) + x_84 = None + x_86 = x_79 + x_85 + x_79 = x_85 = None + x_87 = torch.nn.functional.layer_norm( + x_86, + (1024,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_88 = torch._C._nn.linear( + x_87, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_87 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_89 = torch._C._nn.gelu(x_88, approximate="none") + x_88 = None + x_90 = torch.nn.functional.dropout(x_89, 0.0, False, False) + x_89 = None + x_91 = torch.nn.functional.layer_norm( + x_90, + (4096,), + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_90 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_92 = torch._C._nn.linear( + x_91, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_91 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_93 = torch.nn.functional.dropout(x_92, 0.0, False, False) + x_92 = None + x_94 = x_86 + x_93 + x_86 = x_93 = None + x_95 = torch.nn.functional.layer_norm( + x_94, + (1024,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_95, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_95 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 197, 3, 16, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_96 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_96.transpose(1, 2) + x_96 = None + x_97 = transpose_7.reshape(1, 197, 1024) + transpose_7 = None + x_98 = torch.nn.functional.layer_norm( + x_97, + (1024,), + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_97 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_ + ) = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + x_101 = x_94 + x_100 + x_94 = x_100 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (1024,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch.nn.functional.layer_norm( + x_105, + (4096,), + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_105 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_107 = torch._C._nn.linear( + x_106, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_106 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_101 + x_108 + x_101 = x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (1024,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_110, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_110 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 197, 3, 16, 64) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_111 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_111.transpose(1, 2) + x_111 = None + x_112 = transpose_8.reshape(1, 197, 1024) + transpose_8 = None + x_113 = torch.nn.functional.layer_norm( + x_112, + (1024,), + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_112 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_ + ) = None + x_114 = torch._C._nn.linear( + x_113, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_113 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_115 = torch.nn.functional.dropout(x_114, 0.0, False, False) + x_114 = None + x_116 = x_109 + x_115 + x_109 = x_115 = None + x_117 = torch.nn.functional.layer_norm( + x_116, + (1024,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_118 = torch._C._nn.linear( + x_117, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_117 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_119 = torch._C._nn.gelu(x_118, approximate="none") + x_118 = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (4096,), + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_120 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_121 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) + x_122 = None + x_124 = x_116 + x_123 + x_116 = x_123 = None + x_125 = torch.nn.functional.layer_norm( + x_124, + (1024,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_125, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_125 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 197, 3, 16, 64) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_126 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_126.transpose(1, 2) + x_126 = None + x_127 = transpose_9.reshape(1, 197, 1024) + transpose_9 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (1024,), + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_127 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) + x_129 = None + x_131 = x_124 + x_130 + x_124 = x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (1024,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_134 = torch._C._nn.gelu(x_133, approximate="none") + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = torch.nn.functional.layer_norm( + x_135, + (4096,), + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_135 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_131 + x_138 + x_131 = x_138 = None + x_140 = torch.nn.functional.layer_norm( + x_139, + (1024,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_140, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_140 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 197, 3, 16, 64) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_141 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_141.transpose(1, 2) + x_141 = None + x_142 = transpose_10.reshape(1, 197, 1024) + transpose_10 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (1024,), + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_142 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_ + ) = None + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_143 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_145 = torch.nn.functional.dropout(x_144, 0.0, False, False) + x_144 = None + x_146 = x_139 + x_145 + x_139 = x_145 = None + x_147 = torch.nn.functional.layer_norm( + x_146, + (1024,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_147 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_149 = torch._C._nn.gelu(x_148, approximate="none") + x_148 = None + x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) + x_149 = None + x_151 = torch.nn.functional.layer_norm( + x_150, + (4096,), + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_150 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_152 = torch._C._nn.linear( + x_151, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_151 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_153 = torch.nn.functional.dropout(x_152, 0.0, False, False) + x_152 = None + x_154 = x_146 + x_153 + x_146 = x_153 = None + x_155 = torch.nn.functional.layer_norm( + x_154, + (1024,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_155, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_155 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 197, 3, 16, 64) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_156 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_156.transpose(1, 2) + x_156 = None + x_157 = transpose_11.reshape(1, 197, 1024) + transpose_11 = None + x_158 = torch.nn.functional.layer_norm( + x_157, + (1024,), + l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_157 = l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_ + ) = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_158 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_154 + x_160 + x_154 = x_160 = None + x_162 = torch.nn.functional.layer_norm( + x_161, + (1024,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_163 = torch._C._nn.linear( + x_162, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_162 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_164 = torch._C._nn.gelu(x_163, approximate="none") + x_163 = None + x_165 = torch.nn.functional.dropout(x_164, 0.0, False, False) + x_164 = None + x_166 = torch.nn.functional.layer_norm( + x_165, + (4096,), + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_165 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_167 = torch._C._nn.linear( + x_166, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_166 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_169 = x_161 + x_168 + x_161 = x_168 = None + x_170 = torch.nn.functional.layer_norm( + x_169, + (1024,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_170, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_170 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 197, 3, 16, 64) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_171 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_171.transpose(1, 2) + x_171 = None + x_172 = transpose_12.reshape(1, 197, 1024) + transpose_12 = None + x_173 = torch.nn.functional.layer_norm( + x_172, + (1024,), + l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_172 = l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_ + ) = None + x_174 = torch._C._nn.linear( + x_173, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_173 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_175 = torch.nn.functional.dropout(x_174, 0.0, False, False) + x_174 = None + x_176 = x_169 + x_175 + x_169 = x_175 = None + x_177 = torch.nn.functional.layer_norm( + x_176, + (1024,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_178 = torch._C._nn.linear( + x_177, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_177 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_179 = torch._C._nn.gelu(x_178, approximate="none") + x_178 = None + x_180 = torch.nn.functional.dropout(x_179, 0.0, False, False) + x_179 = None + x_181 = torch.nn.functional.layer_norm( + x_180, + (4096,), + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_180 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_182 = torch._C._nn.linear( + x_181, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_181 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_183 = torch.nn.functional.dropout(x_182, 0.0, False, False) + x_182 = None + x_184 = x_176 + x_183 + x_176 = x_183 = None + x_185 = torch.nn.functional.layer_norm( + x_184, + (1024,), + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ + ) = None + linear_48 = torch._C._nn.linear( + x_185, + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_, + ) + x_185 = ( + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_24 = linear_48.reshape(1, 197, 3, 16, 64) + linear_48 = None + qkv_12 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_12 = unbind_12[0] + k_12 = unbind_12[1] + v_12 = unbind_12[2] + unbind_12 = None + x_186 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_13 = x_186.transpose(1, 2) + x_186 = None + x_187 = transpose_13.reshape(1, 197, 1024) + transpose_13 = None + x_188 = torch.nn.functional.layer_norm( + x_187, + (1024,), + l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_187 = l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_ + ) = None + x_189 = torch._C._nn.linear( + x_188, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_, + ) + x_188 = l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ + ) = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = x_184 + x_190 + x_184 = x_190 = None + x_192 = torch.nn.functional.layer_norm( + x_191, + (1024,), + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ + ) = None + x_193 = torch._C._nn.linear( + x_192, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_192 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_194 = torch._C._nn.gelu(x_193, approximate="none") + x_193 = None + x_195 = torch.nn.functional.dropout(x_194, 0.0, False, False) + x_194 = None + x_196 = torch.nn.functional.layer_norm( + x_195, + (4096,), + l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_195 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_197 = torch._C._nn.linear( + x_196, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_196 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_198 = torch.nn.functional.dropout(x_197, 0.0, False, False) + x_197 = None + x_199 = x_191 + x_198 + x_191 = x_198 = None + x_200 = torch.nn.functional.layer_norm( + x_199, + (1024,), + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ + ) = None + linear_52 = torch._C._nn.linear( + x_200, + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_, + ) + x_200 = ( + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_26 = linear_52.reshape(1, 197, 3, 16, 64) + linear_52 = None + qkv_13 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_13 = unbind_13[0] + k_13 = unbind_13[1] + v_13 = unbind_13[2] + unbind_13 = None + x_201 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_14 = x_201.transpose(1, 2) + x_201 = None + x_202 = transpose_14.reshape(1, 197, 1024) + transpose_14 = None + x_203 = torch.nn.functional.layer_norm( + x_202, + (1024,), + l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_202 = l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_ + ) = None + x_204 = torch._C._nn.linear( + x_203, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_, + ) + x_203 = l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ + ) = None + x_205 = torch.nn.functional.dropout(x_204, 0.0, False, False) + x_204 = None + x_206 = x_199 + x_205 + x_199 = x_205 = None + x_207 = torch.nn.functional.layer_norm( + x_206, + (1024,), + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ + ) = None + x_208 = torch._C._nn.linear( + x_207, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_207 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_209 = torch._C._nn.gelu(x_208, approximate="none") + x_208 = None + x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) + x_209 = None + x_211 = torch.nn.functional.layer_norm( + x_210, + (4096,), + l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_210 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_212 = torch._C._nn.linear( + x_211, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_211 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_213 = torch.nn.functional.dropout(x_212, 0.0, False, False) + x_212 = None + x_214 = x_206 + x_213 + x_206 = x_213 = None + x_215 = torch.nn.functional.layer_norm( + x_214, + (1024,), + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ + ) = None + linear_56 = torch._C._nn.linear( + x_215, + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_, + ) + x_215 = ( + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_28 = linear_56.reshape(1, 197, 3, 16, 64) + linear_56 = None + qkv_14 = reshape_28.permute(2, 0, 3, 1, 4) + reshape_28 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_14 = unbind_14[0] + k_14 = unbind_14[1] + v_14 = unbind_14[2] + unbind_14 = None + x_216 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_15 = x_216.transpose(1, 2) + x_216 = None + x_217 = transpose_15.reshape(1, 197, 1024) + transpose_15 = None + x_218 = torch.nn.functional.layer_norm( + x_217, + (1024,), + l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_217 = l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_ + ) = None + x_219 = torch._C._nn.linear( + x_218, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_, + ) + x_218 = l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ + ) = None + x_220 = torch.nn.functional.dropout(x_219, 0.0, False, False) + x_219 = None + x_221 = x_214 + x_220 + x_214 = x_220 = None + x_222 = torch.nn.functional.layer_norm( + x_221, + (1024,), + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ + ) = None + x_223 = torch._C._nn.linear( + x_222, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_222 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_224 = torch._C._nn.gelu(x_223, approximate="none") + x_223 = None + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = torch.nn.functional.layer_norm( + x_225, + (4096,), + l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_225 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_227 = torch._C._nn.linear( + x_226, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_226 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_228 = torch.nn.functional.dropout(x_227, 0.0, False, False) + x_227 = None + x_229 = x_221 + x_228 + x_221 = x_228 = None + x_230 = torch.nn.functional.layer_norm( + x_229, + (1024,), + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ + ) = None + linear_60 = torch._C._nn.linear( + x_230, + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_, + ) + x_230 = ( + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_30 = linear_60.reshape(1, 197, 3, 16, 64) + linear_60 = None + qkv_15 = reshape_30.permute(2, 0, 3, 1, 4) + reshape_30 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_15 = unbind_15[0] + k_15 = unbind_15[1] + v_15 = unbind_15[2] + unbind_15 = None + x_231 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_16 = x_231.transpose(1, 2) + x_231 = None + x_232 = transpose_16.reshape(1, 197, 1024) + transpose_16 = None + x_233 = torch.nn.functional.layer_norm( + x_232, + (1024,), + l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_232 = l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_ + ) = None + x_234 = torch._C._nn.linear( + x_233, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_, + ) + x_233 = l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ + ) = None + x_235 = torch.nn.functional.dropout(x_234, 0.0, False, False) + x_234 = None + x_236 = x_229 + x_235 + x_229 = x_235 = None + x_237 = torch.nn.functional.layer_norm( + x_236, + (1024,), + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ + ) = None + x_238 = torch._C._nn.linear( + x_237, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_237 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_239 = torch._C._nn.gelu(x_238, approximate="none") + x_238 = None + x_240 = torch.nn.functional.dropout(x_239, 0.0, False, False) + x_239 = None + x_241 = torch.nn.functional.layer_norm( + x_240, + (4096,), + l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_240 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_242 = torch._C._nn.linear( + x_241, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_241 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_243 = torch.nn.functional.dropout(x_242, 0.0, False, False) + x_242 = None + x_244 = x_236 + x_243 + x_236 = x_243 = None + x_245 = torch.nn.functional.layer_norm( + x_244, + (1024,), + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ + ) = None + linear_64 = torch._C._nn.linear( + x_245, + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_, + ) + x_245 = ( + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_32 = linear_64.reshape(1, 197, 3, 16, 64) + linear_64 = None + qkv_16 = reshape_32.permute(2, 0, 3, 1, 4) + reshape_32 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_16 = unbind_16[0] + k_16 = unbind_16[1] + v_16 = unbind_16[2] + unbind_16 = None + x_246 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, attn_mask=None, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_17 = x_246.transpose(1, 2) + x_246 = None + x_247 = transpose_17.reshape(1, 197, 1024) + transpose_17 = None + x_248 = torch.nn.functional.layer_norm( + x_247, + (1024,), + l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_247 = l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_ + ) = None + x_249 = torch._C._nn.linear( + x_248, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_, + ) + x_248 = l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ + ) = None + x_250 = torch.nn.functional.dropout(x_249, 0.0, False, False) + x_249 = None + x_251 = x_244 + x_250 + x_244 = x_250 = None + x_252 = torch.nn.functional.layer_norm( + x_251, + (1024,), + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ + ) = None + x_253 = torch._C._nn.linear( + x_252, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_252 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_254 = torch._C._nn.gelu(x_253, approximate="none") + x_253 = None + x_255 = torch.nn.functional.dropout(x_254, 0.0, False, False) + x_254 = None + x_256 = torch.nn.functional.layer_norm( + x_255, + (4096,), + l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_255 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_257 = torch._C._nn.linear( + x_256, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_256 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_258 = torch.nn.functional.dropout(x_257, 0.0, False, False) + x_257 = None + x_259 = x_251 + x_258 + x_251 = x_258 = None + x_260 = torch.nn.functional.layer_norm( + x_259, + (1024,), + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ + ) = None + linear_68 = torch._C._nn.linear( + x_260, + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_, + ) + x_260 = ( + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_34 = linear_68.reshape(1, 197, 3, 16, 64) + linear_68 = None + qkv_17 = reshape_34.permute(2, 0, 3, 1, 4) + reshape_34 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_17 = unbind_17[0] + k_17 = unbind_17[1] + v_17 = unbind_17[2] + unbind_17 = None + x_261 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, attn_mask=None, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_18 = x_261.transpose(1, 2) + x_261 = None + x_262 = transpose_18.reshape(1, 197, 1024) + transpose_18 = None + x_263 = torch.nn.functional.layer_norm( + x_262, + (1024,), + l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_262 = l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_ + ) = None + x_264 = torch._C._nn.linear( + x_263, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_, + ) + x_263 = l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ + ) = None + x_265 = torch.nn.functional.dropout(x_264, 0.0, False, False) + x_264 = None + x_266 = x_259 + x_265 + x_259 = x_265 = None + x_267 = torch.nn.functional.layer_norm( + x_266, + (1024,), + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ + ) = None + x_268 = torch._C._nn.linear( + x_267, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_267 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_269 = torch._C._nn.gelu(x_268, approximate="none") + x_268 = None + x_270 = torch.nn.functional.dropout(x_269, 0.0, False, False) + x_269 = None + x_271 = torch.nn.functional.layer_norm( + x_270, + (4096,), + l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_270 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_272 = torch._C._nn.linear( + x_271, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_271 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_273 = torch.nn.functional.dropout(x_272, 0.0, False, False) + x_272 = None + x_274 = x_266 + x_273 + x_266 = x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (1024,), + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ + ) = None + linear_72 = torch._C._nn.linear( + x_275, + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_, + ) + x_275 = ( + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_36 = linear_72.reshape(1, 197, 3, 16, 64) + linear_72 = None + qkv_18 = reshape_36.permute(2, 0, 3, 1, 4) + reshape_36 = None + unbind_18 = qkv_18.unbind(0) + qkv_18 = None + q_18 = unbind_18[0] + k_18 = unbind_18[1] + v_18 = unbind_18[2] + unbind_18 = None + x_276 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_19 = x_276.transpose(1, 2) + x_276 = None + x_277 = transpose_19.reshape(1, 197, 1024) + transpose_19 = None + x_278 = torch.nn.functional.layer_norm( + x_277, + (1024,), + l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_277 = l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_ + ) = None + x_279 = torch._C._nn.linear( + x_278, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_, + ) + x_278 = l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ + ) = None + x_280 = torch.nn.functional.dropout(x_279, 0.0, False, False) + x_279 = None + x_281 = x_274 + x_280 + x_274 = x_280 = None + x_282 = torch.nn.functional.layer_norm( + x_281, + (1024,), + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ + ) = None + x_283 = torch._C._nn.linear( + x_282, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_282 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_284 = torch._C._nn.gelu(x_283, approximate="none") + x_283 = None + x_285 = torch.nn.functional.dropout(x_284, 0.0, False, False) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (4096,), + l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_285 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_286 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) + x_287 = None + x_289 = x_281 + x_288 + x_281 = x_288 = None + x_290 = torch.nn.functional.layer_norm( + x_289, + (1024,), + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ + ) = None + linear_76 = torch._C._nn.linear( + x_290, + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_, + ) + x_290 = ( + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_38 = linear_76.reshape(1, 197, 3, 16, 64) + linear_76 = None + qkv_19 = reshape_38.permute(2, 0, 3, 1, 4) + reshape_38 = None + unbind_19 = qkv_19.unbind(0) + qkv_19 = None + q_19 = unbind_19[0] + k_19 = unbind_19[1] + v_19 = unbind_19[2] + unbind_19 = None + x_291 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, attn_mask=None, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_20 = x_291.transpose(1, 2) + x_291 = None + x_292 = transpose_20.reshape(1, 197, 1024) + transpose_20 = None + x_293 = torch.nn.functional.layer_norm( + x_292, + (1024,), + l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_292 = l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_ + ) = None + x_294 = torch._C._nn.linear( + x_293, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_, + ) + x_293 = l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ + ) = None + x_295 = torch.nn.functional.dropout(x_294, 0.0, False, False) + x_294 = None + x_296 = x_289 + x_295 + x_289 = x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (1024,), + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ + ) = None + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_299 = torch._C._nn.gelu(x_298, approximate="none") + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_301 = torch.nn.functional.layer_norm( + x_300, + (4096,), + l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_300 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_302 = torch._C._nn.linear( + x_301, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_301 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_303 = torch.nn.functional.dropout(x_302, 0.0, False, False) + x_302 = None + x_304 = x_296 + x_303 + x_296 = x_303 = None + x_305 = torch.nn.functional.layer_norm( + x_304, + (1024,), + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ + ) = None + linear_80 = torch._C._nn.linear( + x_305, + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_, + ) + x_305 = ( + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_40 = linear_80.reshape(1, 197, 3, 16, 64) + linear_80 = None + qkv_20 = reshape_40.permute(2, 0, 3, 1, 4) + reshape_40 = None + unbind_20 = qkv_20.unbind(0) + qkv_20 = None + q_20 = unbind_20[0] + k_20 = unbind_20[1] + v_20 = unbind_20[2] + unbind_20 = None + x_306 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, attn_mask=None, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_21 = x_306.transpose(1, 2) + x_306 = None + x_307 = transpose_21.reshape(1, 197, 1024) + transpose_21 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (1024,), + l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_307 = l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_ + ) = None + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_, + ) + x_308 = l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ + ) = None + x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) + x_309 = None + x_311 = x_304 + x_310 + x_304 = x_310 = None + x_312 = torch.nn.functional.layer_norm( + x_311, + (1024,), + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ + ) = None + x_313 = torch._C._nn.linear( + x_312, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_312 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_314 = torch._C._nn.gelu(x_313, approximate="none") + x_313 = None + x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) + x_314 = None + x_316 = torch.nn.functional.layer_norm( + x_315, + (4096,), + l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_315 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_317 = torch._C._nn.linear( + x_316, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_316 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_318 = torch.nn.functional.dropout(x_317, 0.0, False, False) + x_317 = None + x_319 = x_311 + x_318 + x_311 = x_318 = None + x_320 = torch.nn.functional.layer_norm( + x_319, + (1024,), + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ + ) = None + linear_84 = torch._C._nn.linear( + x_320, + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_, + ) + x_320 = ( + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_42 = linear_84.reshape(1, 197, 3, 16, 64) + linear_84 = None + qkv_21 = reshape_42.permute(2, 0, 3, 1, 4) + reshape_42 = None + unbind_21 = qkv_21.unbind(0) + qkv_21 = None + q_21 = unbind_21[0] + k_21 = unbind_21[1] + v_21 = unbind_21[2] + unbind_21 = None + x_321 = torch._C._nn.scaled_dot_product_attention( + q_21, k_21, v_21, attn_mask=None, dropout_p=0.0 + ) + q_21 = k_21 = v_21 = None + transpose_22 = x_321.transpose(1, 2) + x_321 = None + x_322 = transpose_22.reshape(1, 197, 1024) + transpose_22 = None + x_323 = torch.nn.functional.layer_norm( + x_322, + (1024,), + l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_322 = l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_ + ) = None + x_324 = torch._C._nn.linear( + x_323, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_, + ) + x_323 = l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ + ) = None + x_325 = torch.nn.functional.dropout(x_324, 0.0, False, False) + x_324 = None + x_326 = x_319 + x_325 + x_319 = x_325 = None + x_327 = torch.nn.functional.layer_norm( + x_326, + (1024,), + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ + ) = None + x_328 = torch._C._nn.linear( + x_327, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_327 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_329 = torch._C._nn.gelu(x_328, approximate="none") + x_328 = None + x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) + x_329 = None + x_331 = torch.nn.functional.layer_norm( + x_330, + (4096,), + l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_330 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_332 = torch._C._nn.linear( + x_331, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_331 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_334 = x_326 + x_333 + x_326 = x_333 = None + x_335 = torch.nn.functional.layer_norm( + x_334, + (1024,), + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ + ) = None + linear_88 = torch._C._nn.linear( + x_335, + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_, + ) + x_335 = ( + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_44 = linear_88.reshape(1, 197, 3, 16, 64) + linear_88 = None + qkv_22 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_22 = qkv_22.unbind(0) + qkv_22 = None + q_22 = unbind_22[0] + k_22 = unbind_22[1] + v_22 = unbind_22[2] + unbind_22 = None + x_336 = torch._C._nn.scaled_dot_product_attention( + q_22, k_22, v_22, attn_mask=None, dropout_p=0.0 + ) + q_22 = k_22 = v_22 = None + transpose_23 = x_336.transpose(1, 2) + x_336 = None + x_337 = transpose_23.reshape(1, 197, 1024) + transpose_23 = None + x_338 = torch.nn.functional.layer_norm( + x_337, + (1024,), + l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_337 = l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_ + ) = None + x_339 = torch._C._nn.linear( + x_338, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_, + ) + x_338 = l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ + ) = None + x_340 = torch.nn.functional.dropout(x_339, 0.0, False, False) + x_339 = None + x_341 = x_334 + x_340 + x_334 = x_340 = None + x_342 = torch.nn.functional.layer_norm( + x_341, + (1024,), + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ + ) = None + x_343 = torch._C._nn.linear( + x_342, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_342 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_344 = torch._C._nn.gelu(x_343, approximate="none") + x_343 = None + x_345 = torch.nn.functional.dropout(x_344, 0.0, False, False) + x_344 = None + x_346 = torch.nn.functional.layer_norm( + x_345, + (4096,), + l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_345 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_347 = torch._C._nn.linear( + x_346, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_346 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) + x_347 = None + x_349 = x_341 + x_348 + x_341 = x_348 = None + x_350 = torch.nn.functional.layer_norm( + x_349, + (1024,), + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ + ) = None + linear_92 = torch._C._nn.linear( + x_350, + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_, + ) + x_350 = ( + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_46 = linear_92.reshape(1, 197, 3, 16, 64) + linear_92 = None + qkv_23 = reshape_46.permute(2, 0, 3, 1, 4) + reshape_46 = None + unbind_23 = qkv_23.unbind(0) + qkv_23 = None + q_23 = unbind_23[0] + k_23 = unbind_23[1] + v_23 = unbind_23[2] + unbind_23 = None + x_351 = torch._C._nn.scaled_dot_product_attention( + q_23, k_23, v_23, attn_mask=None, dropout_p=0.0 + ) + q_23 = k_23 = v_23 = None + transpose_24 = x_351.transpose(1, 2) + x_351 = None + x_352 = transpose_24.reshape(1, 197, 1024) + transpose_24 = None + x_353 = torch.nn.functional.layer_norm( + x_352, + (1024,), + l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_352 = l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_ + ) = None + x_354 = torch._C._nn.linear( + x_353, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_, + ) + x_353 = l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ + ) = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_356 = x_349 + x_355 + x_349 = x_355 = None + x_357 = torch.nn.functional.layer_norm( + x_356, + (1024,), + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ + ) = None + x_358 = torch._C._nn.linear( + x_357, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_357 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_359 = torch._C._nn.gelu(x_358, approximate="none") + x_358 = None + x_360 = torch.nn.functional.dropout(x_359, 0.0, False, False) + x_359 = None + x_361 = torch.nn.functional.layer_norm( + x_360, + (4096,), + l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_360 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_362 = torch._C._nn.linear( + x_361, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_361 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) + x_362 = None + x_364 = x_356 + x_363 + x_356 = x_363 = None + x_365 = x_364[ + (slice(None, None, None), slice(l_self_num_prefix_tokens, None, None)) + ] + x_364 = l_self_num_prefix_tokens = None + x_366 = x_365.mean(dim=1) + x_365 = None + x_367 = torch.nn.functional.layer_norm( + x_366, + (1024,), + l_self_modules_fc_norm_parameters_weight_, + l_self_modules_fc_norm_parameters_bias_, + 1e-05, + ) + x_366 = ( + l_self_modules_fc_norm_parameters_weight_ + ) = l_self_modules_fc_norm_parameters_bias_ = None + x_368 = torch.nn.functional.dropout(x_367, 0.0, False, False) + x_367 = None + x_369 = torch._C._nn.linear( + x_368, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_368 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_369,) diff --git a/samples/timm/beit3_large_patch16_224.in22k_ft_in1k/weight_meta.py b/samples/timm/beit3_large_patch16_224.in22k_ft_in1k/weight_meta.py new file mode 100644 index 0000000000..d61e903b7c --- /dev/null +++ b/samples/timm/beit3_large_patch16_224.in22k_ft_in1k/weight_meta.py @@ -0,0 +1,4090 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.221 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [1024, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 197, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_num_prefix_tokens: + name = "L_self_num_prefix_tokens" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_self_modules_fc_norm_parameters_weight_: + name = "L_self_modules_fc_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_fc_norm_parameters_bias_: + name = "L_self_modules_fc_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None From d27510d8e96a2a2c4ec0199dababe6aa8fca114d Mon Sep 17 00:00:00 2001 From: Liyulingyue <852433440@qq.com> Date: Mon, 4 Aug 2025 22:50:26 +0800 Subject: [PATCH 2/7] timm group --- .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../model.py | 4246 +++++++++++++++ .../weight_meta.py | 4090 ++++++++++++++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../model.py | 4234 +++++++++++++++ .../weight_meta.py | 4070 ++++++++++++++ .../beit3_large_patch16_224.pt/graph_hash.txt | 1 + .../beit3_large_patch16_224.pt/graph_net.json | 5 + .../beit3_large_patch16_224.pt/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/beit3_large_patch16_224.pt/model.py | 4234 +++++++++++++++ .../beit3_large_patch16_224.pt/weight_meta.py | 4070 ++++++++++++++ .../caformer_b36.sail_in1k/graph_net.json | 5 + .../timm/caformer_b36.sail_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/caformer_b36.sail_in1k/model.py | 4758 +++++++++++++++++ .../caformer_b36.sail_in1k/weight_meta.py | 3968 ++++++++++++++ .../caformer_b36.sail_in1k_384/graph_net.json | 5 + .../caformer_b36.sail_in1k_384/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/caformer_b36.sail_in1k_384/model.py | 4758 +++++++++++++++++ .../caformer_b36.sail_in1k_384/weight_meta.py | 3968 ++++++++++++++ .../caformer_b36.sail_in22k/graph_net.json | 5 + .../caformer_b36.sail_in22k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/caformer_b36.sail_in22k/model.py | 4758 +++++++++++++++++ .../caformer_b36.sail_in22k/weight_meta.py | 3968 ++++++++++++++ .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../caformer_b36.sail_in22k_ft_in1k/model.py | 4758 +++++++++++++++++ .../weight_meta.py | 3968 ++++++++++++++ .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../model.py | 4758 +++++++++++++++++ .../weight_meta.py | 3968 ++++++++++++++ .../caformer_m36.sail_in1k/graph_net.json | 5 + .../timm/caformer_m36.sail_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/caformer_m36.sail_in1k/model.py | 4754 ++++++++++++++++ .../caformer_m36.sail_in1k/weight_meta.py | 3968 ++++++++++++++ .../caformer_m36.sail_in1k_384/graph_net.json | 5 + .../caformer_m36.sail_in1k_384/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/caformer_m36.sail_in1k_384/model.py | 4754 ++++++++++++++++ .../caformer_m36.sail_in1k_384/weight_meta.py | 3968 ++++++++++++++ .../caformer_m36.sail_in22k/graph_net.json | 5 + .../caformer_m36.sail_in22k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/caformer_m36.sail_in22k/model.py | 4754 ++++++++++++++++ .../caformer_m36.sail_in22k/weight_meta.py | 3968 ++++++++++++++ .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../caformer_m36.sail_in22k_ft_in1k/model.py | 4754 ++++++++++++++++ .../weight_meta.py | 3968 ++++++++++++++ .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../model.py | 4754 ++++++++++++++++ .../weight_meta.py | 3968 ++++++++++++++ .../caformer_s18.sail_in1k/graph_net.json | 5 + .../timm/caformer_s18.sail_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/caformer_s18.sail_in1k/model.py | 2468 +++++++++ .../caformer_s18.sail_in1k/weight_meta.py | 2078 +++++++ .../caformer_s18.sail_in1k_384/graph_net.json | 5 + .../caformer_s18.sail_in1k_384/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/caformer_s18.sail_in1k_384/model.py | 2468 +++++++++ .../caformer_s18.sail_in1k_384/weight_meta.py | 2078 +++++++ .../caformer_s18.sail_in22k/graph_net.json | 5 + .../caformer_s18.sail_in22k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/caformer_s18.sail_in22k/model.py | 2468 +++++++++ .../caformer_s18.sail_in22k/weight_meta.py | 2078 +++++++ timm_success_list.txt | 0 84 files changed, 125905 insertions(+) create mode 100644 samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/graph_hash.txt create mode 100644 samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/graph_net.json create mode 100644 samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/input_meta.py create mode 100644 samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/input_tensor_constraints.py create mode 100644 samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/model.py create mode 100644 samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/weight_meta.py create mode 100644 samples/timm/beit3_large_patch16_224.indomain_pt/graph_hash.txt create mode 100644 samples/timm/beit3_large_patch16_224.indomain_pt/graph_net.json create mode 100644 samples/timm/beit3_large_patch16_224.indomain_pt/input_meta.py create mode 100644 samples/timm/beit3_large_patch16_224.indomain_pt/input_tensor_constraints.py create mode 100644 samples/timm/beit3_large_patch16_224.indomain_pt/model.py create mode 100644 samples/timm/beit3_large_patch16_224.indomain_pt/weight_meta.py create mode 100644 samples/timm/beit3_large_patch16_224.pt/graph_hash.txt create mode 100644 samples/timm/beit3_large_patch16_224.pt/graph_net.json create mode 100644 samples/timm/beit3_large_patch16_224.pt/input_meta.py create mode 100644 samples/timm/beit3_large_patch16_224.pt/input_tensor_constraints.py create mode 100644 samples/timm/beit3_large_patch16_224.pt/model.py create mode 100644 samples/timm/beit3_large_patch16_224.pt/weight_meta.py create mode 100644 samples/timm/caformer_b36.sail_in1k/graph_net.json create mode 100644 samples/timm/caformer_b36.sail_in1k/input_meta.py create mode 100644 samples/timm/caformer_b36.sail_in1k/input_tensor_constraints.py create mode 100644 samples/timm/caformer_b36.sail_in1k/model.py create mode 100644 samples/timm/caformer_b36.sail_in1k/weight_meta.py create mode 100644 samples/timm/caformer_b36.sail_in1k_384/graph_net.json create mode 100644 samples/timm/caformer_b36.sail_in1k_384/input_meta.py create mode 100644 samples/timm/caformer_b36.sail_in1k_384/input_tensor_constraints.py create mode 100644 samples/timm/caformer_b36.sail_in1k_384/model.py create mode 100644 samples/timm/caformer_b36.sail_in1k_384/weight_meta.py create mode 100644 samples/timm/caformer_b36.sail_in22k/graph_net.json create mode 100644 samples/timm/caformer_b36.sail_in22k/input_meta.py create mode 100644 samples/timm/caformer_b36.sail_in22k/input_tensor_constraints.py create mode 100644 samples/timm/caformer_b36.sail_in22k/model.py create mode 100644 samples/timm/caformer_b36.sail_in22k/weight_meta.py create mode 100644 samples/timm/caformer_b36.sail_in22k_ft_in1k/graph_net.json create mode 100644 samples/timm/caformer_b36.sail_in22k_ft_in1k/input_meta.py create mode 100644 samples/timm/caformer_b36.sail_in22k_ft_in1k/input_tensor_constraints.py create mode 100644 samples/timm/caformer_b36.sail_in22k_ft_in1k/model.py create mode 100644 samples/timm/caformer_b36.sail_in22k_ft_in1k/weight_meta.py create mode 100644 samples/timm/caformer_b36.sail_in22k_ft_in1k_384/graph_net.json create mode 100644 samples/timm/caformer_b36.sail_in22k_ft_in1k_384/input_meta.py create mode 100644 samples/timm/caformer_b36.sail_in22k_ft_in1k_384/input_tensor_constraints.py create mode 100644 samples/timm/caformer_b36.sail_in22k_ft_in1k_384/model.py create mode 100644 samples/timm/caformer_b36.sail_in22k_ft_in1k_384/weight_meta.py create mode 100644 samples/timm/caformer_m36.sail_in1k/graph_net.json create mode 100644 samples/timm/caformer_m36.sail_in1k/input_meta.py create mode 100644 samples/timm/caformer_m36.sail_in1k/input_tensor_constraints.py create mode 100644 samples/timm/caformer_m36.sail_in1k/model.py create mode 100644 samples/timm/caformer_m36.sail_in1k/weight_meta.py create mode 100644 samples/timm/caformer_m36.sail_in1k_384/graph_net.json create mode 100644 samples/timm/caformer_m36.sail_in1k_384/input_meta.py create mode 100644 samples/timm/caformer_m36.sail_in1k_384/input_tensor_constraints.py create mode 100644 samples/timm/caformer_m36.sail_in1k_384/model.py create mode 100644 samples/timm/caformer_m36.sail_in1k_384/weight_meta.py create mode 100644 samples/timm/caformer_m36.sail_in22k/graph_net.json create mode 100644 samples/timm/caformer_m36.sail_in22k/input_meta.py create mode 100644 samples/timm/caformer_m36.sail_in22k/input_tensor_constraints.py create mode 100644 samples/timm/caformer_m36.sail_in22k/model.py create mode 100644 samples/timm/caformer_m36.sail_in22k/weight_meta.py create mode 100644 samples/timm/caformer_m36.sail_in22k_ft_in1k/graph_net.json create mode 100644 samples/timm/caformer_m36.sail_in22k_ft_in1k/input_meta.py create mode 100644 samples/timm/caformer_m36.sail_in22k_ft_in1k/input_tensor_constraints.py create mode 100644 samples/timm/caformer_m36.sail_in22k_ft_in1k/model.py create mode 100644 samples/timm/caformer_m36.sail_in22k_ft_in1k/weight_meta.py create mode 100644 samples/timm/caformer_m36.sail_in22k_ft_in1k_384/graph_net.json create mode 100644 samples/timm/caformer_m36.sail_in22k_ft_in1k_384/input_meta.py create mode 100644 samples/timm/caformer_m36.sail_in22k_ft_in1k_384/input_tensor_constraints.py create mode 100644 samples/timm/caformer_m36.sail_in22k_ft_in1k_384/model.py create mode 100644 samples/timm/caformer_m36.sail_in22k_ft_in1k_384/weight_meta.py create mode 100644 samples/timm/caformer_s18.sail_in1k/graph_net.json create mode 100644 samples/timm/caformer_s18.sail_in1k/input_meta.py create mode 100644 samples/timm/caformer_s18.sail_in1k/input_tensor_constraints.py create mode 100644 samples/timm/caformer_s18.sail_in1k/model.py create mode 100644 samples/timm/caformer_s18.sail_in1k/weight_meta.py create mode 100644 samples/timm/caformer_s18.sail_in1k_384/graph_net.json create mode 100644 samples/timm/caformer_s18.sail_in1k_384/input_meta.py create mode 100644 samples/timm/caformer_s18.sail_in1k_384/input_tensor_constraints.py create mode 100644 samples/timm/caformer_s18.sail_in1k_384/model.py create mode 100644 samples/timm/caformer_s18.sail_in1k_384/weight_meta.py create mode 100644 samples/timm/caformer_s18.sail_in22k/graph_net.json create mode 100644 samples/timm/caformer_s18.sail_in22k/input_meta.py create mode 100644 samples/timm/caformer_s18.sail_in22k/input_tensor_constraints.py create mode 100644 samples/timm/caformer_s18.sail_in22k/model.py create mode 100644 samples/timm/caformer_s18.sail_in22k/weight_meta.py create mode 100644 timm_success_list.txt diff --git a/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/graph_hash.txt b/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/graph_hash.txt new file mode 100644 index 0000000000..44b074954d --- /dev/null +++ b/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/graph_hash.txt @@ -0,0 +1 @@ +a79caac738b040f293c01cd92e1770f312f059371fd757fa138f608c37b3cf6e \ No newline at end of file diff --git a/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/graph_net.json b/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/input_meta.py b/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/input_tensor_constraints.py b/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/model.py b/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/model.py new file mode 100644 index 0000000000..c424406a3e --- /dev/null +++ b/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/model.py @@ -0,0 +1,4246 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_num_prefix_tokens: torch.SymInt, + L_self_modules_fc_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_fc_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_num_prefix_tokens = L_self_num_prefix_tokens + l_self_modules_fc_norm_parameters_weight_ = ( + L_self_modules_fc_norm_parameters_weight_ + ) + l_self_modules_fc_norm_parameters_bias_ = ( + L_self_modules_fc_norm_parameters_bias_ + ) + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + x_2 = torch.cat([expand, x_1], dim=1) + expand = x_1 = None + x_3 = x_2 + l_self_parameters_pos_embed_ + x_2 = l_self_parameters_pos_embed_ = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (1024,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 197, 3, 16, 64) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 197, 1024) + transpose_1 = None + x_8 = torch.nn.functional.layer_norm( + x_7, + (1024,), + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_ + ) = None + x_9 = torch._C._nn.linear( + x_8, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_8 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = x_4 + x_10 + x_4 = x_10 = None + x_12 = torch.nn.functional.layer_norm( + x_11, + (1024,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_13 = torch._C._nn.linear( + x_12, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_12 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_14 = torch._C._nn.gelu(x_13, approximate="none") + x_13 = None + x_15 = torch.nn.functional.dropout(x_14, 0.0, False, False) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (4096,), + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_15 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_16 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_18 = torch.nn.functional.dropout(x_17, 0.0, False, False) + x_17 = None + x_19 = x_11 + x_18 + x_11 = x_18 = None + x_20 = torch.nn.functional.layer_norm( + x_19, + (1024,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 197, 3, 16, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_21 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_21.transpose(1, 2) + x_21 = None + x_22 = transpose_2.reshape(1, 197, 1024) + transpose_2 = None + x_23 = torch.nn.functional.layer_norm( + x_22, + (1024,), + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_22 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_ + ) = None + x_24 = torch._C._nn.linear( + x_23, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_23 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_25 = torch.nn.functional.dropout(x_24, 0.0, False, False) + x_24 = None + x_26 = x_19 + x_25 + x_19 = x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (1024,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_29 = torch._C._nn.gelu(x_28, approximate="none") + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (4096,), + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_30 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_32 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + x_34 = x_26 + x_33 + x_26 = x_33 = None + x_35 = torch.nn.functional.layer_norm( + x_34, + (1024,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_35, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_35 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 197, 3, 16, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_36 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_36.transpose(1, 2) + x_36 = None + x_37 = transpose_3.reshape(1, 197, 1024) + transpose_3 = None + x_38 = torch.nn.functional.layer_norm( + x_37, + (1024,), + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_ + ) = None + x_39 = torch._C._nn.linear( + x_38, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_38 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = x_34 + x_40 + x_34 = x_40 = None + x_42 = torch.nn.functional.layer_norm( + x_41, + (1024,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_43 = torch._C._nn.linear( + x_42, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_42 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_44 = torch._C._nn.gelu(x_43, approximate="none") + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = torch.nn.functional.layer_norm( + x_45, + (4096,), + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_45 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_41 + x_48 + x_41 = x_48 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (1024,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 197, 3, 16, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_51 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_51.transpose(1, 2) + x_51 = None + x_52 = transpose_4.reshape(1, 197, 1024) + transpose_4 = None + x_53 = torch.nn.functional.layer_norm( + x_52, + (1024,), + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_52 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_ + ) = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = x_49 + x_55 + x_49 = x_55 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (1024,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_58 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_59 = torch._C._nn.gelu(x_58, approximate="none") + x_58 = None + x_60 = torch.nn.functional.dropout(x_59, 0.0, False, False) + x_59 = None + x_61 = torch.nn.functional.layer_norm( + x_60, + (4096,), + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_60 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_62 = torch._C._nn.linear( + x_61, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_61 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_63 = torch.nn.functional.dropout(x_62, 0.0, False, False) + x_62 = None + x_64 = x_56 + x_63 + x_56 = x_63 = None + x_65 = torch.nn.functional.layer_norm( + x_64, + (1024,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_65, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_65 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 197, 3, 16, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_66 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_66.transpose(1, 2) + x_66 = None + x_67 = transpose_5.reshape(1, 197, 1024) + transpose_5 = None + x_68 = torch.nn.functional.layer_norm( + x_67, + (1024,), + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_67 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_ + ) = None + x_69 = torch._C._nn.linear( + x_68, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_68 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_71 = x_64 + x_70 + x_64 = x_70 = None + x_72 = torch.nn.functional.layer_norm( + x_71, + (1024,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_74 = torch._C._nn.gelu(x_73, approximate="none") + x_73 = None + x_75 = torch.nn.functional.dropout(x_74, 0.0, False, False) + x_74 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (4096,), + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_75 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_78 = torch.nn.functional.dropout(x_77, 0.0, False, False) + x_77 = None + x_79 = x_71 + x_78 + x_71 = x_78 = None + x_80 = torch.nn.functional.layer_norm( + x_79, + (1024,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_80, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_80 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 197, 3, 16, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_81 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_81.transpose(1, 2) + x_81 = None + x_82 = transpose_6.reshape(1, 197, 1024) + transpose_6 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (1024,), + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_82 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_ + ) = None + x_84 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_85 = torch.nn.functional.dropout(x_84, 0.0, False, False) + x_84 = None + x_86 = x_79 + x_85 + x_79 = x_85 = None + x_87 = torch.nn.functional.layer_norm( + x_86, + (1024,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_88 = torch._C._nn.linear( + x_87, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_87 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_89 = torch._C._nn.gelu(x_88, approximate="none") + x_88 = None + x_90 = torch.nn.functional.dropout(x_89, 0.0, False, False) + x_89 = None + x_91 = torch.nn.functional.layer_norm( + x_90, + (4096,), + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_90 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_92 = torch._C._nn.linear( + x_91, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_91 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_93 = torch.nn.functional.dropout(x_92, 0.0, False, False) + x_92 = None + x_94 = x_86 + x_93 + x_86 = x_93 = None + x_95 = torch.nn.functional.layer_norm( + x_94, + (1024,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_95, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_95 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 197, 3, 16, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_96 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_96.transpose(1, 2) + x_96 = None + x_97 = transpose_7.reshape(1, 197, 1024) + transpose_7 = None + x_98 = torch.nn.functional.layer_norm( + x_97, + (1024,), + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_97 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_ + ) = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + x_101 = x_94 + x_100 + x_94 = x_100 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (1024,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch.nn.functional.layer_norm( + x_105, + (4096,), + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_105 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_107 = torch._C._nn.linear( + x_106, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_106 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_101 + x_108 + x_101 = x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (1024,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_110, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_110 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 197, 3, 16, 64) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_111 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_111.transpose(1, 2) + x_111 = None + x_112 = transpose_8.reshape(1, 197, 1024) + transpose_8 = None + x_113 = torch.nn.functional.layer_norm( + x_112, + (1024,), + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_112 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_ + ) = None + x_114 = torch._C._nn.linear( + x_113, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_113 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_115 = torch.nn.functional.dropout(x_114, 0.0, False, False) + x_114 = None + x_116 = x_109 + x_115 + x_109 = x_115 = None + x_117 = torch.nn.functional.layer_norm( + x_116, + (1024,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_118 = torch._C._nn.linear( + x_117, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_117 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_119 = torch._C._nn.gelu(x_118, approximate="none") + x_118 = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (4096,), + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_120 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_121 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) + x_122 = None + x_124 = x_116 + x_123 + x_116 = x_123 = None + x_125 = torch.nn.functional.layer_norm( + x_124, + (1024,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_125, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_125 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 197, 3, 16, 64) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_126 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_126.transpose(1, 2) + x_126 = None + x_127 = transpose_9.reshape(1, 197, 1024) + transpose_9 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (1024,), + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_127 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) + x_129 = None + x_131 = x_124 + x_130 + x_124 = x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (1024,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_134 = torch._C._nn.gelu(x_133, approximate="none") + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = torch.nn.functional.layer_norm( + x_135, + (4096,), + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_135 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_131 + x_138 + x_131 = x_138 = None + x_140 = torch.nn.functional.layer_norm( + x_139, + (1024,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_140, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_140 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 197, 3, 16, 64) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_141 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_141.transpose(1, 2) + x_141 = None + x_142 = transpose_10.reshape(1, 197, 1024) + transpose_10 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (1024,), + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_142 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_ + ) = None + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_143 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_145 = torch.nn.functional.dropout(x_144, 0.0, False, False) + x_144 = None + x_146 = x_139 + x_145 + x_139 = x_145 = None + x_147 = torch.nn.functional.layer_norm( + x_146, + (1024,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_147 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_149 = torch._C._nn.gelu(x_148, approximate="none") + x_148 = None + x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) + x_149 = None + x_151 = torch.nn.functional.layer_norm( + x_150, + (4096,), + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_150 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_152 = torch._C._nn.linear( + x_151, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_151 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_153 = torch.nn.functional.dropout(x_152, 0.0, False, False) + x_152 = None + x_154 = x_146 + x_153 + x_146 = x_153 = None + x_155 = torch.nn.functional.layer_norm( + x_154, + (1024,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_155, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_155 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 197, 3, 16, 64) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_156 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_156.transpose(1, 2) + x_156 = None + x_157 = transpose_11.reshape(1, 197, 1024) + transpose_11 = None + x_158 = torch.nn.functional.layer_norm( + x_157, + (1024,), + l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_157 = l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_ + ) = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_158 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_154 + x_160 + x_154 = x_160 = None + x_162 = torch.nn.functional.layer_norm( + x_161, + (1024,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_163 = torch._C._nn.linear( + x_162, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_162 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_164 = torch._C._nn.gelu(x_163, approximate="none") + x_163 = None + x_165 = torch.nn.functional.dropout(x_164, 0.0, False, False) + x_164 = None + x_166 = torch.nn.functional.layer_norm( + x_165, + (4096,), + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_165 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_167 = torch._C._nn.linear( + x_166, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_166 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_169 = x_161 + x_168 + x_161 = x_168 = None + x_170 = torch.nn.functional.layer_norm( + x_169, + (1024,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_170, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_170 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 197, 3, 16, 64) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_171 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_171.transpose(1, 2) + x_171 = None + x_172 = transpose_12.reshape(1, 197, 1024) + transpose_12 = None + x_173 = torch.nn.functional.layer_norm( + x_172, + (1024,), + l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_172 = l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_ + ) = None + x_174 = torch._C._nn.linear( + x_173, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_173 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_175 = torch.nn.functional.dropout(x_174, 0.0, False, False) + x_174 = None + x_176 = x_169 + x_175 + x_169 = x_175 = None + x_177 = torch.nn.functional.layer_norm( + x_176, + (1024,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_178 = torch._C._nn.linear( + x_177, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_177 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_179 = torch._C._nn.gelu(x_178, approximate="none") + x_178 = None + x_180 = torch.nn.functional.dropout(x_179, 0.0, False, False) + x_179 = None + x_181 = torch.nn.functional.layer_norm( + x_180, + (4096,), + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_180 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_182 = torch._C._nn.linear( + x_181, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_181 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_183 = torch.nn.functional.dropout(x_182, 0.0, False, False) + x_182 = None + x_184 = x_176 + x_183 + x_176 = x_183 = None + x_185 = torch.nn.functional.layer_norm( + x_184, + (1024,), + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ + ) = None + linear_48 = torch._C._nn.linear( + x_185, + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_, + ) + x_185 = ( + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_24 = linear_48.reshape(1, 197, 3, 16, 64) + linear_48 = None + qkv_12 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_12 = unbind_12[0] + k_12 = unbind_12[1] + v_12 = unbind_12[2] + unbind_12 = None + x_186 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_13 = x_186.transpose(1, 2) + x_186 = None + x_187 = transpose_13.reshape(1, 197, 1024) + transpose_13 = None + x_188 = torch.nn.functional.layer_norm( + x_187, + (1024,), + l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_187 = l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_ + ) = None + x_189 = torch._C._nn.linear( + x_188, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_, + ) + x_188 = l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ + ) = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = x_184 + x_190 + x_184 = x_190 = None + x_192 = torch.nn.functional.layer_norm( + x_191, + (1024,), + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ + ) = None + x_193 = torch._C._nn.linear( + x_192, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_192 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_194 = torch._C._nn.gelu(x_193, approximate="none") + x_193 = None + x_195 = torch.nn.functional.dropout(x_194, 0.0, False, False) + x_194 = None + x_196 = torch.nn.functional.layer_norm( + x_195, + (4096,), + l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_195 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_197 = torch._C._nn.linear( + x_196, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_196 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_198 = torch.nn.functional.dropout(x_197, 0.0, False, False) + x_197 = None + x_199 = x_191 + x_198 + x_191 = x_198 = None + x_200 = torch.nn.functional.layer_norm( + x_199, + (1024,), + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ + ) = None + linear_52 = torch._C._nn.linear( + x_200, + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_, + ) + x_200 = ( + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_26 = linear_52.reshape(1, 197, 3, 16, 64) + linear_52 = None + qkv_13 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_13 = unbind_13[0] + k_13 = unbind_13[1] + v_13 = unbind_13[2] + unbind_13 = None + x_201 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_14 = x_201.transpose(1, 2) + x_201 = None + x_202 = transpose_14.reshape(1, 197, 1024) + transpose_14 = None + x_203 = torch.nn.functional.layer_norm( + x_202, + (1024,), + l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_202 = l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_ + ) = None + x_204 = torch._C._nn.linear( + x_203, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_, + ) + x_203 = l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ + ) = None + x_205 = torch.nn.functional.dropout(x_204, 0.0, False, False) + x_204 = None + x_206 = x_199 + x_205 + x_199 = x_205 = None + x_207 = torch.nn.functional.layer_norm( + x_206, + (1024,), + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ + ) = None + x_208 = torch._C._nn.linear( + x_207, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_207 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_209 = torch._C._nn.gelu(x_208, approximate="none") + x_208 = None + x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) + x_209 = None + x_211 = torch.nn.functional.layer_norm( + x_210, + (4096,), + l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_210 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_212 = torch._C._nn.linear( + x_211, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_211 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_213 = torch.nn.functional.dropout(x_212, 0.0, False, False) + x_212 = None + x_214 = x_206 + x_213 + x_206 = x_213 = None + x_215 = torch.nn.functional.layer_norm( + x_214, + (1024,), + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ + ) = None + linear_56 = torch._C._nn.linear( + x_215, + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_, + ) + x_215 = ( + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_28 = linear_56.reshape(1, 197, 3, 16, 64) + linear_56 = None + qkv_14 = reshape_28.permute(2, 0, 3, 1, 4) + reshape_28 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_14 = unbind_14[0] + k_14 = unbind_14[1] + v_14 = unbind_14[2] + unbind_14 = None + x_216 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_15 = x_216.transpose(1, 2) + x_216 = None + x_217 = transpose_15.reshape(1, 197, 1024) + transpose_15 = None + x_218 = torch.nn.functional.layer_norm( + x_217, + (1024,), + l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_217 = l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_ + ) = None + x_219 = torch._C._nn.linear( + x_218, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_, + ) + x_218 = l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ + ) = None + x_220 = torch.nn.functional.dropout(x_219, 0.0, False, False) + x_219 = None + x_221 = x_214 + x_220 + x_214 = x_220 = None + x_222 = torch.nn.functional.layer_norm( + x_221, + (1024,), + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ + ) = None + x_223 = torch._C._nn.linear( + x_222, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_222 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_224 = torch._C._nn.gelu(x_223, approximate="none") + x_223 = None + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = torch.nn.functional.layer_norm( + x_225, + (4096,), + l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_225 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_227 = torch._C._nn.linear( + x_226, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_226 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_228 = torch.nn.functional.dropout(x_227, 0.0, False, False) + x_227 = None + x_229 = x_221 + x_228 + x_221 = x_228 = None + x_230 = torch.nn.functional.layer_norm( + x_229, + (1024,), + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ + ) = None + linear_60 = torch._C._nn.linear( + x_230, + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_, + ) + x_230 = ( + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_30 = linear_60.reshape(1, 197, 3, 16, 64) + linear_60 = None + qkv_15 = reshape_30.permute(2, 0, 3, 1, 4) + reshape_30 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_15 = unbind_15[0] + k_15 = unbind_15[1] + v_15 = unbind_15[2] + unbind_15 = None + x_231 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_16 = x_231.transpose(1, 2) + x_231 = None + x_232 = transpose_16.reshape(1, 197, 1024) + transpose_16 = None + x_233 = torch.nn.functional.layer_norm( + x_232, + (1024,), + l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_232 = l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_ + ) = None + x_234 = torch._C._nn.linear( + x_233, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_, + ) + x_233 = l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ + ) = None + x_235 = torch.nn.functional.dropout(x_234, 0.0, False, False) + x_234 = None + x_236 = x_229 + x_235 + x_229 = x_235 = None + x_237 = torch.nn.functional.layer_norm( + x_236, + (1024,), + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ + ) = None + x_238 = torch._C._nn.linear( + x_237, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_237 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_239 = torch._C._nn.gelu(x_238, approximate="none") + x_238 = None + x_240 = torch.nn.functional.dropout(x_239, 0.0, False, False) + x_239 = None + x_241 = torch.nn.functional.layer_norm( + x_240, + (4096,), + l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_240 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_242 = torch._C._nn.linear( + x_241, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_241 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_243 = torch.nn.functional.dropout(x_242, 0.0, False, False) + x_242 = None + x_244 = x_236 + x_243 + x_236 = x_243 = None + x_245 = torch.nn.functional.layer_norm( + x_244, + (1024,), + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ + ) = None + linear_64 = torch._C._nn.linear( + x_245, + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_, + ) + x_245 = ( + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_32 = linear_64.reshape(1, 197, 3, 16, 64) + linear_64 = None + qkv_16 = reshape_32.permute(2, 0, 3, 1, 4) + reshape_32 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_16 = unbind_16[0] + k_16 = unbind_16[1] + v_16 = unbind_16[2] + unbind_16 = None + x_246 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, attn_mask=None, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_17 = x_246.transpose(1, 2) + x_246 = None + x_247 = transpose_17.reshape(1, 197, 1024) + transpose_17 = None + x_248 = torch.nn.functional.layer_norm( + x_247, + (1024,), + l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_247 = l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_ + ) = None + x_249 = torch._C._nn.linear( + x_248, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_, + ) + x_248 = l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ + ) = None + x_250 = torch.nn.functional.dropout(x_249, 0.0, False, False) + x_249 = None + x_251 = x_244 + x_250 + x_244 = x_250 = None + x_252 = torch.nn.functional.layer_norm( + x_251, + (1024,), + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ + ) = None + x_253 = torch._C._nn.linear( + x_252, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_252 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_254 = torch._C._nn.gelu(x_253, approximate="none") + x_253 = None + x_255 = torch.nn.functional.dropout(x_254, 0.0, False, False) + x_254 = None + x_256 = torch.nn.functional.layer_norm( + x_255, + (4096,), + l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_255 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_257 = torch._C._nn.linear( + x_256, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_256 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_258 = torch.nn.functional.dropout(x_257, 0.0, False, False) + x_257 = None + x_259 = x_251 + x_258 + x_251 = x_258 = None + x_260 = torch.nn.functional.layer_norm( + x_259, + (1024,), + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ + ) = None + linear_68 = torch._C._nn.linear( + x_260, + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_, + ) + x_260 = ( + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_34 = linear_68.reshape(1, 197, 3, 16, 64) + linear_68 = None + qkv_17 = reshape_34.permute(2, 0, 3, 1, 4) + reshape_34 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_17 = unbind_17[0] + k_17 = unbind_17[1] + v_17 = unbind_17[2] + unbind_17 = None + x_261 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, attn_mask=None, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_18 = x_261.transpose(1, 2) + x_261 = None + x_262 = transpose_18.reshape(1, 197, 1024) + transpose_18 = None + x_263 = torch.nn.functional.layer_norm( + x_262, + (1024,), + l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_262 = l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_ + ) = None + x_264 = torch._C._nn.linear( + x_263, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_, + ) + x_263 = l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ + ) = None + x_265 = torch.nn.functional.dropout(x_264, 0.0, False, False) + x_264 = None + x_266 = x_259 + x_265 + x_259 = x_265 = None + x_267 = torch.nn.functional.layer_norm( + x_266, + (1024,), + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ + ) = None + x_268 = torch._C._nn.linear( + x_267, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_267 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_269 = torch._C._nn.gelu(x_268, approximate="none") + x_268 = None + x_270 = torch.nn.functional.dropout(x_269, 0.0, False, False) + x_269 = None + x_271 = torch.nn.functional.layer_norm( + x_270, + (4096,), + l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_270 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_272 = torch._C._nn.linear( + x_271, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_271 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_273 = torch.nn.functional.dropout(x_272, 0.0, False, False) + x_272 = None + x_274 = x_266 + x_273 + x_266 = x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (1024,), + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ + ) = None + linear_72 = torch._C._nn.linear( + x_275, + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_, + ) + x_275 = ( + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_36 = linear_72.reshape(1, 197, 3, 16, 64) + linear_72 = None + qkv_18 = reshape_36.permute(2, 0, 3, 1, 4) + reshape_36 = None + unbind_18 = qkv_18.unbind(0) + qkv_18 = None + q_18 = unbind_18[0] + k_18 = unbind_18[1] + v_18 = unbind_18[2] + unbind_18 = None + x_276 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_19 = x_276.transpose(1, 2) + x_276 = None + x_277 = transpose_19.reshape(1, 197, 1024) + transpose_19 = None + x_278 = torch.nn.functional.layer_norm( + x_277, + (1024,), + l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_277 = l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_ + ) = None + x_279 = torch._C._nn.linear( + x_278, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_, + ) + x_278 = l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ + ) = None + x_280 = torch.nn.functional.dropout(x_279, 0.0, False, False) + x_279 = None + x_281 = x_274 + x_280 + x_274 = x_280 = None + x_282 = torch.nn.functional.layer_norm( + x_281, + (1024,), + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ + ) = None + x_283 = torch._C._nn.linear( + x_282, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_282 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_284 = torch._C._nn.gelu(x_283, approximate="none") + x_283 = None + x_285 = torch.nn.functional.dropout(x_284, 0.0, False, False) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (4096,), + l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_285 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_286 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) + x_287 = None + x_289 = x_281 + x_288 + x_281 = x_288 = None + x_290 = torch.nn.functional.layer_norm( + x_289, + (1024,), + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ + ) = None + linear_76 = torch._C._nn.linear( + x_290, + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_, + ) + x_290 = ( + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_38 = linear_76.reshape(1, 197, 3, 16, 64) + linear_76 = None + qkv_19 = reshape_38.permute(2, 0, 3, 1, 4) + reshape_38 = None + unbind_19 = qkv_19.unbind(0) + qkv_19 = None + q_19 = unbind_19[0] + k_19 = unbind_19[1] + v_19 = unbind_19[2] + unbind_19 = None + x_291 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, attn_mask=None, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_20 = x_291.transpose(1, 2) + x_291 = None + x_292 = transpose_20.reshape(1, 197, 1024) + transpose_20 = None + x_293 = torch.nn.functional.layer_norm( + x_292, + (1024,), + l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_292 = l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_ + ) = None + x_294 = torch._C._nn.linear( + x_293, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_, + ) + x_293 = l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ + ) = None + x_295 = torch.nn.functional.dropout(x_294, 0.0, False, False) + x_294 = None + x_296 = x_289 + x_295 + x_289 = x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (1024,), + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ + ) = None + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_299 = torch._C._nn.gelu(x_298, approximate="none") + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_301 = torch.nn.functional.layer_norm( + x_300, + (4096,), + l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_300 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_302 = torch._C._nn.linear( + x_301, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_301 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_303 = torch.nn.functional.dropout(x_302, 0.0, False, False) + x_302 = None + x_304 = x_296 + x_303 + x_296 = x_303 = None + x_305 = torch.nn.functional.layer_norm( + x_304, + (1024,), + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ + ) = None + linear_80 = torch._C._nn.linear( + x_305, + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_, + ) + x_305 = ( + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_40 = linear_80.reshape(1, 197, 3, 16, 64) + linear_80 = None + qkv_20 = reshape_40.permute(2, 0, 3, 1, 4) + reshape_40 = None + unbind_20 = qkv_20.unbind(0) + qkv_20 = None + q_20 = unbind_20[0] + k_20 = unbind_20[1] + v_20 = unbind_20[2] + unbind_20 = None + x_306 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, attn_mask=None, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_21 = x_306.transpose(1, 2) + x_306 = None + x_307 = transpose_21.reshape(1, 197, 1024) + transpose_21 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (1024,), + l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_307 = l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_ + ) = None + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_, + ) + x_308 = l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ + ) = None + x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) + x_309 = None + x_311 = x_304 + x_310 + x_304 = x_310 = None + x_312 = torch.nn.functional.layer_norm( + x_311, + (1024,), + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ + ) = None + x_313 = torch._C._nn.linear( + x_312, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_312 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_314 = torch._C._nn.gelu(x_313, approximate="none") + x_313 = None + x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) + x_314 = None + x_316 = torch.nn.functional.layer_norm( + x_315, + (4096,), + l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_315 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_317 = torch._C._nn.linear( + x_316, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_316 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_318 = torch.nn.functional.dropout(x_317, 0.0, False, False) + x_317 = None + x_319 = x_311 + x_318 + x_311 = x_318 = None + x_320 = torch.nn.functional.layer_norm( + x_319, + (1024,), + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ + ) = None + linear_84 = torch._C._nn.linear( + x_320, + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_, + ) + x_320 = ( + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_42 = linear_84.reshape(1, 197, 3, 16, 64) + linear_84 = None + qkv_21 = reshape_42.permute(2, 0, 3, 1, 4) + reshape_42 = None + unbind_21 = qkv_21.unbind(0) + qkv_21 = None + q_21 = unbind_21[0] + k_21 = unbind_21[1] + v_21 = unbind_21[2] + unbind_21 = None + x_321 = torch._C._nn.scaled_dot_product_attention( + q_21, k_21, v_21, attn_mask=None, dropout_p=0.0 + ) + q_21 = k_21 = v_21 = None + transpose_22 = x_321.transpose(1, 2) + x_321 = None + x_322 = transpose_22.reshape(1, 197, 1024) + transpose_22 = None + x_323 = torch.nn.functional.layer_norm( + x_322, + (1024,), + l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_322 = l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_ + ) = None + x_324 = torch._C._nn.linear( + x_323, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_, + ) + x_323 = l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ + ) = None + x_325 = torch.nn.functional.dropout(x_324, 0.0, False, False) + x_324 = None + x_326 = x_319 + x_325 + x_319 = x_325 = None + x_327 = torch.nn.functional.layer_norm( + x_326, + (1024,), + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ + ) = None + x_328 = torch._C._nn.linear( + x_327, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_327 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_329 = torch._C._nn.gelu(x_328, approximate="none") + x_328 = None + x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) + x_329 = None + x_331 = torch.nn.functional.layer_norm( + x_330, + (4096,), + l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_330 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_332 = torch._C._nn.linear( + x_331, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_331 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_334 = x_326 + x_333 + x_326 = x_333 = None + x_335 = torch.nn.functional.layer_norm( + x_334, + (1024,), + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ + ) = None + linear_88 = torch._C._nn.linear( + x_335, + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_, + ) + x_335 = ( + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_44 = linear_88.reshape(1, 197, 3, 16, 64) + linear_88 = None + qkv_22 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_22 = qkv_22.unbind(0) + qkv_22 = None + q_22 = unbind_22[0] + k_22 = unbind_22[1] + v_22 = unbind_22[2] + unbind_22 = None + x_336 = torch._C._nn.scaled_dot_product_attention( + q_22, k_22, v_22, attn_mask=None, dropout_p=0.0 + ) + q_22 = k_22 = v_22 = None + transpose_23 = x_336.transpose(1, 2) + x_336 = None + x_337 = transpose_23.reshape(1, 197, 1024) + transpose_23 = None + x_338 = torch.nn.functional.layer_norm( + x_337, + (1024,), + l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_337 = l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_ + ) = None + x_339 = torch._C._nn.linear( + x_338, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_, + ) + x_338 = l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ + ) = None + x_340 = torch.nn.functional.dropout(x_339, 0.0, False, False) + x_339 = None + x_341 = x_334 + x_340 + x_334 = x_340 = None + x_342 = torch.nn.functional.layer_norm( + x_341, + (1024,), + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ + ) = None + x_343 = torch._C._nn.linear( + x_342, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_342 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_344 = torch._C._nn.gelu(x_343, approximate="none") + x_343 = None + x_345 = torch.nn.functional.dropout(x_344, 0.0, False, False) + x_344 = None + x_346 = torch.nn.functional.layer_norm( + x_345, + (4096,), + l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_345 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_347 = torch._C._nn.linear( + x_346, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_346 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) + x_347 = None + x_349 = x_341 + x_348 + x_341 = x_348 = None + x_350 = torch.nn.functional.layer_norm( + x_349, + (1024,), + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ + ) = None + linear_92 = torch._C._nn.linear( + x_350, + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_, + ) + x_350 = ( + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_46 = linear_92.reshape(1, 197, 3, 16, 64) + linear_92 = None + qkv_23 = reshape_46.permute(2, 0, 3, 1, 4) + reshape_46 = None + unbind_23 = qkv_23.unbind(0) + qkv_23 = None + q_23 = unbind_23[0] + k_23 = unbind_23[1] + v_23 = unbind_23[2] + unbind_23 = None + x_351 = torch._C._nn.scaled_dot_product_attention( + q_23, k_23, v_23, attn_mask=None, dropout_p=0.0 + ) + q_23 = k_23 = v_23 = None + transpose_24 = x_351.transpose(1, 2) + x_351 = None + x_352 = transpose_24.reshape(1, 197, 1024) + transpose_24 = None + x_353 = torch.nn.functional.layer_norm( + x_352, + (1024,), + l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_352 = l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_ + ) = None + x_354 = torch._C._nn.linear( + x_353, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_, + ) + x_353 = l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ + ) = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_356 = x_349 + x_355 + x_349 = x_355 = None + x_357 = torch.nn.functional.layer_norm( + x_356, + (1024,), + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ + ) = None + x_358 = torch._C._nn.linear( + x_357, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_357 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_359 = torch._C._nn.gelu(x_358, approximate="none") + x_358 = None + x_360 = torch.nn.functional.dropout(x_359, 0.0, False, False) + x_359 = None + x_361 = torch.nn.functional.layer_norm( + x_360, + (4096,), + l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_360 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_362 = torch._C._nn.linear( + x_361, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_361 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) + x_362 = None + x_364 = x_356 + x_363 + x_356 = x_363 = None + x_365 = x_364[ + (slice(None, None, None), slice(l_self_num_prefix_tokens, None, None)) + ] + x_364 = l_self_num_prefix_tokens = None + x_366 = x_365.mean(dim=1) + x_365 = None + x_367 = torch.nn.functional.layer_norm( + x_366, + (1024,), + l_self_modules_fc_norm_parameters_weight_, + l_self_modules_fc_norm_parameters_bias_, + 1e-05, + ) + x_366 = ( + l_self_modules_fc_norm_parameters_weight_ + ) = l_self_modules_fc_norm_parameters_bias_ = None + x_368 = torch.nn.functional.dropout(x_367, 0.0, False, False) + x_367 = None + x_369 = torch._C._nn.linear( + x_368, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_368 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_369,) diff --git a/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/weight_meta.py b/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/weight_meta.py new file mode 100644 index 0000000000..2f8b8e8681 --- /dev/null +++ b/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/weight_meta.py @@ -0,0 +1,4090 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.222 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [1024, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 197, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_num_prefix_tokens: + name = "L_self_num_prefix_tokens" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_self_modules_fc_norm_parameters_weight_: + name = "L_self_modules_fc_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_fc_norm_parameters_bias_: + name = "L_self_modules_fc_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/beit3_large_patch16_224.indomain_pt/graph_hash.txt b/samples/timm/beit3_large_patch16_224.indomain_pt/graph_hash.txt new file mode 100644 index 0000000000..44b074954d --- /dev/null +++ b/samples/timm/beit3_large_patch16_224.indomain_pt/graph_hash.txt @@ -0,0 +1 @@ +a79caac738b040f293c01cd92e1770f312f059371fd757fa138f608c37b3cf6e \ No newline at end of file diff --git a/samples/timm/beit3_large_patch16_224.indomain_pt/graph_net.json b/samples/timm/beit3_large_patch16_224.indomain_pt/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/beit3_large_patch16_224.indomain_pt/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/beit3_large_patch16_224.indomain_pt/input_meta.py b/samples/timm/beit3_large_patch16_224.indomain_pt/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/beit3_large_patch16_224.indomain_pt/input_tensor_constraints.py b/samples/timm/beit3_large_patch16_224.indomain_pt/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/beit3_large_patch16_224.indomain_pt/model.py b/samples/timm/beit3_large_patch16_224.indomain_pt/model.py new file mode 100644 index 0000000000..59f9368528 --- /dev/null +++ b/samples/timm/beit3_large_patch16_224.indomain_pt/model.py @@ -0,0 +1,4234 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_num_prefix_tokens: torch.SymInt, + L_self_modules_fc_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_fc_norm_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_num_prefix_tokens = L_self_num_prefix_tokens + l_self_modules_fc_norm_parameters_weight_ = ( + L_self_modules_fc_norm_parameters_weight_ + ) + l_self_modules_fc_norm_parameters_bias_ = ( + L_self_modules_fc_norm_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + x_2 = torch.cat([expand, x_1], dim=1) + expand = x_1 = None + x_3 = x_2 + l_self_parameters_pos_embed_ + x_2 = l_self_parameters_pos_embed_ = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (1024,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 197, 3, 16, 64) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 197, 1024) + transpose_1 = None + x_8 = torch.nn.functional.layer_norm( + x_7, + (1024,), + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_ + ) = None + x_9 = torch._C._nn.linear( + x_8, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_8 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = x_4 + x_10 + x_4 = x_10 = None + x_12 = torch.nn.functional.layer_norm( + x_11, + (1024,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_13 = torch._C._nn.linear( + x_12, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_12 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_14 = torch._C._nn.gelu(x_13, approximate="none") + x_13 = None + x_15 = torch.nn.functional.dropout(x_14, 0.0, False, False) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (4096,), + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_15 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_16 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_18 = torch.nn.functional.dropout(x_17, 0.0, False, False) + x_17 = None + x_19 = x_11 + x_18 + x_11 = x_18 = None + x_20 = torch.nn.functional.layer_norm( + x_19, + (1024,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 197, 3, 16, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_21 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_21.transpose(1, 2) + x_21 = None + x_22 = transpose_2.reshape(1, 197, 1024) + transpose_2 = None + x_23 = torch.nn.functional.layer_norm( + x_22, + (1024,), + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_22 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_ + ) = None + x_24 = torch._C._nn.linear( + x_23, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_23 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_25 = torch.nn.functional.dropout(x_24, 0.0, False, False) + x_24 = None + x_26 = x_19 + x_25 + x_19 = x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (1024,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_29 = torch._C._nn.gelu(x_28, approximate="none") + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (4096,), + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_30 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_32 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + x_34 = x_26 + x_33 + x_26 = x_33 = None + x_35 = torch.nn.functional.layer_norm( + x_34, + (1024,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_35, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_35 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 197, 3, 16, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_36 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_36.transpose(1, 2) + x_36 = None + x_37 = transpose_3.reshape(1, 197, 1024) + transpose_3 = None + x_38 = torch.nn.functional.layer_norm( + x_37, + (1024,), + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_ + ) = None + x_39 = torch._C._nn.linear( + x_38, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_38 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = x_34 + x_40 + x_34 = x_40 = None + x_42 = torch.nn.functional.layer_norm( + x_41, + (1024,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_43 = torch._C._nn.linear( + x_42, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_42 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_44 = torch._C._nn.gelu(x_43, approximate="none") + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = torch.nn.functional.layer_norm( + x_45, + (4096,), + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_45 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_41 + x_48 + x_41 = x_48 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (1024,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 197, 3, 16, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_51 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_51.transpose(1, 2) + x_51 = None + x_52 = transpose_4.reshape(1, 197, 1024) + transpose_4 = None + x_53 = torch.nn.functional.layer_norm( + x_52, + (1024,), + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_52 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_ + ) = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = x_49 + x_55 + x_49 = x_55 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (1024,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_58 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_59 = torch._C._nn.gelu(x_58, approximate="none") + x_58 = None + x_60 = torch.nn.functional.dropout(x_59, 0.0, False, False) + x_59 = None + x_61 = torch.nn.functional.layer_norm( + x_60, + (4096,), + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_60 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_62 = torch._C._nn.linear( + x_61, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_61 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_63 = torch.nn.functional.dropout(x_62, 0.0, False, False) + x_62 = None + x_64 = x_56 + x_63 + x_56 = x_63 = None + x_65 = torch.nn.functional.layer_norm( + x_64, + (1024,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_65, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_65 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 197, 3, 16, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_66 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_66.transpose(1, 2) + x_66 = None + x_67 = transpose_5.reshape(1, 197, 1024) + transpose_5 = None + x_68 = torch.nn.functional.layer_norm( + x_67, + (1024,), + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_67 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_ + ) = None + x_69 = torch._C._nn.linear( + x_68, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_68 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_71 = x_64 + x_70 + x_64 = x_70 = None + x_72 = torch.nn.functional.layer_norm( + x_71, + (1024,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_74 = torch._C._nn.gelu(x_73, approximate="none") + x_73 = None + x_75 = torch.nn.functional.dropout(x_74, 0.0, False, False) + x_74 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (4096,), + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_75 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_78 = torch.nn.functional.dropout(x_77, 0.0, False, False) + x_77 = None + x_79 = x_71 + x_78 + x_71 = x_78 = None + x_80 = torch.nn.functional.layer_norm( + x_79, + (1024,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_80, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_80 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 197, 3, 16, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_81 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_81.transpose(1, 2) + x_81 = None + x_82 = transpose_6.reshape(1, 197, 1024) + transpose_6 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (1024,), + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_82 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_ + ) = None + x_84 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_85 = torch.nn.functional.dropout(x_84, 0.0, False, False) + x_84 = None + x_86 = x_79 + x_85 + x_79 = x_85 = None + x_87 = torch.nn.functional.layer_norm( + x_86, + (1024,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_88 = torch._C._nn.linear( + x_87, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_87 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_89 = torch._C._nn.gelu(x_88, approximate="none") + x_88 = None + x_90 = torch.nn.functional.dropout(x_89, 0.0, False, False) + x_89 = None + x_91 = torch.nn.functional.layer_norm( + x_90, + (4096,), + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_90 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_92 = torch._C._nn.linear( + x_91, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_91 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_93 = torch.nn.functional.dropout(x_92, 0.0, False, False) + x_92 = None + x_94 = x_86 + x_93 + x_86 = x_93 = None + x_95 = torch.nn.functional.layer_norm( + x_94, + (1024,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_95, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_95 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 197, 3, 16, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_96 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_96.transpose(1, 2) + x_96 = None + x_97 = transpose_7.reshape(1, 197, 1024) + transpose_7 = None + x_98 = torch.nn.functional.layer_norm( + x_97, + (1024,), + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_97 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_ + ) = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + x_101 = x_94 + x_100 + x_94 = x_100 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (1024,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch.nn.functional.layer_norm( + x_105, + (4096,), + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_105 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_107 = torch._C._nn.linear( + x_106, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_106 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_101 + x_108 + x_101 = x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (1024,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_110, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_110 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 197, 3, 16, 64) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_111 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_111.transpose(1, 2) + x_111 = None + x_112 = transpose_8.reshape(1, 197, 1024) + transpose_8 = None + x_113 = torch.nn.functional.layer_norm( + x_112, + (1024,), + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_112 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_ + ) = None + x_114 = torch._C._nn.linear( + x_113, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_113 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_115 = torch.nn.functional.dropout(x_114, 0.0, False, False) + x_114 = None + x_116 = x_109 + x_115 + x_109 = x_115 = None + x_117 = torch.nn.functional.layer_norm( + x_116, + (1024,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_118 = torch._C._nn.linear( + x_117, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_117 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_119 = torch._C._nn.gelu(x_118, approximate="none") + x_118 = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (4096,), + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_120 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_121 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) + x_122 = None + x_124 = x_116 + x_123 + x_116 = x_123 = None + x_125 = torch.nn.functional.layer_norm( + x_124, + (1024,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_125, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_125 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 197, 3, 16, 64) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_126 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_126.transpose(1, 2) + x_126 = None + x_127 = transpose_9.reshape(1, 197, 1024) + transpose_9 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (1024,), + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_127 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) + x_129 = None + x_131 = x_124 + x_130 + x_124 = x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (1024,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_134 = torch._C._nn.gelu(x_133, approximate="none") + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = torch.nn.functional.layer_norm( + x_135, + (4096,), + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_135 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_131 + x_138 + x_131 = x_138 = None + x_140 = torch.nn.functional.layer_norm( + x_139, + (1024,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_140, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_140 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 197, 3, 16, 64) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_141 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_141.transpose(1, 2) + x_141 = None + x_142 = transpose_10.reshape(1, 197, 1024) + transpose_10 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (1024,), + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_142 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_ + ) = None + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_143 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_145 = torch.nn.functional.dropout(x_144, 0.0, False, False) + x_144 = None + x_146 = x_139 + x_145 + x_139 = x_145 = None + x_147 = torch.nn.functional.layer_norm( + x_146, + (1024,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_147 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_149 = torch._C._nn.gelu(x_148, approximate="none") + x_148 = None + x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) + x_149 = None + x_151 = torch.nn.functional.layer_norm( + x_150, + (4096,), + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_150 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_152 = torch._C._nn.linear( + x_151, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_151 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_153 = torch.nn.functional.dropout(x_152, 0.0, False, False) + x_152 = None + x_154 = x_146 + x_153 + x_146 = x_153 = None + x_155 = torch.nn.functional.layer_norm( + x_154, + (1024,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_155, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_155 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 197, 3, 16, 64) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_156 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_156.transpose(1, 2) + x_156 = None + x_157 = transpose_11.reshape(1, 197, 1024) + transpose_11 = None + x_158 = torch.nn.functional.layer_norm( + x_157, + (1024,), + l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_157 = l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_ + ) = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_158 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_154 + x_160 + x_154 = x_160 = None + x_162 = torch.nn.functional.layer_norm( + x_161, + (1024,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_163 = torch._C._nn.linear( + x_162, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_162 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_164 = torch._C._nn.gelu(x_163, approximate="none") + x_163 = None + x_165 = torch.nn.functional.dropout(x_164, 0.0, False, False) + x_164 = None + x_166 = torch.nn.functional.layer_norm( + x_165, + (4096,), + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_165 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_167 = torch._C._nn.linear( + x_166, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_166 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_169 = x_161 + x_168 + x_161 = x_168 = None + x_170 = torch.nn.functional.layer_norm( + x_169, + (1024,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_170, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_170 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 197, 3, 16, 64) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_171 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_171.transpose(1, 2) + x_171 = None + x_172 = transpose_12.reshape(1, 197, 1024) + transpose_12 = None + x_173 = torch.nn.functional.layer_norm( + x_172, + (1024,), + l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_172 = l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_ + ) = None + x_174 = torch._C._nn.linear( + x_173, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_173 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_175 = torch.nn.functional.dropout(x_174, 0.0, False, False) + x_174 = None + x_176 = x_169 + x_175 + x_169 = x_175 = None + x_177 = torch.nn.functional.layer_norm( + x_176, + (1024,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_178 = torch._C._nn.linear( + x_177, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_177 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_179 = torch._C._nn.gelu(x_178, approximate="none") + x_178 = None + x_180 = torch.nn.functional.dropout(x_179, 0.0, False, False) + x_179 = None + x_181 = torch.nn.functional.layer_norm( + x_180, + (4096,), + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_180 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_182 = torch._C._nn.linear( + x_181, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_181 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_183 = torch.nn.functional.dropout(x_182, 0.0, False, False) + x_182 = None + x_184 = x_176 + x_183 + x_176 = x_183 = None + x_185 = torch.nn.functional.layer_norm( + x_184, + (1024,), + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ + ) = None + linear_48 = torch._C._nn.linear( + x_185, + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_, + ) + x_185 = ( + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_24 = linear_48.reshape(1, 197, 3, 16, 64) + linear_48 = None + qkv_12 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_12 = unbind_12[0] + k_12 = unbind_12[1] + v_12 = unbind_12[2] + unbind_12 = None + x_186 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_13 = x_186.transpose(1, 2) + x_186 = None + x_187 = transpose_13.reshape(1, 197, 1024) + transpose_13 = None + x_188 = torch.nn.functional.layer_norm( + x_187, + (1024,), + l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_187 = l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_ + ) = None + x_189 = torch._C._nn.linear( + x_188, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_, + ) + x_188 = l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ + ) = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = x_184 + x_190 + x_184 = x_190 = None + x_192 = torch.nn.functional.layer_norm( + x_191, + (1024,), + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ + ) = None + x_193 = torch._C._nn.linear( + x_192, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_192 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_194 = torch._C._nn.gelu(x_193, approximate="none") + x_193 = None + x_195 = torch.nn.functional.dropout(x_194, 0.0, False, False) + x_194 = None + x_196 = torch.nn.functional.layer_norm( + x_195, + (4096,), + l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_195 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_197 = torch._C._nn.linear( + x_196, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_196 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_198 = torch.nn.functional.dropout(x_197, 0.0, False, False) + x_197 = None + x_199 = x_191 + x_198 + x_191 = x_198 = None + x_200 = torch.nn.functional.layer_norm( + x_199, + (1024,), + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ + ) = None + linear_52 = torch._C._nn.linear( + x_200, + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_, + ) + x_200 = ( + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_26 = linear_52.reshape(1, 197, 3, 16, 64) + linear_52 = None + qkv_13 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_13 = unbind_13[0] + k_13 = unbind_13[1] + v_13 = unbind_13[2] + unbind_13 = None + x_201 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_14 = x_201.transpose(1, 2) + x_201 = None + x_202 = transpose_14.reshape(1, 197, 1024) + transpose_14 = None + x_203 = torch.nn.functional.layer_norm( + x_202, + (1024,), + l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_202 = l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_ + ) = None + x_204 = torch._C._nn.linear( + x_203, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_, + ) + x_203 = l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ + ) = None + x_205 = torch.nn.functional.dropout(x_204, 0.0, False, False) + x_204 = None + x_206 = x_199 + x_205 + x_199 = x_205 = None + x_207 = torch.nn.functional.layer_norm( + x_206, + (1024,), + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ + ) = None + x_208 = torch._C._nn.linear( + x_207, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_207 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_209 = torch._C._nn.gelu(x_208, approximate="none") + x_208 = None + x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) + x_209 = None + x_211 = torch.nn.functional.layer_norm( + x_210, + (4096,), + l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_210 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_212 = torch._C._nn.linear( + x_211, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_211 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_213 = torch.nn.functional.dropout(x_212, 0.0, False, False) + x_212 = None + x_214 = x_206 + x_213 + x_206 = x_213 = None + x_215 = torch.nn.functional.layer_norm( + x_214, + (1024,), + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ + ) = None + linear_56 = torch._C._nn.linear( + x_215, + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_, + ) + x_215 = ( + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_28 = linear_56.reshape(1, 197, 3, 16, 64) + linear_56 = None + qkv_14 = reshape_28.permute(2, 0, 3, 1, 4) + reshape_28 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_14 = unbind_14[0] + k_14 = unbind_14[1] + v_14 = unbind_14[2] + unbind_14 = None + x_216 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_15 = x_216.transpose(1, 2) + x_216 = None + x_217 = transpose_15.reshape(1, 197, 1024) + transpose_15 = None + x_218 = torch.nn.functional.layer_norm( + x_217, + (1024,), + l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_217 = l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_ + ) = None + x_219 = torch._C._nn.linear( + x_218, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_, + ) + x_218 = l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ + ) = None + x_220 = torch.nn.functional.dropout(x_219, 0.0, False, False) + x_219 = None + x_221 = x_214 + x_220 + x_214 = x_220 = None + x_222 = torch.nn.functional.layer_norm( + x_221, + (1024,), + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ + ) = None + x_223 = torch._C._nn.linear( + x_222, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_222 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_224 = torch._C._nn.gelu(x_223, approximate="none") + x_223 = None + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = torch.nn.functional.layer_norm( + x_225, + (4096,), + l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_225 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_227 = torch._C._nn.linear( + x_226, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_226 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_228 = torch.nn.functional.dropout(x_227, 0.0, False, False) + x_227 = None + x_229 = x_221 + x_228 + x_221 = x_228 = None + x_230 = torch.nn.functional.layer_norm( + x_229, + (1024,), + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ + ) = None + linear_60 = torch._C._nn.linear( + x_230, + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_, + ) + x_230 = ( + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_30 = linear_60.reshape(1, 197, 3, 16, 64) + linear_60 = None + qkv_15 = reshape_30.permute(2, 0, 3, 1, 4) + reshape_30 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_15 = unbind_15[0] + k_15 = unbind_15[1] + v_15 = unbind_15[2] + unbind_15 = None + x_231 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_16 = x_231.transpose(1, 2) + x_231 = None + x_232 = transpose_16.reshape(1, 197, 1024) + transpose_16 = None + x_233 = torch.nn.functional.layer_norm( + x_232, + (1024,), + l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_232 = l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_ + ) = None + x_234 = torch._C._nn.linear( + x_233, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_, + ) + x_233 = l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ + ) = None + x_235 = torch.nn.functional.dropout(x_234, 0.0, False, False) + x_234 = None + x_236 = x_229 + x_235 + x_229 = x_235 = None + x_237 = torch.nn.functional.layer_norm( + x_236, + (1024,), + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ + ) = None + x_238 = torch._C._nn.linear( + x_237, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_237 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_239 = torch._C._nn.gelu(x_238, approximate="none") + x_238 = None + x_240 = torch.nn.functional.dropout(x_239, 0.0, False, False) + x_239 = None + x_241 = torch.nn.functional.layer_norm( + x_240, + (4096,), + l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_240 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_242 = torch._C._nn.linear( + x_241, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_241 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_243 = torch.nn.functional.dropout(x_242, 0.0, False, False) + x_242 = None + x_244 = x_236 + x_243 + x_236 = x_243 = None + x_245 = torch.nn.functional.layer_norm( + x_244, + (1024,), + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ + ) = None + linear_64 = torch._C._nn.linear( + x_245, + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_, + ) + x_245 = ( + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_32 = linear_64.reshape(1, 197, 3, 16, 64) + linear_64 = None + qkv_16 = reshape_32.permute(2, 0, 3, 1, 4) + reshape_32 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_16 = unbind_16[0] + k_16 = unbind_16[1] + v_16 = unbind_16[2] + unbind_16 = None + x_246 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, attn_mask=None, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_17 = x_246.transpose(1, 2) + x_246 = None + x_247 = transpose_17.reshape(1, 197, 1024) + transpose_17 = None + x_248 = torch.nn.functional.layer_norm( + x_247, + (1024,), + l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_247 = l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_ + ) = None + x_249 = torch._C._nn.linear( + x_248, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_, + ) + x_248 = l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ + ) = None + x_250 = torch.nn.functional.dropout(x_249, 0.0, False, False) + x_249 = None + x_251 = x_244 + x_250 + x_244 = x_250 = None + x_252 = torch.nn.functional.layer_norm( + x_251, + (1024,), + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ + ) = None + x_253 = torch._C._nn.linear( + x_252, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_252 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_254 = torch._C._nn.gelu(x_253, approximate="none") + x_253 = None + x_255 = torch.nn.functional.dropout(x_254, 0.0, False, False) + x_254 = None + x_256 = torch.nn.functional.layer_norm( + x_255, + (4096,), + l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_255 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_257 = torch._C._nn.linear( + x_256, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_256 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_258 = torch.nn.functional.dropout(x_257, 0.0, False, False) + x_257 = None + x_259 = x_251 + x_258 + x_251 = x_258 = None + x_260 = torch.nn.functional.layer_norm( + x_259, + (1024,), + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ + ) = None + linear_68 = torch._C._nn.linear( + x_260, + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_, + ) + x_260 = ( + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_34 = linear_68.reshape(1, 197, 3, 16, 64) + linear_68 = None + qkv_17 = reshape_34.permute(2, 0, 3, 1, 4) + reshape_34 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_17 = unbind_17[0] + k_17 = unbind_17[1] + v_17 = unbind_17[2] + unbind_17 = None + x_261 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, attn_mask=None, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_18 = x_261.transpose(1, 2) + x_261 = None + x_262 = transpose_18.reshape(1, 197, 1024) + transpose_18 = None + x_263 = torch.nn.functional.layer_norm( + x_262, + (1024,), + l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_262 = l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_ + ) = None + x_264 = torch._C._nn.linear( + x_263, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_, + ) + x_263 = l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ + ) = None + x_265 = torch.nn.functional.dropout(x_264, 0.0, False, False) + x_264 = None + x_266 = x_259 + x_265 + x_259 = x_265 = None + x_267 = torch.nn.functional.layer_norm( + x_266, + (1024,), + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ + ) = None + x_268 = torch._C._nn.linear( + x_267, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_267 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_269 = torch._C._nn.gelu(x_268, approximate="none") + x_268 = None + x_270 = torch.nn.functional.dropout(x_269, 0.0, False, False) + x_269 = None + x_271 = torch.nn.functional.layer_norm( + x_270, + (4096,), + l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_270 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_272 = torch._C._nn.linear( + x_271, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_271 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_273 = torch.nn.functional.dropout(x_272, 0.0, False, False) + x_272 = None + x_274 = x_266 + x_273 + x_266 = x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (1024,), + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ + ) = None + linear_72 = torch._C._nn.linear( + x_275, + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_, + ) + x_275 = ( + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_36 = linear_72.reshape(1, 197, 3, 16, 64) + linear_72 = None + qkv_18 = reshape_36.permute(2, 0, 3, 1, 4) + reshape_36 = None + unbind_18 = qkv_18.unbind(0) + qkv_18 = None + q_18 = unbind_18[0] + k_18 = unbind_18[1] + v_18 = unbind_18[2] + unbind_18 = None + x_276 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_19 = x_276.transpose(1, 2) + x_276 = None + x_277 = transpose_19.reshape(1, 197, 1024) + transpose_19 = None + x_278 = torch.nn.functional.layer_norm( + x_277, + (1024,), + l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_277 = l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_ + ) = None + x_279 = torch._C._nn.linear( + x_278, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_, + ) + x_278 = l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ + ) = None + x_280 = torch.nn.functional.dropout(x_279, 0.0, False, False) + x_279 = None + x_281 = x_274 + x_280 + x_274 = x_280 = None + x_282 = torch.nn.functional.layer_norm( + x_281, + (1024,), + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ + ) = None + x_283 = torch._C._nn.linear( + x_282, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_282 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_284 = torch._C._nn.gelu(x_283, approximate="none") + x_283 = None + x_285 = torch.nn.functional.dropout(x_284, 0.0, False, False) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (4096,), + l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_285 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_286 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) + x_287 = None + x_289 = x_281 + x_288 + x_281 = x_288 = None + x_290 = torch.nn.functional.layer_norm( + x_289, + (1024,), + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ + ) = None + linear_76 = torch._C._nn.linear( + x_290, + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_, + ) + x_290 = ( + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_38 = linear_76.reshape(1, 197, 3, 16, 64) + linear_76 = None + qkv_19 = reshape_38.permute(2, 0, 3, 1, 4) + reshape_38 = None + unbind_19 = qkv_19.unbind(0) + qkv_19 = None + q_19 = unbind_19[0] + k_19 = unbind_19[1] + v_19 = unbind_19[2] + unbind_19 = None + x_291 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, attn_mask=None, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_20 = x_291.transpose(1, 2) + x_291 = None + x_292 = transpose_20.reshape(1, 197, 1024) + transpose_20 = None + x_293 = torch.nn.functional.layer_norm( + x_292, + (1024,), + l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_292 = l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_ + ) = None + x_294 = torch._C._nn.linear( + x_293, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_, + ) + x_293 = l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ + ) = None + x_295 = torch.nn.functional.dropout(x_294, 0.0, False, False) + x_294 = None + x_296 = x_289 + x_295 + x_289 = x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (1024,), + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ + ) = None + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_299 = torch._C._nn.gelu(x_298, approximate="none") + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_301 = torch.nn.functional.layer_norm( + x_300, + (4096,), + l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_300 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_302 = torch._C._nn.linear( + x_301, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_301 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_303 = torch.nn.functional.dropout(x_302, 0.0, False, False) + x_302 = None + x_304 = x_296 + x_303 + x_296 = x_303 = None + x_305 = torch.nn.functional.layer_norm( + x_304, + (1024,), + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ + ) = None + linear_80 = torch._C._nn.linear( + x_305, + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_, + ) + x_305 = ( + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_40 = linear_80.reshape(1, 197, 3, 16, 64) + linear_80 = None + qkv_20 = reshape_40.permute(2, 0, 3, 1, 4) + reshape_40 = None + unbind_20 = qkv_20.unbind(0) + qkv_20 = None + q_20 = unbind_20[0] + k_20 = unbind_20[1] + v_20 = unbind_20[2] + unbind_20 = None + x_306 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, attn_mask=None, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_21 = x_306.transpose(1, 2) + x_306 = None + x_307 = transpose_21.reshape(1, 197, 1024) + transpose_21 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (1024,), + l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_307 = l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_ + ) = None + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_, + ) + x_308 = l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ + ) = None + x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) + x_309 = None + x_311 = x_304 + x_310 + x_304 = x_310 = None + x_312 = torch.nn.functional.layer_norm( + x_311, + (1024,), + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ + ) = None + x_313 = torch._C._nn.linear( + x_312, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_312 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_314 = torch._C._nn.gelu(x_313, approximate="none") + x_313 = None + x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) + x_314 = None + x_316 = torch.nn.functional.layer_norm( + x_315, + (4096,), + l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_315 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_317 = torch._C._nn.linear( + x_316, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_316 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_318 = torch.nn.functional.dropout(x_317, 0.0, False, False) + x_317 = None + x_319 = x_311 + x_318 + x_311 = x_318 = None + x_320 = torch.nn.functional.layer_norm( + x_319, + (1024,), + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ + ) = None + linear_84 = torch._C._nn.linear( + x_320, + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_, + ) + x_320 = ( + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_42 = linear_84.reshape(1, 197, 3, 16, 64) + linear_84 = None + qkv_21 = reshape_42.permute(2, 0, 3, 1, 4) + reshape_42 = None + unbind_21 = qkv_21.unbind(0) + qkv_21 = None + q_21 = unbind_21[0] + k_21 = unbind_21[1] + v_21 = unbind_21[2] + unbind_21 = None + x_321 = torch._C._nn.scaled_dot_product_attention( + q_21, k_21, v_21, attn_mask=None, dropout_p=0.0 + ) + q_21 = k_21 = v_21 = None + transpose_22 = x_321.transpose(1, 2) + x_321 = None + x_322 = transpose_22.reshape(1, 197, 1024) + transpose_22 = None + x_323 = torch.nn.functional.layer_norm( + x_322, + (1024,), + l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_322 = l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_ + ) = None + x_324 = torch._C._nn.linear( + x_323, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_, + ) + x_323 = l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ + ) = None + x_325 = torch.nn.functional.dropout(x_324, 0.0, False, False) + x_324 = None + x_326 = x_319 + x_325 + x_319 = x_325 = None + x_327 = torch.nn.functional.layer_norm( + x_326, + (1024,), + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ + ) = None + x_328 = torch._C._nn.linear( + x_327, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_327 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_329 = torch._C._nn.gelu(x_328, approximate="none") + x_328 = None + x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) + x_329 = None + x_331 = torch.nn.functional.layer_norm( + x_330, + (4096,), + l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_330 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_332 = torch._C._nn.linear( + x_331, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_331 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_334 = x_326 + x_333 + x_326 = x_333 = None + x_335 = torch.nn.functional.layer_norm( + x_334, + (1024,), + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ + ) = None + linear_88 = torch._C._nn.linear( + x_335, + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_, + ) + x_335 = ( + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_44 = linear_88.reshape(1, 197, 3, 16, 64) + linear_88 = None + qkv_22 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_22 = qkv_22.unbind(0) + qkv_22 = None + q_22 = unbind_22[0] + k_22 = unbind_22[1] + v_22 = unbind_22[2] + unbind_22 = None + x_336 = torch._C._nn.scaled_dot_product_attention( + q_22, k_22, v_22, attn_mask=None, dropout_p=0.0 + ) + q_22 = k_22 = v_22 = None + transpose_23 = x_336.transpose(1, 2) + x_336 = None + x_337 = transpose_23.reshape(1, 197, 1024) + transpose_23 = None + x_338 = torch.nn.functional.layer_norm( + x_337, + (1024,), + l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_337 = l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_ + ) = None + x_339 = torch._C._nn.linear( + x_338, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_, + ) + x_338 = l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ + ) = None + x_340 = torch.nn.functional.dropout(x_339, 0.0, False, False) + x_339 = None + x_341 = x_334 + x_340 + x_334 = x_340 = None + x_342 = torch.nn.functional.layer_norm( + x_341, + (1024,), + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ + ) = None + x_343 = torch._C._nn.linear( + x_342, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_342 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_344 = torch._C._nn.gelu(x_343, approximate="none") + x_343 = None + x_345 = torch.nn.functional.dropout(x_344, 0.0, False, False) + x_344 = None + x_346 = torch.nn.functional.layer_norm( + x_345, + (4096,), + l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_345 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_347 = torch._C._nn.linear( + x_346, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_346 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) + x_347 = None + x_349 = x_341 + x_348 + x_341 = x_348 = None + x_350 = torch.nn.functional.layer_norm( + x_349, + (1024,), + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ + ) = None + linear_92 = torch._C._nn.linear( + x_350, + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_, + ) + x_350 = ( + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_46 = linear_92.reshape(1, 197, 3, 16, 64) + linear_92 = None + qkv_23 = reshape_46.permute(2, 0, 3, 1, 4) + reshape_46 = None + unbind_23 = qkv_23.unbind(0) + qkv_23 = None + q_23 = unbind_23[0] + k_23 = unbind_23[1] + v_23 = unbind_23[2] + unbind_23 = None + x_351 = torch._C._nn.scaled_dot_product_attention( + q_23, k_23, v_23, attn_mask=None, dropout_p=0.0 + ) + q_23 = k_23 = v_23 = None + transpose_24 = x_351.transpose(1, 2) + x_351 = None + x_352 = transpose_24.reshape(1, 197, 1024) + transpose_24 = None + x_353 = torch.nn.functional.layer_norm( + x_352, + (1024,), + l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_352 = l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_ + ) = None + x_354 = torch._C._nn.linear( + x_353, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_, + ) + x_353 = l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ + ) = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_356 = x_349 + x_355 + x_349 = x_355 = None + x_357 = torch.nn.functional.layer_norm( + x_356, + (1024,), + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ + ) = None + x_358 = torch._C._nn.linear( + x_357, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_357 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_359 = torch._C._nn.gelu(x_358, approximate="none") + x_358 = None + x_360 = torch.nn.functional.dropout(x_359, 0.0, False, False) + x_359 = None + x_361 = torch.nn.functional.layer_norm( + x_360, + (4096,), + l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_360 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_362 = torch._C._nn.linear( + x_361, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_361 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) + x_362 = None + x_364 = x_356 + x_363 + x_356 = x_363 = None + x_365 = x_364[ + (slice(None, None, None), slice(l_self_num_prefix_tokens, None, None)) + ] + x_364 = l_self_num_prefix_tokens = None + x_366 = x_365.mean(dim=1) + x_365 = None + x_367 = torch.nn.functional.layer_norm( + x_366, + (1024,), + l_self_modules_fc_norm_parameters_weight_, + l_self_modules_fc_norm_parameters_bias_, + 1e-05, + ) + x_366 = ( + l_self_modules_fc_norm_parameters_weight_ + ) = l_self_modules_fc_norm_parameters_bias_ = None + x_368 = torch.nn.functional.dropout(x_367, 0.0, False, False) + x_367 = None + return (x_368,) diff --git a/samples/timm/beit3_large_patch16_224.indomain_pt/weight_meta.py b/samples/timm/beit3_large_patch16_224.indomain_pt/weight_meta.py new file mode 100644 index 0000000000..5bda823d60 --- /dev/null +++ b/samples/timm/beit3_large_patch16_224.indomain_pt/weight_meta.py @@ -0,0 +1,4070 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.224 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [1024, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 197, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_num_prefix_tokens: + name = "L_self_num_prefix_tokens" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_self_modules_fc_norm_parameters_weight_: + name = "L_self_modules_fc_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_fc_norm_parameters_bias_: + name = "L_self_modules_fc_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/beit3_large_patch16_224.pt/graph_hash.txt b/samples/timm/beit3_large_patch16_224.pt/graph_hash.txt new file mode 100644 index 0000000000..44b074954d --- /dev/null +++ b/samples/timm/beit3_large_patch16_224.pt/graph_hash.txt @@ -0,0 +1 @@ +a79caac738b040f293c01cd92e1770f312f059371fd757fa138f608c37b3cf6e \ No newline at end of file diff --git a/samples/timm/beit3_large_patch16_224.pt/graph_net.json b/samples/timm/beit3_large_patch16_224.pt/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/beit3_large_patch16_224.pt/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/beit3_large_patch16_224.pt/input_meta.py b/samples/timm/beit3_large_patch16_224.pt/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/beit3_large_patch16_224.pt/input_tensor_constraints.py b/samples/timm/beit3_large_patch16_224.pt/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/beit3_large_patch16_224.pt/model.py b/samples/timm/beit3_large_patch16_224.pt/model.py new file mode 100644 index 0000000000..59f9368528 --- /dev/null +++ b/samples/timm/beit3_large_patch16_224.pt/model.py @@ -0,0 +1,4234 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_num_prefix_tokens: torch.SymInt, + L_self_modules_fc_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_fc_norm_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_ + l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_num_prefix_tokens = L_self_num_prefix_tokens + l_self_modules_fc_norm_parameters_weight_ = ( + L_self_modules_fc_norm_parameters_weight_ + ) + l_self_modules_fc_norm_parameters_bias_ = ( + L_self_modules_fc_norm_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + x_2 = torch.cat([expand, x_1], dim=1) + expand = x_1 = None + x_3 = x_2 + l_self_parameters_pos_embed_ + x_2 = l_self_parameters_pos_embed_ = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (1024,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 197, 3, 16, 64) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 197, 1024) + transpose_1 = None + x_8 = torch.nn.functional.layer_norm( + x_7, + (1024,), + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_ + ) = None + x_9 = torch._C._nn.linear( + x_8, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_8 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = x_4 + x_10 + x_4 = x_10 = None + x_12 = torch.nn.functional.layer_norm( + x_11, + (1024,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_13 = torch._C._nn.linear( + x_12, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_12 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_14 = torch._C._nn.gelu(x_13, approximate="none") + x_13 = None + x_15 = torch.nn.functional.dropout(x_14, 0.0, False, False) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (4096,), + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_15 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_16 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_18 = torch.nn.functional.dropout(x_17, 0.0, False, False) + x_17 = None + x_19 = x_11 + x_18 + x_11 = x_18 = None + x_20 = torch.nn.functional.layer_norm( + x_19, + (1024,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 197, 3, 16, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_21 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_21.transpose(1, 2) + x_21 = None + x_22 = transpose_2.reshape(1, 197, 1024) + transpose_2 = None + x_23 = torch.nn.functional.layer_norm( + x_22, + (1024,), + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_22 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_ + ) = None + x_24 = torch._C._nn.linear( + x_23, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_23 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_25 = torch.nn.functional.dropout(x_24, 0.0, False, False) + x_24 = None + x_26 = x_19 + x_25 + x_19 = x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (1024,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_29 = torch._C._nn.gelu(x_28, approximate="none") + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (4096,), + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_30 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_32 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + x_34 = x_26 + x_33 + x_26 = x_33 = None + x_35 = torch.nn.functional.layer_norm( + x_34, + (1024,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_35, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_35 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 197, 3, 16, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_36 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_36.transpose(1, 2) + x_36 = None + x_37 = transpose_3.reshape(1, 197, 1024) + transpose_3 = None + x_38 = torch.nn.functional.layer_norm( + x_37, + (1024,), + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_ + ) = None + x_39 = torch._C._nn.linear( + x_38, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_38 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = x_34 + x_40 + x_34 = x_40 = None + x_42 = torch.nn.functional.layer_norm( + x_41, + (1024,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_43 = torch._C._nn.linear( + x_42, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_42 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_44 = torch._C._nn.gelu(x_43, approximate="none") + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = torch.nn.functional.layer_norm( + x_45, + (4096,), + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_45 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_41 + x_48 + x_41 = x_48 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (1024,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 197, 3, 16, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_51 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_51.transpose(1, 2) + x_51 = None + x_52 = transpose_4.reshape(1, 197, 1024) + transpose_4 = None + x_53 = torch.nn.functional.layer_norm( + x_52, + (1024,), + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_52 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_ + ) = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = x_49 + x_55 + x_49 = x_55 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (1024,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_58 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_59 = torch._C._nn.gelu(x_58, approximate="none") + x_58 = None + x_60 = torch.nn.functional.dropout(x_59, 0.0, False, False) + x_59 = None + x_61 = torch.nn.functional.layer_norm( + x_60, + (4096,), + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_60 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_62 = torch._C._nn.linear( + x_61, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_61 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_63 = torch.nn.functional.dropout(x_62, 0.0, False, False) + x_62 = None + x_64 = x_56 + x_63 + x_56 = x_63 = None + x_65 = torch.nn.functional.layer_norm( + x_64, + (1024,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_65, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_65 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 197, 3, 16, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_66 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_66.transpose(1, 2) + x_66 = None + x_67 = transpose_5.reshape(1, 197, 1024) + transpose_5 = None + x_68 = torch.nn.functional.layer_norm( + x_67, + (1024,), + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_67 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_ + ) = None + x_69 = torch._C._nn.linear( + x_68, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_68 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_71 = x_64 + x_70 + x_64 = x_70 = None + x_72 = torch.nn.functional.layer_norm( + x_71, + (1024,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_74 = torch._C._nn.gelu(x_73, approximate="none") + x_73 = None + x_75 = torch.nn.functional.dropout(x_74, 0.0, False, False) + x_74 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (4096,), + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_75 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_78 = torch.nn.functional.dropout(x_77, 0.0, False, False) + x_77 = None + x_79 = x_71 + x_78 + x_71 = x_78 = None + x_80 = torch.nn.functional.layer_norm( + x_79, + (1024,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_80, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_80 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 197, 3, 16, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_81 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_81.transpose(1, 2) + x_81 = None + x_82 = transpose_6.reshape(1, 197, 1024) + transpose_6 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (1024,), + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_82 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_ + ) = None + x_84 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_85 = torch.nn.functional.dropout(x_84, 0.0, False, False) + x_84 = None + x_86 = x_79 + x_85 + x_79 = x_85 = None + x_87 = torch.nn.functional.layer_norm( + x_86, + (1024,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_88 = torch._C._nn.linear( + x_87, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_87 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_89 = torch._C._nn.gelu(x_88, approximate="none") + x_88 = None + x_90 = torch.nn.functional.dropout(x_89, 0.0, False, False) + x_89 = None + x_91 = torch.nn.functional.layer_norm( + x_90, + (4096,), + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_90 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_92 = torch._C._nn.linear( + x_91, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_91 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_93 = torch.nn.functional.dropout(x_92, 0.0, False, False) + x_92 = None + x_94 = x_86 + x_93 + x_86 = x_93 = None + x_95 = torch.nn.functional.layer_norm( + x_94, + (1024,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_95, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_95 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 197, 3, 16, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_96 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_96.transpose(1, 2) + x_96 = None + x_97 = transpose_7.reshape(1, 197, 1024) + transpose_7 = None + x_98 = torch.nn.functional.layer_norm( + x_97, + (1024,), + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_97 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_ + ) = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + x_101 = x_94 + x_100 + x_94 = x_100 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (1024,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch.nn.functional.layer_norm( + x_105, + (4096,), + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_105 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_107 = torch._C._nn.linear( + x_106, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_106 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_101 + x_108 + x_101 = x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (1024,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_110, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_110 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 197, 3, 16, 64) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_111 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_111.transpose(1, 2) + x_111 = None + x_112 = transpose_8.reshape(1, 197, 1024) + transpose_8 = None + x_113 = torch.nn.functional.layer_norm( + x_112, + (1024,), + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_112 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_ + ) = None + x_114 = torch._C._nn.linear( + x_113, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_113 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_115 = torch.nn.functional.dropout(x_114, 0.0, False, False) + x_114 = None + x_116 = x_109 + x_115 + x_109 = x_115 = None + x_117 = torch.nn.functional.layer_norm( + x_116, + (1024,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_118 = torch._C._nn.linear( + x_117, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_117 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_119 = torch._C._nn.gelu(x_118, approximate="none") + x_118 = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (4096,), + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_120 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_121 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) + x_122 = None + x_124 = x_116 + x_123 + x_116 = x_123 = None + x_125 = torch.nn.functional.layer_norm( + x_124, + (1024,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_125, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_125 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 197, 3, 16, 64) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_126 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_126.transpose(1, 2) + x_126 = None + x_127 = transpose_9.reshape(1, 197, 1024) + transpose_9 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (1024,), + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_127 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) + x_129 = None + x_131 = x_124 + x_130 + x_124 = x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (1024,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_134 = torch._C._nn.gelu(x_133, approximate="none") + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = torch.nn.functional.layer_norm( + x_135, + (4096,), + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_135 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_131 + x_138 + x_131 = x_138 = None + x_140 = torch.nn.functional.layer_norm( + x_139, + (1024,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_140, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_140 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 197, 3, 16, 64) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_141 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_141.transpose(1, 2) + x_141 = None + x_142 = transpose_10.reshape(1, 197, 1024) + transpose_10 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (1024,), + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_142 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_ + ) = None + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_143 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_145 = torch.nn.functional.dropout(x_144, 0.0, False, False) + x_144 = None + x_146 = x_139 + x_145 + x_139 = x_145 = None + x_147 = torch.nn.functional.layer_norm( + x_146, + (1024,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_147 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_149 = torch._C._nn.gelu(x_148, approximate="none") + x_148 = None + x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) + x_149 = None + x_151 = torch.nn.functional.layer_norm( + x_150, + (4096,), + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_150 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_152 = torch._C._nn.linear( + x_151, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_151 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_153 = torch.nn.functional.dropout(x_152, 0.0, False, False) + x_152 = None + x_154 = x_146 + x_153 + x_146 = x_153 = None + x_155 = torch.nn.functional.layer_norm( + x_154, + (1024,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_155, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_155 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 197, 3, 16, 64) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_156 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_156.transpose(1, 2) + x_156 = None + x_157 = transpose_11.reshape(1, 197, 1024) + transpose_11 = None + x_158 = torch.nn.functional.layer_norm( + x_157, + (1024,), + l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_157 = l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_ + ) = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_158 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_154 + x_160 + x_154 = x_160 = None + x_162 = torch.nn.functional.layer_norm( + x_161, + (1024,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_163 = torch._C._nn.linear( + x_162, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_162 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_164 = torch._C._nn.gelu(x_163, approximate="none") + x_163 = None + x_165 = torch.nn.functional.dropout(x_164, 0.0, False, False) + x_164 = None + x_166 = torch.nn.functional.layer_norm( + x_165, + (4096,), + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_165 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_167 = torch._C._nn.linear( + x_166, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_166 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_169 = x_161 + x_168 + x_161 = x_168 = None + x_170 = torch.nn.functional.layer_norm( + x_169, + (1024,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_170, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_170 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 197, 3, 16, 64) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_171 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_171.transpose(1, 2) + x_171 = None + x_172 = transpose_12.reshape(1, 197, 1024) + transpose_12 = None + x_173 = torch.nn.functional.layer_norm( + x_172, + (1024,), + l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_172 = l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_ + ) = None + x_174 = torch._C._nn.linear( + x_173, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_173 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_175 = torch.nn.functional.dropout(x_174, 0.0, False, False) + x_174 = None + x_176 = x_169 + x_175 + x_169 = x_175 = None + x_177 = torch.nn.functional.layer_norm( + x_176, + (1024,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_178 = torch._C._nn.linear( + x_177, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_177 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_179 = torch._C._nn.gelu(x_178, approximate="none") + x_178 = None + x_180 = torch.nn.functional.dropout(x_179, 0.0, False, False) + x_179 = None + x_181 = torch.nn.functional.layer_norm( + x_180, + (4096,), + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_180 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_182 = torch._C._nn.linear( + x_181, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_181 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_183 = torch.nn.functional.dropout(x_182, 0.0, False, False) + x_182 = None + x_184 = x_176 + x_183 + x_176 = x_183 = None + x_185 = torch.nn.functional.layer_norm( + x_184, + (1024,), + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ + ) = None + linear_48 = torch._C._nn.linear( + x_185, + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_, + ) + x_185 = ( + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_24 = linear_48.reshape(1, 197, 3, 16, 64) + linear_48 = None + qkv_12 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_12 = unbind_12[0] + k_12 = unbind_12[1] + v_12 = unbind_12[2] + unbind_12 = None + x_186 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_13 = x_186.transpose(1, 2) + x_186 = None + x_187 = transpose_13.reshape(1, 197, 1024) + transpose_13 = None + x_188 = torch.nn.functional.layer_norm( + x_187, + (1024,), + l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_187 = l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_ + ) = None + x_189 = torch._C._nn.linear( + x_188, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_, + ) + x_188 = l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ + ) = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = x_184 + x_190 + x_184 = x_190 = None + x_192 = torch.nn.functional.layer_norm( + x_191, + (1024,), + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ + ) = None + x_193 = torch._C._nn.linear( + x_192, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_192 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_194 = torch._C._nn.gelu(x_193, approximate="none") + x_193 = None + x_195 = torch.nn.functional.dropout(x_194, 0.0, False, False) + x_194 = None + x_196 = torch.nn.functional.layer_norm( + x_195, + (4096,), + l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_195 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_197 = torch._C._nn.linear( + x_196, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_196 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_198 = torch.nn.functional.dropout(x_197, 0.0, False, False) + x_197 = None + x_199 = x_191 + x_198 + x_191 = x_198 = None + x_200 = torch.nn.functional.layer_norm( + x_199, + (1024,), + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ + ) = None + linear_52 = torch._C._nn.linear( + x_200, + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_, + ) + x_200 = ( + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_26 = linear_52.reshape(1, 197, 3, 16, 64) + linear_52 = None + qkv_13 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_13 = unbind_13[0] + k_13 = unbind_13[1] + v_13 = unbind_13[2] + unbind_13 = None + x_201 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_14 = x_201.transpose(1, 2) + x_201 = None + x_202 = transpose_14.reshape(1, 197, 1024) + transpose_14 = None + x_203 = torch.nn.functional.layer_norm( + x_202, + (1024,), + l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_202 = l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_ + ) = None + x_204 = torch._C._nn.linear( + x_203, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_, + ) + x_203 = l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ + ) = None + x_205 = torch.nn.functional.dropout(x_204, 0.0, False, False) + x_204 = None + x_206 = x_199 + x_205 + x_199 = x_205 = None + x_207 = torch.nn.functional.layer_norm( + x_206, + (1024,), + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ + ) = None + x_208 = torch._C._nn.linear( + x_207, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_207 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_209 = torch._C._nn.gelu(x_208, approximate="none") + x_208 = None + x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) + x_209 = None + x_211 = torch.nn.functional.layer_norm( + x_210, + (4096,), + l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_210 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_212 = torch._C._nn.linear( + x_211, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_211 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_213 = torch.nn.functional.dropout(x_212, 0.0, False, False) + x_212 = None + x_214 = x_206 + x_213 + x_206 = x_213 = None + x_215 = torch.nn.functional.layer_norm( + x_214, + (1024,), + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ + ) = None + linear_56 = torch._C._nn.linear( + x_215, + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_, + ) + x_215 = ( + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_28 = linear_56.reshape(1, 197, 3, 16, 64) + linear_56 = None + qkv_14 = reshape_28.permute(2, 0, 3, 1, 4) + reshape_28 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_14 = unbind_14[0] + k_14 = unbind_14[1] + v_14 = unbind_14[2] + unbind_14 = None + x_216 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_15 = x_216.transpose(1, 2) + x_216 = None + x_217 = transpose_15.reshape(1, 197, 1024) + transpose_15 = None + x_218 = torch.nn.functional.layer_norm( + x_217, + (1024,), + l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_217 = l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_ + ) = None + x_219 = torch._C._nn.linear( + x_218, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_, + ) + x_218 = l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ + ) = None + x_220 = torch.nn.functional.dropout(x_219, 0.0, False, False) + x_219 = None + x_221 = x_214 + x_220 + x_214 = x_220 = None + x_222 = torch.nn.functional.layer_norm( + x_221, + (1024,), + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ + ) = None + x_223 = torch._C._nn.linear( + x_222, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_222 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_224 = torch._C._nn.gelu(x_223, approximate="none") + x_223 = None + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = torch.nn.functional.layer_norm( + x_225, + (4096,), + l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_225 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_227 = torch._C._nn.linear( + x_226, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_226 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_228 = torch.nn.functional.dropout(x_227, 0.0, False, False) + x_227 = None + x_229 = x_221 + x_228 + x_221 = x_228 = None + x_230 = torch.nn.functional.layer_norm( + x_229, + (1024,), + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ + ) = None + linear_60 = torch._C._nn.linear( + x_230, + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_, + ) + x_230 = ( + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_30 = linear_60.reshape(1, 197, 3, 16, 64) + linear_60 = None + qkv_15 = reshape_30.permute(2, 0, 3, 1, 4) + reshape_30 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_15 = unbind_15[0] + k_15 = unbind_15[1] + v_15 = unbind_15[2] + unbind_15 = None + x_231 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_16 = x_231.transpose(1, 2) + x_231 = None + x_232 = transpose_16.reshape(1, 197, 1024) + transpose_16 = None + x_233 = torch.nn.functional.layer_norm( + x_232, + (1024,), + l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_232 = l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_ + ) = None + x_234 = torch._C._nn.linear( + x_233, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_, + ) + x_233 = l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ + ) = None + x_235 = torch.nn.functional.dropout(x_234, 0.0, False, False) + x_234 = None + x_236 = x_229 + x_235 + x_229 = x_235 = None + x_237 = torch.nn.functional.layer_norm( + x_236, + (1024,), + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ + ) = None + x_238 = torch._C._nn.linear( + x_237, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_237 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_239 = torch._C._nn.gelu(x_238, approximate="none") + x_238 = None + x_240 = torch.nn.functional.dropout(x_239, 0.0, False, False) + x_239 = None + x_241 = torch.nn.functional.layer_norm( + x_240, + (4096,), + l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_240 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_242 = torch._C._nn.linear( + x_241, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_241 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_243 = torch.nn.functional.dropout(x_242, 0.0, False, False) + x_242 = None + x_244 = x_236 + x_243 + x_236 = x_243 = None + x_245 = torch.nn.functional.layer_norm( + x_244, + (1024,), + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ + ) = None + linear_64 = torch._C._nn.linear( + x_245, + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_, + ) + x_245 = ( + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_32 = linear_64.reshape(1, 197, 3, 16, 64) + linear_64 = None + qkv_16 = reshape_32.permute(2, 0, 3, 1, 4) + reshape_32 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_16 = unbind_16[0] + k_16 = unbind_16[1] + v_16 = unbind_16[2] + unbind_16 = None + x_246 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, attn_mask=None, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_17 = x_246.transpose(1, 2) + x_246 = None + x_247 = transpose_17.reshape(1, 197, 1024) + transpose_17 = None + x_248 = torch.nn.functional.layer_norm( + x_247, + (1024,), + l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_247 = l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_ + ) = None + x_249 = torch._C._nn.linear( + x_248, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_, + ) + x_248 = l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ + ) = None + x_250 = torch.nn.functional.dropout(x_249, 0.0, False, False) + x_249 = None + x_251 = x_244 + x_250 + x_244 = x_250 = None + x_252 = torch.nn.functional.layer_norm( + x_251, + (1024,), + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ + ) = None + x_253 = torch._C._nn.linear( + x_252, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_252 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_254 = torch._C._nn.gelu(x_253, approximate="none") + x_253 = None + x_255 = torch.nn.functional.dropout(x_254, 0.0, False, False) + x_254 = None + x_256 = torch.nn.functional.layer_norm( + x_255, + (4096,), + l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_255 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_257 = torch._C._nn.linear( + x_256, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_256 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_258 = torch.nn.functional.dropout(x_257, 0.0, False, False) + x_257 = None + x_259 = x_251 + x_258 + x_251 = x_258 = None + x_260 = torch.nn.functional.layer_norm( + x_259, + (1024,), + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ + ) = None + linear_68 = torch._C._nn.linear( + x_260, + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_, + ) + x_260 = ( + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_34 = linear_68.reshape(1, 197, 3, 16, 64) + linear_68 = None + qkv_17 = reshape_34.permute(2, 0, 3, 1, 4) + reshape_34 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_17 = unbind_17[0] + k_17 = unbind_17[1] + v_17 = unbind_17[2] + unbind_17 = None + x_261 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, attn_mask=None, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_18 = x_261.transpose(1, 2) + x_261 = None + x_262 = transpose_18.reshape(1, 197, 1024) + transpose_18 = None + x_263 = torch.nn.functional.layer_norm( + x_262, + (1024,), + l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_262 = l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_ + ) = None + x_264 = torch._C._nn.linear( + x_263, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_, + ) + x_263 = l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ + ) = None + x_265 = torch.nn.functional.dropout(x_264, 0.0, False, False) + x_264 = None + x_266 = x_259 + x_265 + x_259 = x_265 = None + x_267 = torch.nn.functional.layer_norm( + x_266, + (1024,), + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ + ) = None + x_268 = torch._C._nn.linear( + x_267, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_267 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_269 = torch._C._nn.gelu(x_268, approximate="none") + x_268 = None + x_270 = torch.nn.functional.dropout(x_269, 0.0, False, False) + x_269 = None + x_271 = torch.nn.functional.layer_norm( + x_270, + (4096,), + l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_270 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_272 = torch._C._nn.linear( + x_271, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_271 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_273 = torch.nn.functional.dropout(x_272, 0.0, False, False) + x_272 = None + x_274 = x_266 + x_273 + x_266 = x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (1024,), + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ + ) = None + linear_72 = torch._C._nn.linear( + x_275, + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_, + ) + x_275 = ( + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_36 = linear_72.reshape(1, 197, 3, 16, 64) + linear_72 = None + qkv_18 = reshape_36.permute(2, 0, 3, 1, 4) + reshape_36 = None + unbind_18 = qkv_18.unbind(0) + qkv_18 = None + q_18 = unbind_18[0] + k_18 = unbind_18[1] + v_18 = unbind_18[2] + unbind_18 = None + x_276 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_19 = x_276.transpose(1, 2) + x_276 = None + x_277 = transpose_19.reshape(1, 197, 1024) + transpose_19 = None + x_278 = torch.nn.functional.layer_norm( + x_277, + (1024,), + l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_277 = l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_ + ) = None + x_279 = torch._C._nn.linear( + x_278, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_, + ) + x_278 = l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ + ) = None + x_280 = torch.nn.functional.dropout(x_279, 0.0, False, False) + x_279 = None + x_281 = x_274 + x_280 + x_274 = x_280 = None + x_282 = torch.nn.functional.layer_norm( + x_281, + (1024,), + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ + ) = None + x_283 = torch._C._nn.linear( + x_282, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_282 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_284 = torch._C._nn.gelu(x_283, approximate="none") + x_283 = None + x_285 = torch.nn.functional.dropout(x_284, 0.0, False, False) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (4096,), + l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_285 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_286 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) + x_287 = None + x_289 = x_281 + x_288 + x_281 = x_288 = None + x_290 = torch.nn.functional.layer_norm( + x_289, + (1024,), + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ + ) = None + linear_76 = torch._C._nn.linear( + x_290, + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_, + ) + x_290 = ( + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_38 = linear_76.reshape(1, 197, 3, 16, 64) + linear_76 = None + qkv_19 = reshape_38.permute(2, 0, 3, 1, 4) + reshape_38 = None + unbind_19 = qkv_19.unbind(0) + qkv_19 = None + q_19 = unbind_19[0] + k_19 = unbind_19[1] + v_19 = unbind_19[2] + unbind_19 = None + x_291 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, attn_mask=None, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_20 = x_291.transpose(1, 2) + x_291 = None + x_292 = transpose_20.reshape(1, 197, 1024) + transpose_20 = None + x_293 = torch.nn.functional.layer_norm( + x_292, + (1024,), + l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_292 = l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_ + ) = None + x_294 = torch._C._nn.linear( + x_293, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_, + ) + x_293 = l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ + ) = None + x_295 = torch.nn.functional.dropout(x_294, 0.0, False, False) + x_294 = None + x_296 = x_289 + x_295 + x_289 = x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (1024,), + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ + ) = None + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_299 = torch._C._nn.gelu(x_298, approximate="none") + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_301 = torch.nn.functional.layer_norm( + x_300, + (4096,), + l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_300 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_302 = torch._C._nn.linear( + x_301, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_301 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_303 = torch.nn.functional.dropout(x_302, 0.0, False, False) + x_302 = None + x_304 = x_296 + x_303 + x_296 = x_303 = None + x_305 = torch.nn.functional.layer_norm( + x_304, + (1024,), + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ + ) = None + linear_80 = torch._C._nn.linear( + x_305, + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_, + ) + x_305 = ( + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_40 = linear_80.reshape(1, 197, 3, 16, 64) + linear_80 = None + qkv_20 = reshape_40.permute(2, 0, 3, 1, 4) + reshape_40 = None + unbind_20 = qkv_20.unbind(0) + qkv_20 = None + q_20 = unbind_20[0] + k_20 = unbind_20[1] + v_20 = unbind_20[2] + unbind_20 = None + x_306 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, attn_mask=None, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_21 = x_306.transpose(1, 2) + x_306 = None + x_307 = transpose_21.reshape(1, 197, 1024) + transpose_21 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (1024,), + l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_307 = l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_ + ) = None + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_, + ) + x_308 = l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ + ) = None + x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) + x_309 = None + x_311 = x_304 + x_310 + x_304 = x_310 = None + x_312 = torch.nn.functional.layer_norm( + x_311, + (1024,), + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ + ) = None + x_313 = torch._C._nn.linear( + x_312, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_312 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_314 = torch._C._nn.gelu(x_313, approximate="none") + x_313 = None + x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) + x_314 = None + x_316 = torch.nn.functional.layer_norm( + x_315, + (4096,), + l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_315 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_317 = torch._C._nn.linear( + x_316, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_316 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_318 = torch.nn.functional.dropout(x_317, 0.0, False, False) + x_317 = None + x_319 = x_311 + x_318 + x_311 = x_318 = None + x_320 = torch.nn.functional.layer_norm( + x_319, + (1024,), + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ + ) = None + linear_84 = torch._C._nn.linear( + x_320, + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_, + ) + x_320 = ( + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_42 = linear_84.reshape(1, 197, 3, 16, 64) + linear_84 = None + qkv_21 = reshape_42.permute(2, 0, 3, 1, 4) + reshape_42 = None + unbind_21 = qkv_21.unbind(0) + qkv_21 = None + q_21 = unbind_21[0] + k_21 = unbind_21[1] + v_21 = unbind_21[2] + unbind_21 = None + x_321 = torch._C._nn.scaled_dot_product_attention( + q_21, k_21, v_21, attn_mask=None, dropout_p=0.0 + ) + q_21 = k_21 = v_21 = None + transpose_22 = x_321.transpose(1, 2) + x_321 = None + x_322 = transpose_22.reshape(1, 197, 1024) + transpose_22 = None + x_323 = torch.nn.functional.layer_norm( + x_322, + (1024,), + l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_322 = l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_ + ) = None + x_324 = torch._C._nn.linear( + x_323, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_, + ) + x_323 = l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ + ) = None + x_325 = torch.nn.functional.dropout(x_324, 0.0, False, False) + x_324 = None + x_326 = x_319 + x_325 + x_319 = x_325 = None + x_327 = torch.nn.functional.layer_norm( + x_326, + (1024,), + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ + ) = None + x_328 = torch._C._nn.linear( + x_327, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_327 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_329 = torch._C._nn.gelu(x_328, approximate="none") + x_328 = None + x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) + x_329 = None + x_331 = torch.nn.functional.layer_norm( + x_330, + (4096,), + l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_330 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_332 = torch._C._nn.linear( + x_331, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_331 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_334 = x_326 + x_333 + x_326 = x_333 = None + x_335 = torch.nn.functional.layer_norm( + x_334, + (1024,), + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ + ) = None + linear_88 = torch._C._nn.linear( + x_335, + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_, + ) + x_335 = ( + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_44 = linear_88.reshape(1, 197, 3, 16, 64) + linear_88 = None + qkv_22 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_22 = qkv_22.unbind(0) + qkv_22 = None + q_22 = unbind_22[0] + k_22 = unbind_22[1] + v_22 = unbind_22[2] + unbind_22 = None + x_336 = torch._C._nn.scaled_dot_product_attention( + q_22, k_22, v_22, attn_mask=None, dropout_p=0.0 + ) + q_22 = k_22 = v_22 = None + transpose_23 = x_336.transpose(1, 2) + x_336 = None + x_337 = transpose_23.reshape(1, 197, 1024) + transpose_23 = None + x_338 = torch.nn.functional.layer_norm( + x_337, + (1024,), + l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_337 = l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_ + ) = None + x_339 = torch._C._nn.linear( + x_338, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_, + ) + x_338 = l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ + ) = None + x_340 = torch.nn.functional.dropout(x_339, 0.0, False, False) + x_339 = None + x_341 = x_334 + x_340 + x_334 = x_340 = None + x_342 = torch.nn.functional.layer_norm( + x_341, + (1024,), + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ + ) = None + x_343 = torch._C._nn.linear( + x_342, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_342 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_344 = torch._C._nn.gelu(x_343, approximate="none") + x_343 = None + x_345 = torch.nn.functional.dropout(x_344, 0.0, False, False) + x_344 = None + x_346 = torch.nn.functional.layer_norm( + x_345, + (4096,), + l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_345 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_347 = torch._C._nn.linear( + x_346, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_346 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) + x_347 = None + x_349 = x_341 + x_348 + x_341 = x_348 = None + x_350 = torch.nn.functional.layer_norm( + x_349, + (1024,), + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ + ) = None + linear_92 = torch._C._nn.linear( + x_350, + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_, + ) + x_350 = ( + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_46 = linear_92.reshape(1, 197, 3, 16, 64) + linear_92 = None + qkv_23 = reshape_46.permute(2, 0, 3, 1, 4) + reshape_46 = None + unbind_23 = qkv_23.unbind(0) + qkv_23 = None + q_23 = unbind_23[0] + k_23 = unbind_23[1] + v_23 = unbind_23[2] + unbind_23 = None + x_351 = torch._C._nn.scaled_dot_product_attention( + q_23, k_23, v_23, attn_mask=None, dropout_p=0.0 + ) + q_23 = k_23 = v_23 = None + transpose_24 = x_351.transpose(1, 2) + x_351 = None + x_352 = transpose_24.reshape(1, 197, 1024) + transpose_24 = None + x_353 = torch.nn.functional.layer_norm( + x_352, + (1024,), + l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_, + 1e-05, + ) + x_352 = l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_ + ) = None + x_354 = torch._C._nn.linear( + x_353, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_, + ) + x_353 = l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ + ) = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_356 = x_349 + x_355 + x_349 = x_355 = None + x_357 = torch.nn.functional.layer_norm( + x_356, + (1024,), + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_, + 1e-05, + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ + ) = None + x_358 = torch._C._nn.linear( + x_357, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_357 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_359 = torch._C._nn.gelu(x_358, approximate="none") + x_358 = None + x_360 = torch.nn.functional.dropout(x_359, 0.0, False, False) + x_359 = None + x_361 = torch.nn.functional.layer_norm( + x_360, + (4096,), + l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_, + 1e-05, + ) + x_360 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_ + ) = None + x_362 = torch._C._nn.linear( + x_361, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_361 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) + x_362 = None + x_364 = x_356 + x_363 + x_356 = x_363 = None + x_365 = x_364[ + (slice(None, None, None), slice(l_self_num_prefix_tokens, None, None)) + ] + x_364 = l_self_num_prefix_tokens = None + x_366 = x_365.mean(dim=1) + x_365 = None + x_367 = torch.nn.functional.layer_norm( + x_366, + (1024,), + l_self_modules_fc_norm_parameters_weight_, + l_self_modules_fc_norm_parameters_bias_, + 1e-05, + ) + x_366 = ( + l_self_modules_fc_norm_parameters_weight_ + ) = l_self_modules_fc_norm_parameters_bias_ = None + x_368 = torch.nn.functional.dropout(x_367, 0.0, False, False) + x_367 = None + return (x_368,) diff --git a/samples/timm/beit3_large_patch16_224.pt/weight_meta.py b/samples/timm/beit3_large_patch16_224.pt/weight_meta.py new file mode 100644 index 0000000000..6f3c28f89d --- /dev/null +++ b/samples/timm/beit3_large_patch16_224.pt/weight_meta.py @@ -0,0 +1,4070 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.228 + std = 1.284 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [1024, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 197, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_" + ) + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_num_prefix_tokens: + name = "L_self_num_prefix_tokens" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_self_modules_fc_norm_parameters_weight_: + name = "L_self_modules_fc_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_fc_norm_parameters_bias_: + name = "L_self_modules_fc_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/caformer_b36.sail_in1k/graph_net.json b/samples/timm/caformer_b36.sail_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/caformer_b36.sail_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/caformer_b36.sail_in1k/input_meta.py b/samples/timm/caformer_b36.sail_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_b36.sail_in1k/input_tensor_constraints.py b/samples/timm/caformer_b36.sail_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_b36.sail_in1k/model.py b/samples/timm/caformer_b36.sail_in1k/model.py new file mode 100644 index 0000000000..08d3de35cc --- /dev/null +++ b/samples/timm/caformer_b36.sail_in1k/model.py @@ -0,0 +1,4758 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv_parameters_bias_ = ( + L_self_modules_stem_modules_conv_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_norm_parameters_weight_ = ( + L_self_modules_stem_modules_norm_parameters_weight_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv_parameters_weight_, + l_self_modules_stem_modules_conv_parameters_bias_, + (4, 4), + (2, 2), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_conv_parameters_weight_ + ) = l_self_modules_stem_modules_conv_parameters_bias_ = None + x_1 = x.permute(0, 2, 3, 1) + x = None + x_2 = torch.nn.functional.layer_norm( + x_1, + (128,), + l_self_modules_stem_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None + x_3 = x_2.permute(0, 3, 1, 2) + x_2 = None + sym_sum = torch.sym_sum([-3, s1]) + s1 = None + floordiv = sym_sum // 4 + sym_sum_1 = torch.sym_sum([1, floordiv]) + floordiv = sym_sum_1 = None + x_4 = x_3.permute(0, 2, 3, 1) + x_5 = torch.nn.functional.layer_norm( + x_4, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu = torch.nn.functional.relu(x_7, inplace=False) + x_7 = None + pow_1 = relu**2 + relu = None + mul = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_1 + ) = None + x_8 = ( + mul + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_11 = x_3 + x_10 + x_3 = x_10 = None + x_12 = x_11.permute(0, 2, 3, 1) + x_13 = torch.nn.functional.layer_norm( + x_12, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_14 = x_13.permute(0, 3, 1, 2) + x_13 = None + x_15 = torch.conv2d( + x_14, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_1 = torch.nn.functional.relu(x_15, inplace=False) + x_15 = None + pow_2 = relu_1**2 + relu_1 = None + mul_1 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_2 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_2 + ) = None + x_16 = ( + mul_1 + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = x_11 + x_19 + x_11 = x_19 = None + x_21 = x_20.permute(0, 2, 3, 1) + x_22 = torch.nn.functional.layer_norm( + x_21, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = torch.conv2d( + x_23, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_2 = torch.nn.functional.relu(x_24, inplace=False) + x_24 = None + pow_3 = relu_2**2 + relu_2 = None + mul_2 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_3 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_3 + ) = None + x_25 = ( + mul_2 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_27 = torch.conv2d( + x_26, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_28 = x_20 + x_27 + x_20 = x_27 = None + x_29 = x_28.permute(0, 2, 3, 1) + x_30 = torch.nn.functional.layer_norm( + x_29, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_3 = torch.nn.functional.relu(x_32, inplace=False) + x_32 = None + pow_4 = relu_3**2 + relu_3 = None + mul_3 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_4 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_4 + ) = None + x_33 = ( + mul_3 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + x_37 = x_28 + x_36 + x_28 = x_36 = None + x_38 = x_37.permute(0, 2, 3, 1) + x_39 = torch.nn.functional.layer_norm( + x_38, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_40 = x_39.permute(0, 3, 1, 2) + x_39 = None + x_41 = torch.conv2d( + x_40, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_4 = torch.nn.functional.relu(x_41, inplace=False) + x_41 = None + pow_5 = relu_4**2 + relu_4 = None + mul_4 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_5 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_5 + ) = None + x_42 = ( + mul_4 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_44 = torch.conv2d( + x_43, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_45 = x_37 + x_44 + x_37 = x_44 = None + x_46 = x_45.permute(0, 2, 3, 1) + x_47 = torch.nn.functional.layer_norm( + x_46, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_5 = torch.nn.functional.relu(x_49, inplace=False) + x_49 = None + pow_6 = relu_5**2 + relu_5 = None + mul_5 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_6 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_6 + ) = None + x_50 = ( + mul_5 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = x_45 + x_53 + x_45 = x_53 = None + x_55 = x_54.permute(0, 2, 3, 1) + x_54 = None + x_56 = torch.nn.functional.layer_norm( + x_55, + (128,), + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) + x_57 = x_56.permute(0, 3, 1, 2) + x_56 = None + x_58 = torch.conv2d( + x_57, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_1 = sym_sum // 8 + sym_sum_2 = torch.sym_sum([1, floordiv_1]) + floordiv_1 = sym_sum_2 = None + x_59 = x_58.permute(0, 2, 3, 1) + x_60 = torch.nn.functional.layer_norm( + x_59, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_61 = x_60.permute(0, 3, 1, 2) + x_60 = None + x_62 = torch.conv2d( + x_61, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_6 = torch.nn.functional.relu(x_62, inplace=False) + x_62 = None + pow_7 = relu_6**2 + relu_6 = None + mul_6 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_7 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_7 + ) = None + x_63 = ( + mul_6 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_66 = x_58 + x_65 + x_58 = x_65 = None + x_67 = x_66.permute(0, 2, 3, 1) + x_68 = torch.nn.functional.layer_norm( + x_67, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + x_70 = torch.conv2d( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_7 = torch.nn.functional.relu(x_70, inplace=False) + x_70 = None + pow_8 = relu_7**2 + relu_7 = None + mul_7 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_8 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_8 + ) = None + x_71 = ( + mul_7 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + x_73 = torch.conv2d( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + x_75 = x_66 + x_74 + x_66 = x_74 = None + x_76 = x_75.permute(0, 2, 3, 1) + x_77 = torch.nn.functional.layer_norm( + x_76, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_8 = torch.nn.functional.relu(x_79, inplace=False) + x_79 = None + pow_9 = relu_8**2 + relu_8 = None + mul_8 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_9 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_9 + ) = None + x_80 = ( + mul_8 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_81 = torch.conv2d( + x_80, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_83 = x_75 + x_82 + x_75 = x_82 = None + x_84 = x_83.permute(0, 2, 3, 1) + x_85 = torch.nn.functional.layer_norm( + x_84, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_86 = x_85.permute(0, 3, 1, 2) + x_85 = None + x_87 = torch.conv2d( + x_86, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_9 = torch.nn.functional.relu(x_87, inplace=False) + x_87 = None + pow_10 = relu_9**2 + relu_9 = None + mul_9 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_10 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_10 + ) = None + x_88 = ( + mul_9 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = x_83 + x_91 + x_83 = x_91 = None + x_93 = x_92.permute(0, 2, 3, 1) + x_94 = torch.nn.functional.layer_norm( + x_93, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = torch.conv2d( + x_95, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_10 = torch.nn.functional.relu(x_96, inplace=False) + x_96 = None + pow_11 = relu_10**2 + relu_10 = None + mul_10 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_11 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_11 + ) = None + x_97 = ( + mul_10 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_98 = torch.conv2d( + x_97, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_99 = torch.conv2d( + x_98, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_100 = x_92 + x_99 + x_92 = x_99 = None + x_101 = x_100.permute(0, 2, 3, 1) + x_102 = torch.nn.functional.layer_norm( + x_101, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_103 = x_102.permute(0, 3, 1, 2) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_11 = torch.nn.functional.relu(x_104, inplace=False) + x_104 = None + pow_12 = relu_11**2 + relu_11 = None + mul_11 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_12 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_12 + ) = None + x_105 = ( + mul_11 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_100 + x_108 + x_100 = x_108 = None + x_110 = x_109.permute(0, 2, 3, 1) + x_111 = torch.nn.functional.layer_norm( + x_110, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_110 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) + x_112 = x_111.permute(0, 3, 1, 2) + x_111 = None + x_113 = torch.conv2d( + x_112, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_112 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_12 = torch.nn.functional.relu(x_113, inplace=False) + x_113 = None + pow_13 = relu_12**2 + relu_12 = None + mul_12 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ + * pow_13 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_13 + ) = None + x_114 = ( + mul_12 + + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_12 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_115 = torch.conv2d( + x_114, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_114 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_117 = x_109 + x_116 + x_109 = x_116 = None + x_118 = x_117.permute(0, 2, 3, 1) + x_119 = torch.nn.functional.layer_norm( + x_118, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_118 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) + x_120 = x_119.permute(0, 3, 1, 2) + x_119 = None + x_121 = torch.conv2d( + x_120, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_120 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_13 = torch.nn.functional.relu(x_121, inplace=False) + x_121 = None + pow_14 = relu_13**2 + relu_13 = None + mul_13 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + * pow_14 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( + pow_14 + ) = None + x_122 = ( + mul_13 + + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + ) + mul_13 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) + x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) + x_122 = None + x_124 = torch.conv2d( + x_123, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_123 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_126 = x_117 + x_125 + x_117 = x_125 = None + x_127 = x_126.permute(0, 2, 3, 1) + x_128 = torch.nn.functional.layer_norm( + x_127, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_127 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) + x_129 = x_128.permute(0, 3, 1, 2) + x_128 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_14 = torch.nn.functional.relu(x_130, inplace=False) + x_130 = None + pow_15 = relu_14**2 + relu_14 = None + mul_14 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ + * pow_15 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_15 + ) = None + x_131 = ( + mul_14 + + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_14 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_132 = torch.conv2d( + x_131, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_131 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_132 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_134 = x_126 + x_133 + x_126 = x_133 = None + x_135 = x_134.permute(0, 2, 3, 1) + x_136 = torch.nn.functional.layer_norm( + x_135, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_135 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) + x_137 = x_136.permute(0, 3, 1, 2) + x_136 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_15 = torch.nn.functional.relu(x_138, inplace=False) + x_138 = None + pow_16 = relu_15**2 + relu_15 = None + mul_15 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + * pow_16 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( + pow_16 + ) = None + x_139 = ( + mul_15 + + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + ) + mul_15 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) + x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) + x_139 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_140 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_134 + x_142 + x_134 = x_142 = None + x_144 = x_143.permute(0, 2, 3, 1) + x_145 = torch.nn.functional.layer_norm( + x_144, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_144 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) + x_146 = x_145.permute(0, 3, 1, 2) + x_145 = None + x_147 = torch.conv2d( + x_146, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_146 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_16 = torch.nn.functional.relu(x_147, inplace=False) + x_147 = None + pow_17 = relu_16**2 + relu_16 = None + mul_16 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ + * pow_17 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_17 + ) = None + x_148 = ( + mul_16 + + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_16 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_149 = torch.conv2d( + x_148, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_148 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_150 = torch.conv2d( + x_149, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_149 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_151 = x_143 + x_150 + x_143 = x_150 = None + x_152 = x_151.permute(0, 2, 3, 1) + x_153 = torch.nn.functional.layer_norm( + x_152, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_152 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) + x_154 = x_153.permute(0, 3, 1, 2) + x_153 = None + x_155 = torch.conv2d( + x_154, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_154 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_17 = torch.nn.functional.relu(x_155, inplace=False) + x_155 = None + pow_18 = relu_17**2 + relu_17 = None + mul_17 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + * pow_18 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( + pow_18 + ) = None + x_156 = ( + mul_17 + + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + ) + mul_17 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch.conv2d( + x_157, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_157 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_151 + x_159 + x_151 = x_159 = None + x_161 = x_160.permute(0, 2, 3, 1) + x_162 = torch.nn.functional.layer_norm( + x_161, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_161 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) + x_163 = x_162.permute(0, 3, 1, 2) + x_162 = None + x_164 = torch.conv2d( + x_163, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_163 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_18 = torch.nn.functional.relu(x_164, inplace=False) + x_164 = None + pow_19 = relu_18**2 + relu_18 = None + mul_18 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ + * pow_19 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_19 + ) = None + x_165 = ( + mul_18 + + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_18 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_166 = torch.conv2d( + x_165, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_165 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_167 = torch.conv2d( + x_166, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_166 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_168 = x_160 + x_167 + x_160 = x_167 = None + x_169 = x_168.permute(0, 2, 3, 1) + x_170 = torch.nn.functional.layer_norm( + x_169, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_169 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) + x_171 = x_170.permute(0, 3, 1, 2) + x_170 = None + x_172 = torch.conv2d( + x_171, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_171 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_19 = torch.nn.functional.relu(x_172, inplace=False) + x_172 = None + pow_20 = relu_19**2 + relu_19 = None + mul_19 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + * pow_20 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( + pow_20 + ) = None + x_173 = ( + mul_19 + + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + ) + mul_19 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) + x_174 = torch.nn.functional.dropout(x_173, 0.0, False, False) + x_173 = None + x_175 = torch.conv2d( + x_174, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_174 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = x_168 + x_176 + x_168 = x_176 = None + x_178 = x_177.permute(0, 2, 3, 1) + x_179 = torch.nn.functional.layer_norm( + x_178, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_178 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) + x_180 = x_179.permute(0, 3, 1, 2) + x_179 = None + x_181 = torch.conv2d( + x_180, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_180 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_20 = torch.nn.functional.relu(x_181, inplace=False) + x_181 = None + pow_21 = relu_20**2 + relu_20 = None + mul_20 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ + * pow_21 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_21 + ) = None + x_182 = ( + mul_20 + + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_20 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_183 = torch.conv2d( + x_182, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_182 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_184 = torch.conv2d( + x_183, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_183 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_185 = x_177 + x_184 + x_177 = x_184 = None + x_186 = x_185.permute(0, 2, 3, 1) + x_187 = torch.nn.functional.layer_norm( + x_186, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) + x_188 = x_187.permute(0, 3, 1, 2) + x_187 = None + x_189 = torch.conv2d( + x_188, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_188 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_21 = torch.nn.functional.relu(x_189, inplace=False) + x_189 = None + pow_22 = relu_21**2 + relu_21 = None + mul_21 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + * pow_22 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( + pow_22 + ) = None + x_190 = ( + mul_21 + + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + ) + mul_21 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) + x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) + x_190 = None + x_192 = torch.conv2d( + x_191, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_191 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_185 + x_193 + x_185 = x_193 = None + x_195 = x_194.permute(0, 2, 3, 1) + x_196 = torch.nn.functional.layer_norm( + x_195, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_195 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) + x_197 = x_196.permute(0, 3, 1, 2) + x_196 = None + x_198 = torch.conv2d( + x_197, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_197 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_22 = torch.nn.functional.relu(x_198, inplace=False) + x_198 = None + pow_23 = relu_22**2 + relu_22 = None + mul_22 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ + * pow_23 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_23 + ) = None + x_199 = ( + mul_22 + + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_22 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_200 = torch.conv2d( + x_199, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_199 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_201 = torch.conv2d( + x_200, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_200 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_202 = x_194 + x_201 + x_194 = x_201 = None + x_203 = x_202.permute(0, 2, 3, 1) + x_204 = torch.nn.functional.layer_norm( + x_203, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_203 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) + x_205 = x_204.permute(0, 3, 1, 2) + x_204 = None + x_206 = torch.conv2d( + x_205, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_205 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_23 = torch.nn.functional.relu(x_206, inplace=False) + x_206 = None + pow_24 = relu_23**2 + relu_23 = None + mul_23 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + * pow_24 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( + pow_24 + ) = None + x_207 = ( + mul_23 + + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + ) + mul_23 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) + x_208 = torch.nn.functional.dropout(x_207, 0.0, False, False) + x_207 = None + x_209 = torch.conv2d( + x_208, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_208 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) + x_209 = None + x_211 = x_202 + x_210 + x_202 = x_210 = None + x_212 = x_211.permute(0, 2, 3, 1) + x_213 = torch.nn.functional.layer_norm( + x_212, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_212 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) + x_214 = x_213.permute(0, 3, 1, 2) + x_213 = None + x_215 = torch.conv2d( + x_214, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_214 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_24 = torch.nn.functional.relu(x_215, inplace=False) + x_215 = None + pow_25 = relu_24**2 + relu_24 = None + mul_24 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ + * pow_25 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_25 + ) = None + x_216 = ( + mul_24 + + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_24 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_217 = torch.conv2d( + x_216, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_216 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_217 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_219 = x_211 + x_218 + x_211 = x_218 = None + x_220 = x_219.permute(0, 2, 3, 1) + x_221 = torch.nn.functional.layer_norm( + x_220, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_220 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) + x_222 = x_221.permute(0, 3, 1, 2) + x_221 = None + x_223 = torch.conv2d( + x_222, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_222 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_25 = torch.nn.functional.relu(x_223, inplace=False) + x_223 = None + pow_26 = relu_25**2 + relu_25 = None + mul_25 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + * pow_26 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( + pow_26 + ) = None + x_224 = ( + mul_25 + + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + ) + mul_25 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = torch.conv2d( + x_225, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_225 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) + x_226 = None + x_228 = x_219 + x_227 + x_219 = x_227 = None + x_229 = x_228.permute(0, 2, 3, 1) + x_230 = torch.nn.functional.layer_norm( + x_229, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_229 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) + x_231 = x_230.permute(0, 3, 1, 2) + x_230 = None + x_232 = torch.conv2d( + x_231, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_231 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_26 = torch.nn.functional.relu(x_232, inplace=False) + x_232 = None + pow_27 = relu_26**2 + relu_26 = None + mul_26 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ + * pow_27 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_27 + ) = None + x_233 = ( + mul_26 + + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_26 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_234 = torch.conv2d( + x_233, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_233 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_235 = torch.conv2d( + x_234, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_234 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_236 = x_228 + x_235 + x_228 = x_235 = None + x_237 = x_236.permute(0, 2, 3, 1) + x_238 = torch.nn.functional.layer_norm( + x_237, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_237 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) + x_239 = x_238.permute(0, 3, 1, 2) + x_238 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_239 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_27 = torch.nn.functional.relu(x_240, inplace=False) + x_240 = None + pow_28 = relu_27**2 + relu_27 = None + mul_27 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + * pow_28 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( + pow_28 + ) = None + x_241 = ( + mul_27 + + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + ) + mul_27 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) + x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) + x_241 = None + x_243 = torch.conv2d( + x_242, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_242 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_244 = torch.nn.functional.dropout(x_243, 0.0, False, False) + x_243 = None + x_245 = x_236 + x_244 + x_236 = x_244 = None + x_246 = x_245.permute(0, 2, 3, 1) + x_247 = torch.nn.functional.layer_norm( + x_246, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_246 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) + x_248 = x_247.permute(0, 3, 1, 2) + x_247 = None + x_249 = torch.conv2d( + x_248, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_248 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_28 = torch.nn.functional.relu(x_249, inplace=False) + x_249 = None + pow_29 = relu_28**2 + relu_28 = None + mul_28 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ + * pow_29 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_29 + ) = None + x_250 = ( + mul_28 + + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_28 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_250 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_252 = torch.conv2d( + x_251, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_251 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_253 = x_245 + x_252 + x_245 = x_252 = None + x_254 = x_253.permute(0, 2, 3, 1) + x_255 = torch.nn.functional.layer_norm( + x_254, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_254 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) + x_256 = x_255.permute(0, 3, 1, 2) + x_255 = None + x_257 = torch.conv2d( + x_256, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_256 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_29 = torch.nn.functional.relu(x_257, inplace=False) + x_257 = None + pow_30 = relu_29**2 + relu_29 = None + mul_29 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + * pow_30 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( + pow_30 + ) = None + x_258 = ( + mul_29 + + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + ) + mul_29 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = torch.conv2d( + x_259, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_259 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_261 = torch.nn.functional.dropout(x_260, 0.0, False, False) + x_260 = None + x_262 = x_253 + x_261 + x_253 = x_261 = None + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (256,), + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) + x_265 = x_264.permute(0, 3, 1, 2) + x_264 = None + x_266 = torch.conv2d( + x_265, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_265 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_2 = sym_sum // 16 + sym_sum_3 = torch.sym_sum([1, floordiv_2]) + floordiv_2 = None + reshape = x_266.reshape(1, 512, -1) + x_266 = None + x_267 = reshape.transpose(1, 2) + reshape = None + view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_30 = x_267 * view + view = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_267, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_52 = sym_sum_3**2 + linear = torch._C._nn.linear( + layer_norm_33, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_33 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_1 = linear.reshape(1, pow_52, 3, 16, 32) + linear = None + qkv = reshape_1.permute(2, 0, 3, 1, 4) + reshape_1 = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_268 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) + q = k = v = None + transpose_1 = x_268.transpose(1, 2) + x_268 = None + x_269 = transpose_1.reshape(1, pow_52, 512) + transpose_1 = None + x_270 = torch._C._nn.linear( + x_269, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_269 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_271 = torch.nn.functional.dropout(x_270, 0.0, False, False) + x_270 = None + x_272 = mul_30 + x_271 + mul_30 = x_271 = None + view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_31 = x_272 * view_1 + view_1 = None + layer_norm_34 = torch.nn.functional.layer_norm( + x_272, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_272 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_273 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_34 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_30 = torch.nn.functional.relu(x_273, inplace=False) + x_273 = None + pow_31 = relu_30**2 + relu_30 = None + mul_32 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_31 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_31 + ) = None + x_274 = ( + mul_32 + + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_275 = torch.nn.functional.dropout(x_274, 0.0, False, False) + x_274 = None + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_277 = torch.nn.functional.dropout(x_276, 0.0, False, False) + x_276 = None + x_278 = mul_31 + x_277 + mul_31 = x_277 = None + view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_33 = x_278 * view_2 + view_2 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_278, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_3 = linear_4.reshape(1, pow_52, 3, 16, 32) + linear_4 = None + qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) + reshape_3 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_279 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) + q_1 = k_1 = v_1 = None + transpose_2 = x_279.transpose(1, 2) + x_279 = None + x_280 = transpose_2.reshape(1, pow_52, 512) + transpose_2 = None + x_281 = torch._C._nn.linear( + x_280, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_280 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) + x_281 = None + x_283 = mul_33 + x_282 + mul_33 = x_282 = None + view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_34 = x_283 * view_3 + view_3 = None + layer_norm_36 = torch.nn.functional.layer_norm( + x_283, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_283 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_284 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_36 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_31 = torch.nn.functional.relu(x_284, inplace=False) + x_284 = None + pow_32 = relu_31**2 + relu_31 = None + mul_35 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_32 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_32 + ) = None + x_285 = ( + mul_35 + + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_286 = torch.nn.functional.dropout(x_285, 0.0, False, False) + x_285 = None + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) + x_287 = None + x_289 = mul_34 + x_288 + mul_34 = x_288 = None + view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_36 = x_289 * view_4 + view_4 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_289, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_37 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_5 = linear_8.reshape(1, pow_52, 3, 16, 32) + linear_8 = None + qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) + reshape_5 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_290 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) + q_2 = k_2 = v_2 = None + transpose_3 = x_290.transpose(1, 2) + x_290 = None + x_291 = transpose_3.reshape(1, pow_52, 512) + transpose_3 = None + x_292 = torch._C._nn.linear( + x_291, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_291 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_293 = torch.nn.functional.dropout(x_292, 0.0, False, False) + x_292 = None + x_294 = mul_36 + x_293 + mul_36 = x_293 = None + view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_37 = x_294 * view_5 + view_5 = None + layer_norm_38 = torch.nn.functional.layer_norm( + x_294, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_294 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_295 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_32 = torch.nn.functional.relu(x_295, inplace=False) + x_295 = None + pow_33 = relu_32**2 + relu_32 = None + mul_38 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_33 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_33 + ) = None + x_296 = ( + mul_38 + + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_297 = torch.nn.functional.dropout(x_296, 0.0, False, False) + x_296 = None + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_299 = torch.nn.functional.dropout(x_298, 0.0, False, False) + x_298 = None + x_300 = mul_37 + x_299 + mul_37 = x_299 = None + view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_39 = x_300 * view_6 + view_6 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_300, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_39 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_7 = linear_12.reshape(1, pow_52, 3, 16, 32) + linear_12 = None + qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) + reshape_7 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_301 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) + q_3 = k_3 = v_3 = None + transpose_4 = x_301.transpose(1, 2) + x_301 = None + x_302 = transpose_4.reshape(1, pow_52, 512) + transpose_4 = None + x_303 = torch._C._nn.linear( + x_302, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_302 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_304 = torch.nn.functional.dropout(x_303, 0.0, False, False) + x_303 = None + x_305 = mul_39 + x_304 + mul_39 = x_304 = None + view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_40 = x_305 * view_7 + view_7 = None + layer_norm_40 = torch.nn.functional.layer_norm( + x_305, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_305 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) + x_306 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_40 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_33 = torch.nn.functional.relu(x_306, inplace=False) + x_306 = None + pow_34 = relu_33**2 + relu_33 = None + mul_41 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + * pow_34 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( + pow_34 + ) = None + x_307 = ( + mul_41 + + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + ) + mul_41 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) + x_308 = torch.nn.functional.dropout(x_307, 0.0, False, False) + x_307 = None + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) + x_309 = None + x_311 = mul_40 + x_310 + mul_40 = x_310 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_42 = x_311 * view_8 + view_8 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_311, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_41 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_9 = linear_16.reshape(1, pow_52, 3, 16, 32) + linear_16 = None + qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) + reshape_9 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_312 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) + q_4 = k_4 = v_4 = None + transpose_5 = x_312.transpose(1, 2) + x_312 = None + x_313 = transpose_5.reshape(1, pow_52, 512) + transpose_5 = None + x_314 = torch._C._nn.linear( + x_313, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_313 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) + x_314 = None + x_316 = mul_42 + x_315 + mul_42 = x_315 = None + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_43 = x_316 * view_9 + view_9 = None + layer_norm_42 = torch.nn.functional.layer_norm( + x_316, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_316 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) + x_317 = torch._C._nn.linear( + layer_norm_42, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_42 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_34 = torch.nn.functional.relu(x_317, inplace=False) + x_317 = None + pow_35 = relu_34**2 + relu_34 = None + mul_44 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + * pow_35 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( + pow_35 + ) = None + x_318 = ( + mul_44 + + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + ) + mul_44 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) + x_319 = torch.nn.functional.dropout(x_318, 0.0, False, False) + x_318 = None + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_321 = torch.nn.functional.dropout(x_320, 0.0, False, False) + x_320 = None + x_322 = mul_43 + x_321 + mul_43 = x_321 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_45 = x_322 * view_10 + view_10 = None + layer_norm_43 = torch.nn.functional.layer_norm( + x_322, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) + linear_20 = torch._C._nn.linear( + layer_norm_43, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_43 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_11 = linear_20.reshape(1, pow_52, 3, 16, 32) + linear_20 = None + qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) + reshape_11 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_323 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) + q_5 = k_5 = v_5 = None + transpose_6 = x_323.transpose(1, 2) + x_323 = None + x_324 = transpose_6.reshape(1, pow_52, 512) + transpose_6 = None + x_325 = torch._C._nn.linear( + x_324, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_324 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_326 = torch.nn.functional.dropout(x_325, 0.0, False, False) + x_325 = None + x_327 = mul_45 + x_326 + mul_45 = x_326 = None + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_46 = x_327 * view_11 + view_11 = None + layer_norm_44 = torch.nn.functional.layer_norm( + x_327, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_327 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) + x_328 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_44 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_35 = torch.nn.functional.relu(x_328, inplace=False) + x_328 = None + pow_36 = relu_35**2 + relu_35 = None + mul_47 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + * pow_36 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( + pow_36 + ) = None + x_329 = ( + mul_47 + + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + ) + mul_47 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) + x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) + x_329 = None + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_332 = torch.nn.functional.dropout(x_331, 0.0, False, False) + x_331 = None + x_333 = mul_46 + x_332 + mul_46 = x_332 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_48 = x_333 * view_12 + view_12 = None + layer_norm_45 = torch.nn.functional.layer_norm( + x_333, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) + linear_24 = torch._C._nn.linear( + layer_norm_45, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_45 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_13 = linear_24.reshape(1, pow_52, 3, 16, 32) + linear_24 = None + qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) + reshape_13 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_334 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) + q_6 = k_6 = v_6 = None + transpose_7 = x_334.transpose(1, 2) + x_334 = None + x_335 = transpose_7.reshape(1, pow_52, 512) + transpose_7 = None + x_336 = torch._C._nn.linear( + x_335, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_335 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_337 = torch.nn.functional.dropout(x_336, 0.0, False, False) + x_336 = None + x_338 = mul_48 + x_337 + mul_48 = x_337 = None + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_49 = x_338 * view_13 + view_13 = None + layer_norm_46 = torch.nn.functional.layer_norm( + x_338, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_338 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) + x_339 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_46 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_36 = torch.nn.functional.relu(x_339, inplace=False) + x_339 = None + pow_37 = relu_36**2 + relu_36 = None + mul_50 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + * pow_37 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( + pow_37 + ) = None + x_340 = ( + mul_50 + + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + ) + mul_50 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) + x_341 = torch.nn.functional.dropout(x_340, 0.0, False, False) + x_340 = None + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_343 = torch.nn.functional.dropout(x_342, 0.0, False, False) + x_342 = None + x_344 = mul_49 + x_343 + mul_49 = x_343 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_51 = x_344 * view_14 + view_14 = None + layer_norm_47 = torch.nn.functional.layer_norm( + x_344, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) + linear_28 = torch._C._nn.linear( + layer_norm_47, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_47 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_15 = linear_28.reshape(1, pow_52, 3, 16, 32) + linear_28 = None + qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) + reshape_15 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_345 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) + q_7 = k_7 = v_7 = None + transpose_8 = x_345.transpose(1, 2) + x_345 = None + x_346 = transpose_8.reshape(1, pow_52, 512) + transpose_8 = None + x_347 = torch._C._nn.linear( + x_346, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_346 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) + x_347 = None + x_349 = mul_51 + x_348 + mul_51 = x_348 = None + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_52 = x_349 * view_15 + view_15 = None + layer_norm_48 = torch.nn.functional.layer_norm( + x_349, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_349 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) + x_350 = torch._C._nn.linear( + layer_norm_48, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_48 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_37 = torch.nn.functional.relu(x_350, inplace=False) + x_350 = None + pow_38 = relu_37**2 + relu_37 = None + mul_53 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + * pow_38 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( + pow_38 + ) = None + x_351 = ( + mul_53 + + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + ) + mul_53 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) + x_352 = torch.nn.functional.dropout(x_351, 0.0, False, False) + x_351 = None + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_354 = torch.nn.functional.dropout(x_353, 0.0, False, False) + x_353 = None + x_355 = mul_52 + x_354 + mul_52 = x_354 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_54 = x_355 * view_16 + view_16 = None + layer_norm_49 = torch.nn.functional.layer_norm( + x_355, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) + linear_32 = torch._C._nn.linear( + layer_norm_49, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_49 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_17 = linear_32.reshape(1, pow_52, 3, 16, 32) + linear_32 = None + qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) + reshape_17 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_356 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) + q_8 = k_8 = v_8 = None + transpose_9 = x_356.transpose(1, 2) + x_356 = None + x_357 = transpose_9.reshape(1, pow_52, 512) + transpose_9 = None + x_358 = torch._C._nn.linear( + x_357, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_357 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_359 = torch.nn.functional.dropout(x_358, 0.0, False, False) + x_358 = None + x_360 = mul_54 + x_359 + mul_54 = x_359 = None + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_55 = x_360 * view_17 + view_17 = None + layer_norm_50 = torch.nn.functional.layer_norm( + x_360, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_360 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) + x_361 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_50 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_38 = torch.nn.functional.relu(x_361, inplace=False) + x_361 = None + pow_39 = relu_38**2 + relu_38 = None + mul_56 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + * pow_39 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( + pow_39 + ) = None + x_362 = ( + mul_56 + + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + ) + mul_56 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) + x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) + x_362 = None + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_365 = torch.nn.functional.dropout(x_364, 0.0, False, False) + x_364 = None + x_366 = mul_55 + x_365 + mul_55 = x_365 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_57 = x_366 * view_18 + view_18 = None + layer_norm_51 = torch.nn.functional.layer_norm( + x_366, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_51, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_51 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_19 = linear_36.reshape(1, pow_52, 3, 16, 32) + linear_36 = None + qkv_9 = reshape_19.permute(2, 0, 3, 1, 4) + reshape_19 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_367 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) + q_9 = k_9 = v_9 = None + transpose_10 = x_367.transpose(1, 2) + x_367 = None + x_368 = transpose_10.reshape(1, pow_52, 512) + transpose_10 = None + x_369 = torch._C._nn.linear( + x_368, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_368 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_370 = torch.nn.functional.dropout(x_369, 0.0, False, False) + x_369 = None + x_371 = mul_57 + x_370 + mul_57 = x_370 = None + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_58 = x_371 * view_19 + view_19 = None + layer_norm_52 = torch.nn.functional.layer_norm( + x_371, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_371 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) + x_372 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_52 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_39 = torch.nn.functional.relu(x_372, inplace=False) + x_372 = None + pow_40 = relu_39**2 + relu_39 = None + mul_59 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + * pow_40 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( + pow_40 + ) = None + x_373 = ( + mul_59 + + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + ) + mul_59 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) + x_374 = torch.nn.functional.dropout(x_373, 0.0, False, False) + x_373 = None + x_375 = torch._C._nn.linear( + x_374, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_374 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_376 = torch.nn.functional.dropout(x_375, 0.0, False, False) + x_375 = None + x_377 = mul_58 + x_376 + mul_58 = x_376 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_60 = x_377 * view_20 + view_20 = None + layer_norm_53 = torch.nn.functional.layer_norm( + x_377, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_377 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_53, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_53 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_21 = linear_40.reshape(1, pow_52, 3, 16, 32) + linear_40 = None + qkv_10 = reshape_21.permute(2, 0, 3, 1, 4) + reshape_21 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_378 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_378.transpose(1, 2) + x_378 = None + x_379 = transpose_11.reshape(1, pow_52, 512) + transpose_11 = None + x_380 = torch._C._nn.linear( + x_379, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_379 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_381 = torch.nn.functional.dropout(x_380, 0.0, False, False) + x_380 = None + x_382 = mul_60 + x_381 + mul_60 = x_381 = None + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_61 = x_382 * view_21 + view_21 = None + layer_norm_54 = torch.nn.functional.layer_norm( + x_382, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_382 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) + x_383 = torch._C._nn.linear( + layer_norm_54, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_54 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_40 = torch.nn.functional.relu(x_383, inplace=False) + x_383 = None + pow_41 = relu_40**2 + relu_40 = None + mul_62 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + * pow_41 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( + pow_41 + ) = None + x_384 = ( + mul_62 + + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + ) + mul_62 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) + x_385 = torch.nn.functional.dropout(x_384, 0.0, False, False) + x_384 = None + x_386 = torch._C._nn.linear( + x_385, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_385 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_387 = torch.nn.functional.dropout(x_386, 0.0, False, False) + x_386 = None + x_388 = mul_61 + x_387 + mul_61 = x_387 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_63 = x_388 * view_22 + view_22 = None + layer_norm_55 = torch.nn.functional.layer_norm( + x_388, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_388 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_55, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_55 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_23 = linear_44.reshape(1, pow_52, 3, 16, 32) + linear_44 = None + qkv_11 = reshape_23.permute(2, 0, 3, 1, 4) + reshape_23 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_389 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_389.transpose(1, 2) + x_389 = None + x_390 = transpose_12.reshape(1, pow_52, 512) + transpose_12 = None + x_391 = torch._C._nn.linear( + x_390, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_390 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_392 = torch.nn.functional.dropout(x_391, 0.0, False, False) + x_391 = None + x_393 = mul_63 + x_392 + mul_63 = x_392 = None + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_64 = x_393 * view_23 + view_23 = None + layer_norm_56 = torch.nn.functional.layer_norm( + x_393, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_393 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) + x_394 = torch._C._nn.linear( + layer_norm_56, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_56 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_41 = torch.nn.functional.relu(x_394, inplace=False) + x_394 = None + pow_42 = relu_41**2 + relu_41 = None + mul_65 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + * pow_42 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( + pow_42 + ) = None + x_395 = ( + mul_65 + + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + ) + mul_65 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) + x_396 = torch.nn.functional.dropout(x_395, 0.0, False, False) + x_395 = None + x_397 = torch._C._nn.linear( + x_396, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_396 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_398 = torch.nn.functional.dropout(x_397, 0.0, False, False) + x_397 = None + x_399 = mul_64 + x_398 + mul_64 = x_398 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_66 = x_399 * view_24 + view_24 = None + layer_norm_57 = torch.nn.functional.layer_norm( + x_399, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_399 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_57, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_57 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_25 = linear_48.reshape(1, pow_52, 3, 16, 32) + linear_48 = None + qkv_12 = reshape_25.permute(2, 0, 3, 1, 4) + reshape_25 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_12 = unbind_12[0] + k_12 = unbind_12[1] + v_12 = unbind_12[2] + unbind_12 = None + x_400 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_13 = x_400.transpose(1, 2) + x_400 = None + x_401 = transpose_13.reshape(1, pow_52, 512) + transpose_13 = None + x_402 = torch._C._nn.linear( + x_401, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_401 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_403 = torch.nn.functional.dropout(x_402, 0.0, False, False) + x_402 = None + x_404 = mul_66 + x_403 + mul_66 = x_403 = None + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_67 = x_404 * view_25 + view_25 = None + layer_norm_58 = torch.nn.functional.layer_norm( + x_404, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_404 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = (None) + x_405 = torch._C._nn.linear( + layer_norm_58, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_58 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_42 = torch.nn.functional.relu(x_405, inplace=False) + x_405 = None + pow_43 = relu_42**2 + relu_42 = None + mul_68 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ + * pow_43 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = ( + pow_43 + ) = None + x_406 = ( + mul_68 + + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ + ) + mul_68 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = (None) + x_407 = torch.nn.functional.dropout(x_406, 0.0, False, False) + x_406 = None + x_408 = torch._C._nn.linear( + x_407, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_407 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_409 = torch.nn.functional.dropout(x_408, 0.0, False, False) + x_408 = None + x_410 = mul_67 + x_409 + mul_67 = x_409 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_69 = x_410 * view_26 + view_26 = None + layer_norm_59 = torch.nn.functional.layer_norm( + x_410, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_410 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = (None) + linear_52 = torch._C._nn.linear( + layer_norm_59, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_59 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_27 = linear_52.reshape(1, pow_52, 3, 16, 32) + linear_52 = None + qkv_13 = reshape_27.permute(2, 0, 3, 1, 4) + reshape_27 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_13 = unbind_13[0] + k_13 = unbind_13[1] + v_13 = unbind_13[2] + unbind_13 = None + x_411 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_14 = x_411.transpose(1, 2) + x_411 = None + x_412 = transpose_14.reshape(1, pow_52, 512) + transpose_14 = None + x_413 = torch._C._nn.linear( + x_412, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_412 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_414 = torch.nn.functional.dropout(x_413, 0.0, False, False) + x_413 = None + x_415 = mul_69 + x_414 + mul_69 = x_414 = None + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_70 = x_415 * view_27 + view_27 = None + layer_norm_60 = torch.nn.functional.layer_norm( + x_415, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_415 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = (None) + x_416 = torch._C._nn.linear( + layer_norm_60, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_60 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_43 = torch.nn.functional.relu(x_416, inplace=False) + x_416 = None + pow_44 = relu_43**2 + relu_43 = None + mul_71 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ + * pow_44 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = ( + pow_44 + ) = None + x_417 = ( + mul_71 + + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ + ) + mul_71 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = (None) + x_418 = torch.nn.functional.dropout(x_417, 0.0, False, False) + x_417 = None + x_419 = torch._C._nn.linear( + x_418, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_418 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_420 = torch.nn.functional.dropout(x_419, 0.0, False, False) + x_419 = None + x_421 = mul_70 + x_420 + mul_70 = x_420 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_72 = x_421 * view_28 + view_28 = None + layer_norm_61 = torch.nn.functional.layer_norm( + x_421, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_421 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_61, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_61 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_29 = linear_56.reshape(1, pow_52, 3, 16, 32) + linear_56 = None + qkv_14 = reshape_29.permute(2, 0, 3, 1, 4) + reshape_29 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_14 = unbind_14[0] + k_14 = unbind_14[1] + v_14 = unbind_14[2] + unbind_14 = None + x_422 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_15 = x_422.transpose(1, 2) + x_422 = None + x_423 = transpose_15.reshape(1, pow_52, 512) + transpose_15 = None + x_424 = torch._C._nn.linear( + x_423, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_423 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_425 = torch.nn.functional.dropout(x_424, 0.0, False, False) + x_424 = None + x_426 = mul_72 + x_425 + mul_72 = x_425 = None + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_73 = x_426 * view_29 + view_29 = None + layer_norm_62 = torch.nn.functional.layer_norm( + x_426, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_426 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = (None) + x_427 = torch._C._nn.linear( + layer_norm_62, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_62 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_44 = torch.nn.functional.relu(x_427, inplace=False) + x_427 = None + pow_45 = relu_44**2 + relu_44 = None + mul_74 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ + * pow_45 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = ( + pow_45 + ) = None + x_428 = ( + mul_74 + + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ + ) + mul_74 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = (None) + x_429 = torch.nn.functional.dropout(x_428, 0.0, False, False) + x_428 = None + x_430 = torch._C._nn.linear( + x_429, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_429 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_431 = torch.nn.functional.dropout(x_430, 0.0, False, False) + x_430 = None + x_432 = mul_73 + x_431 + mul_73 = x_431 = None + view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_75 = x_432 * view_30 + view_30 = None + layer_norm_63 = torch.nn.functional.layer_norm( + x_432, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_432 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = (None) + linear_60 = torch._C._nn.linear( + layer_norm_63, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_63 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_31 = linear_60.reshape(1, pow_52, 3, 16, 32) + linear_60 = None + qkv_15 = reshape_31.permute(2, 0, 3, 1, 4) + reshape_31 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_15 = unbind_15[0] + k_15 = unbind_15[1] + v_15 = unbind_15[2] + unbind_15 = None + x_433 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_16 = x_433.transpose(1, 2) + x_433 = None + x_434 = transpose_16.reshape(1, pow_52, 512) + transpose_16 = None + x_435 = torch._C._nn.linear( + x_434, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_434 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_436 = torch.nn.functional.dropout(x_435, 0.0, False, False) + x_435 = None + x_437 = mul_75 + x_436 + mul_75 = x_436 = None + view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_76 = x_437 * view_31 + view_31 = None + layer_norm_64 = torch.nn.functional.layer_norm( + x_437, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_437 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = (None) + x_438 = torch._C._nn.linear( + layer_norm_64, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_64 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_45 = torch.nn.functional.relu(x_438, inplace=False) + x_438 = None + pow_46 = relu_45**2 + relu_45 = None + mul_77 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ + * pow_46 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = ( + pow_46 + ) = None + x_439 = ( + mul_77 + + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ + ) + mul_77 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = (None) + x_440 = torch.nn.functional.dropout(x_439, 0.0, False, False) + x_439 = None + x_441 = torch._C._nn.linear( + x_440, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_440 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_442 = torch.nn.functional.dropout(x_441, 0.0, False, False) + x_441 = None + x_443 = mul_76 + x_442 + mul_76 = x_442 = None + view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_78 = x_443 * view_32 + view_32 = None + layer_norm_65 = torch.nn.functional.layer_norm( + x_443, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_443 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_65, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_65 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_33 = linear_64.reshape(1, pow_52, 3, 16, 32) + linear_64 = None + qkv_16 = reshape_33.permute(2, 0, 3, 1, 4) + reshape_33 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_16 = unbind_16[0] + k_16 = unbind_16[1] + v_16 = unbind_16[2] + unbind_16 = None + x_444 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_17 = x_444.transpose(1, 2) + x_444 = None + x_445 = transpose_17.reshape(1, pow_52, 512) + transpose_17 = None + x_446 = torch._C._nn.linear( + x_445, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_445 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_447 = torch.nn.functional.dropout(x_446, 0.0, False, False) + x_446 = None + x_448 = mul_78 + x_447 + mul_78 = x_447 = None + view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_79 = x_448 * view_33 + view_33 = None + layer_norm_66 = torch.nn.functional.layer_norm( + x_448, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_448 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = (None) + x_449 = torch._C._nn.linear( + layer_norm_66, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_66 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_46 = torch.nn.functional.relu(x_449, inplace=False) + x_449 = None + pow_47 = relu_46**2 + relu_46 = None + mul_80 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ + * pow_47 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = ( + pow_47 + ) = None + x_450 = ( + mul_80 + + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ + ) + mul_80 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = (None) + x_451 = torch.nn.functional.dropout(x_450, 0.0, False, False) + x_450 = None + x_452 = torch._C._nn.linear( + x_451, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_451 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_453 = torch.nn.functional.dropout(x_452, 0.0, False, False) + x_452 = None + x_454 = mul_79 + x_453 + mul_79 = x_453 = None + view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_81 = x_454 * view_34 + view_34 = None + layer_norm_67 = torch.nn.functional.layer_norm( + x_454, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_454 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = (None) + linear_68 = torch._C._nn.linear( + layer_norm_67, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_67 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_35 = linear_68.reshape(1, pow_52, 3, 16, 32) + linear_68 = None + qkv_17 = reshape_35.permute(2, 0, 3, 1, 4) + reshape_35 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_17 = unbind_17[0] + k_17 = unbind_17[1] + v_17 = unbind_17[2] + unbind_17 = None + x_455 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_18 = x_455.transpose(1, 2) + x_455 = None + x_456 = transpose_18.reshape(1, pow_52, 512) + transpose_18 = pow_52 = None + x_457 = torch._C._nn.linear( + x_456, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_456 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_458 = torch.nn.functional.dropout(x_457, 0.0, False, False) + x_457 = None + x_459 = mul_81 + x_458 + mul_81 = x_458 = None + view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_82 = x_459 * view_35 + view_35 = None + layer_norm_68 = torch.nn.functional.layer_norm( + x_459, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_459 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = (None) + x_460 = torch._C._nn.linear( + layer_norm_68, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_68 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_47 = torch.nn.functional.relu(x_460, inplace=False) + x_460 = None + pow_48 = relu_47**2 + relu_47 = None + mul_83 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ + * pow_48 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = ( + pow_48 + ) = None + x_461 = ( + mul_83 + + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ + ) + mul_83 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = (None) + x_462 = torch.nn.functional.dropout(x_461, 0.0, False, False) + x_461 = None + x_463 = torch._C._nn.linear( + x_462, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_462 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_464 = torch.nn.functional.dropout(x_463, 0.0, False, False) + x_463 = None + x_465 = mul_82 + x_464 + mul_82 = x_464 = None + transpose_19 = x_465.transpose(1, 2) + x_465 = None + x_466 = transpose_19.reshape(1, 512, sym_sum_3, sym_sum_3) + transpose_19 = sym_sum_3 = None + x_467 = x_466.permute(0, 2, 3, 1) + x_466 = None + x_468 = torch.nn.functional.layer_norm( + x_467, + (512,), + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_467 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) + x_469 = x_468.permute(0, 3, 1, 2) + x_468 = None + x_470 = torch.conv2d( + x_469, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_469 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_3 = sym_sum // 32 + sym_sum = None + sym_sum_4 = torch.sym_sum([1, floordiv_3]) + floordiv_3 = None + reshape_38 = x_470.reshape(1, 768, -1) + x_470 = None + x_471 = reshape_38.transpose(1, 2) + reshape_38 = None + view_36 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_84 = x_471 * view_36 + view_36 = None + layer_norm_70 = torch.nn.functional.layer_norm( + x_471, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_471 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_53 = sym_sum_4**2 + linear_72 = torch._C._nn.linear( + layer_norm_70, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_70 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_39 = linear_72.reshape(1, pow_53, 3, 24, 32) + linear_72 = None + qkv_18 = reshape_39.permute(2, 0, 3, 1, 4) + reshape_39 = None + unbind_18 = qkv_18.unbind(0) + qkv_18 = None + q_18 = unbind_18[0] + k_18 = unbind_18[1] + v_18 = unbind_18[2] + unbind_18 = None + x_472 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_21 = x_472.transpose(1, 2) + x_472 = None + x_473 = transpose_21.reshape(1, pow_53, 768) + transpose_21 = None + x_474 = torch._C._nn.linear( + x_473, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_473 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_475 = torch.nn.functional.dropout(x_474, 0.0, False, False) + x_474 = None + x_476 = mul_84 + x_475 + mul_84 = x_475 = None + view_37 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_85 = x_476 * view_37 + view_37 = None + layer_norm_71 = torch.nn.functional.layer_norm( + x_476, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_476 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_477 = torch._C._nn.linear( + layer_norm_71, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_71 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_48 = torch.nn.functional.relu(x_477, inplace=False) + x_477 = None + pow_49 = relu_48**2 + relu_48 = None + mul_86 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_49 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_49 + ) = None + x_478 = ( + mul_86 + + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_86 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_479 = torch.nn.functional.dropout(x_478, 0.0, False, False) + x_478 = None + x_480 = torch._C._nn.linear( + x_479, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_479 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_481 = torch.nn.functional.dropout(x_480, 0.0, False, False) + x_480 = None + x_482 = mul_85 + x_481 + mul_85 = x_481 = None + view_38 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_87 = x_482 * view_38 + view_38 = None + layer_norm_72 = torch.nn.functional.layer_norm( + x_482, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_482 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_76 = torch._C._nn.linear( + layer_norm_72, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_72 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_41 = linear_76.reshape(1, pow_53, 3, 24, 32) + linear_76 = None + qkv_19 = reshape_41.permute(2, 0, 3, 1, 4) + reshape_41 = None + unbind_19 = qkv_19.unbind(0) + qkv_19 = None + q_19 = unbind_19[0] + k_19 = unbind_19[1] + v_19 = unbind_19[2] + unbind_19 = None + x_483 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_22 = x_483.transpose(1, 2) + x_483 = None + x_484 = transpose_22.reshape(1, pow_53, 768) + transpose_22 = None + x_485 = torch._C._nn.linear( + x_484, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_484 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_486 = torch.nn.functional.dropout(x_485, 0.0, False, False) + x_485 = None + x_487 = mul_87 + x_486 + mul_87 = x_486 = None + view_39 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_88 = x_487 * view_39 + view_39 = None + layer_norm_73 = torch.nn.functional.layer_norm( + x_487, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_487 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_488 = torch._C._nn.linear( + layer_norm_73, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_73 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_49 = torch.nn.functional.relu(x_488, inplace=False) + x_488 = None + pow_50 = relu_49**2 + relu_49 = None + mul_89 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_50 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_50 + ) = None + x_489 = ( + mul_89 + + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_89 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_490 = torch.nn.functional.dropout(x_489, 0.0, False, False) + x_489 = None + x_491 = torch._C._nn.linear( + x_490, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_490 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_492 = torch.nn.functional.dropout(x_491, 0.0, False, False) + x_491 = None + x_493 = mul_88 + x_492 + mul_88 = x_492 = None + view_40 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_90 = x_493 * view_40 + view_40 = None + layer_norm_74 = torch.nn.functional.layer_norm( + x_493, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_493 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_74, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_74 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_43 = linear_80.reshape(1, pow_53, 3, 24, 32) + linear_80 = None + qkv_20 = reshape_43.permute(2, 0, 3, 1, 4) + reshape_43 = None + unbind_20 = qkv_20.unbind(0) + qkv_20 = None + q_20 = unbind_20[0] + k_20 = unbind_20[1] + v_20 = unbind_20[2] + unbind_20 = None + x_494 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_23 = x_494.transpose(1, 2) + x_494 = None + x_495 = transpose_23.reshape(1, pow_53, 768) + transpose_23 = pow_53 = None + x_496 = torch._C._nn.linear( + x_495, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_495 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_497 = torch.nn.functional.dropout(x_496, 0.0, False, False) + x_496 = None + x_498 = mul_90 + x_497 + mul_90 = x_497 = None + view_41 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_91 = x_498 * view_41 + view_41 = None + layer_norm_75 = torch.nn.functional.layer_norm( + x_498, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_498 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_499 = torch._C._nn.linear( + layer_norm_75, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_75 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_50 = torch.nn.functional.relu(x_499, inplace=False) + x_499 = None + pow_51 = relu_50**2 + relu_50 = None + mul_92 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_51 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_51 + ) = None + x_500 = ( + mul_92 + + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_92 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_501 = torch.nn.functional.dropout(x_500, 0.0, False, False) + x_500 = None + x_502 = torch._C._nn.linear( + x_501, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_501 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_503 = torch.nn.functional.dropout(x_502, 0.0, False, False) + x_502 = None + x_504 = mul_91 + x_503 + mul_91 = x_503 = None + transpose_24 = x_504.transpose(1, 2) + x_504 = None + x_505 = transpose_24.reshape(1, 768, sym_sum_4, sym_sum_4) + transpose_24 = sym_sum_4 = None + x_506 = torch.nn.functional.adaptive_avg_pool2d(x_505, 1) + x_505 = None + x_507 = x_506.permute(0, 2, 3, 1) + x_506 = None + x_508 = torch.nn.functional.layer_norm( + x_507, + (768,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_507 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_509 = x_508.permute(0, 3, 1, 2) + x_508 = None + x_510 = x_509.flatten(1, -1) + x_509 = None + x_511 = torch.nn.functional.dropout(x_510, 0.0, False, False) + x_510 = None + x_512 = torch._C._nn.linear( + x_511, + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, + ) + x_511 = ( + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None + relu_51 = torch.nn.functional.relu(x_512, inplace=False) + x_512 = None + x_513 = torch.square(relu_51) + relu_51 = None + x_514 = torch.nn.functional.layer_norm( + x_513, + (3072,), + l_self_modules_head_modules_fc_modules_norm_parameters_weight_, + l_self_modules_head_modules_fc_modules_norm_parameters_bias_, + 1e-06, + ) + x_513 = ( + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None + x_515 = torch.nn.functional.dropout(x_514, 0.0, False, False) + x_514 = None + x_516 = torch._C._nn.linear( + x_515, + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, + ) + x_515 = ( + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None + return (x_516,) diff --git a/samples/timm/caformer_b36.sail_in1k/weight_meta.py b/samples/timm/caformer_b36.sail_in1k/weight_meta.py new file mode 100644 index 0000000000..1067172347 --- /dev/null +++ b/samples/timm/caformer_b36.sail_in1k/weight_meta.py @@ -0,0 +1,3968 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv_parameters_weight_" + shape = [128, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: + name = "L_self_modules_stem_modules_conv_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.228 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: + name = "L_self_modules_stem_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" + shape = [768, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" + shape = [1000, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/caformer_b36.sail_in1k_384/graph_net.json b/samples/timm/caformer_b36.sail_in1k_384/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/caformer_b36.sail_in1k_384/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/caformer_b36.sail_in1k_384/input_meta.py b/samples/timm/caformer_b36.sail_in1k_384/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_b36.sail_in1k_384/input_tensor_constraints.py b/samples/timm/caformer_b36.sail_in1k_384/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_b36.sail_in1k_384/model.py b/samples/timm/caformer_b36.sail_in1k_384/model.py new file mode 100644 index 0000000000..08d3de35cc --- /dev/null +++ b/samples/timm/caformer_b36.sail_in1k_384/model.py @@ -0,0 +1,4758 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv_parameters_bias_ = ( + L_self_modules_stem_modules_conv_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_norm_parameters_weight_ = ( + L_self_modules_stem_modules_norm_parameters_weight_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv_parameters_weight_, + l_self_modules_stem_modules_conv_parameters_bias_, + (4, 4), + (2, 2), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_conv_parameters_weight_ + ) = l_self_modules_stem_modules_conv_parameters_bias_ = None + x_1 = x.permute(0, 2, 3, 1) + x = None + x_2 = torch.nn.functional.layer_norm( + x_1, + (128,), + l_self_modules_stem_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None + x_3 = x_2.permute(0, 3, 1, 2) + x_2 = None + sym_sum = torch.sym_sum([-3, s1]) + s1 = None + floordiv = sym_sum // 4 + sym_sum_1 = torch.sym_sum([1, floordiv]) + floordiv = sym_sum_1 = None + x_4 = x_3.permute(0, 2, 3, 1) + x_5 = torch.nn.functional.layer_norm( + x_4, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu = torch.nn.functional.relu(x_7, inplace=False) + x_7 = None + pow_1 = relu**2 + relu = None + mul = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_1 + ) = None + x_8 = ( + mul + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_11 = x_3 + x_10 + x_3 = x_10 = None + x_12 = x_11.permute(0, 2, 3, 1) + x_13 = torch.nn.functional.layer_norm( + x_12, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_14 = x_13.permute(0, 3, 1, 2) + x_13 = None + x_15 = torch.conv2d( + x_14, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_1 = torch.nn.functional.relu(x_15, inplace=False) + x_15 = None + pow_2 = relu_1**2 + relu_1 = None + mul_1 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_2 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_2 + ) = None + x_16 = ( + mul_1 + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = x_11 + x_19 + x_11 = x_19 = None + x_21 = x_20.permute(0, 2, 3, 1) + x_22 = torch.nn.functional.layer_norm( + x_21, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = torch.conv2d( + x_23, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_2 = torch.nn.functional.relu(x_24, inplace=False) + x_24 = None + pow_3 = relu_2**2 + relu_2 = None + mul_2 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_3 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_3 + ) = None + x_25 = ( + mul_2 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_27 = torch.conv2d( + x_26, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_28 = x_20 + x_27 + x_20 = x_27 = None + x_29 = x_28.permute(0, 2, 3, 1) + x_30 = torch.nn.functional.layer_norm( + x_29, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_3 = torch.nn.functional.relu(x_32, inplace=False) + x_32 = None + pow_4 = relu_3**2 + relu_3 = None + mul_3 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_4 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_4 + ) = None + x_33 = ( + mul_3 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + x_37 = x_28 + x_36 + x_28 = x_36 = None + x_38 = x_37.permute(0, 2, 3, 1) + x_39 = torch.nn.functional.layer_norm( + x_38, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_40 = x_39.permute(0, 3, 1, 2) + x_39 = None + x_41 = torch.conv2d( + x_40, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_4 = torch.nn.functional.relu(x_41, inplace=False) + x_41 = None + pow_5 = relu_4**2 + relu_4 = None + mul_4 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_5 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_5 + ) = None + x_42 = ( + mul_4 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_44 = torch.conv2d( + x_43, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_45 = x_37 + x_44 + x_37 = x_44 = None + x_46 = x_45.permute(0, 2, 3, 1) + x_47 = torch.nn.functional.layer_norm( + x_46, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_5 = torch.nn.functional.relu(x_49, inplace=False) + x_49 = None + pow_6 = relu_5**2 + relu_5 = None + mul_5 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_6 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_6 + ) = None + x_50 = ( + mul_5 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = x_45 + x_53 + x_45 = x_53 = None + x_55 = x_54.permute(0, 2, 3, 1) + x_54 = None + x_56 = torch.nn.functional.layer_norm( + x_55, + (128,), + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) + x_57 = x_56.permute(0, 3, 1, 2) + x_56 = None + x_58 = torch.conv2d( + x_57, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_1 = sym_sum // 8 + sym_sum_2 = torch.sym_sum([1, floordiv_1]) + floordiv_1 = sym_sum_2 = None + x_59 = x_58.permute(0, 2, 3, 1) + x_60 = torch.nn.functional.layer_norm( + x_59, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_61 = x_60.permute(0, 3, 1, 2) + x_60 = None + x_62 = torch.conv2d( + x_61, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_6 = torch.nn.functional.relu(x_62, inplace=False) + x_62 = None + pow_7 = relu_6**2 + relu_6 = None + mul_6 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_7 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_7 + ) = None + x_63 = ( + mul_6 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_66 = x_58 + x_65 + x_58 = x_65 = None + x_67 = x_66.permute(0, 2, 3, 1) + x_68 = torch.nn.functional.layer_norm( + x_67, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + x_70 = torch.conv2d( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_7 = torch.nn.functional.relu(x_70, inplace=False) + x_70 = None + pow_8 = relu_7**2 + relu_7 = None + mul_7 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_8 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_8 + ) = None + x_71 = ( + mul_7 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + x_73 = torch.conv2d( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + x_75 = x_66 + x_74 + x_66 = x_74 = None + x_76 = x_75.permute(0, 2, 3, 1) + x_77 = torch.nn.functional.layer_norm( + x_76, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_8 = torch.nn.functional.relu(x_79, inplace=False) + x_79 = None + pow_9 = relu_8**2 + relu_8 = None + mul_8 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_9 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_9 + ) = None + x_80 = ( + mul_8 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_81 = torch.conv2d( + x_80, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_83 = x_75 + x_82 + x_75 = x_82 = None + x_84 = x_83.permute(0, 2, 3, 1) + x_85 = torch.nn.functional.layer_norm( + x_84, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_86 = x_85.permute(0, 3, 1, 2) + x_85 = None + x_87 = torch.conv2d( + x_86, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_9 = torch.nn.functional.relu(x_87, inplace=False) + x_87 = None + pow_10 = relu_9**2 + relu_9 = None + mul_9 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_10 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_10 + ) = None + x_88 = ( + mul_9 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = x_83 + x_91 + x_83 = x_91 = None + x_93 = x_92.permute(0, 2, 3, 1) + x_94 = torch.nn.functional.layer_norm( + x_93, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = torch.conv2d( + x_95, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_10 = torch.nn.functional.relu(x_96, inplace=False) + x_96 = None + pow_11 = relu_10**2 + relu_10 = None + mul_10 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_11 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_11 + ) = None + x_97 = ( + mul_10 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_98 = torch.conv2d( + x_97, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_99 = torch.conv2d( + x_98, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_100 = x_92 + x_99 + x_92 = x_99 = None + x_101 = x_100.permute(0, 2, 3, 1) + x_102 = torch.nn.functional.layer_norm( + x_101, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_103 = x_102.permute(0, 3, 1, 2) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_11 = torch.nn.functional.relu(x_104, inplace=False) + x_104 = None + pow_12 = relu_11**2 + relu_11 = None + mul_11 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_12 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_12 + ) = None + x_105 = ( + mul_11 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_100 + x_108 + x_100 = x_108 = None + x_110 = x_109.permute(0, 2, 3, 1) + x_111 = torch.nn.functional.layer_norm( + x_110, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_110 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) + x_112 = x_111.permute(0, 3, 1, 2) + x_111 = None + x_113 = torch.conv2d( + x_112, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_112 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_12 = torch.nn.functional.relu(x_113, inplace=False) + x_113 = None + pow_13 = relu_12**2 + relu_12 = None + mul_12 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ + * pow_13 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_13 + ) = None + x_114 = ( + mul_12 + + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_12 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_115 = torch.conv2d( + x_114, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_114 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_117 = x_109 + x_116 + x_109 = x_116 = None + x_118 = x_117.permute(0, 2, 3, 1) + x_119 = torch.nn.functional.layer_norm( + x_118, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_118 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) + x_120 = x_119.permute(0, 3, 1, 2) + x_119 = None + x_121 = torch.conv2d( + x_120, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_120 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_13 = torch.nn.functional.relu(x_121, inplace=False) + x_121 = None + pow_14 = relu_13**2 + relu_13 = None + mul_13 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + * pow_14 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( + pow_14 + ) = None + x_122 = ( + mul_13 + + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + ) + mul_13 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) + x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) + x_122 = None + x_124 = torch.conv2d( + x_123, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_123 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_126 = x_117 + x_125 + x_117 = x_125 = None + x_127 = x_126.permute(0, 2, 3, 1) + x_128 = torch.nn.functional.layer_norm( + x_127, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_127 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) + x_129 = x_128.permute(0, 3, 1, 2) + x_128 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_14 = torch.nn.functional.relu(x_130, inplace=False) + x_130 = None + pow_15 = relu_14**2 + relu_14 = None + mul_14 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ + * pow_15 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_15 + ) = None + x_131 = ( + mul_14 + + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_14 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_132 = torch.conv2d( + x_131, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_131 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_132 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_134 = x_126 + x_133 + x_126 = x_133 = None + x_135 = x_134.permute(0, 2, 3, 1) + x_136 = torch.nn.functional.layer_norm( + x_135, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_135 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) + x_137 = x_136.permute(0, 3, 1, 2) + x_136 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_15 = torch.nn.functional.relu(x_138, inplace=False) + x_138 = None + pow_16 = relu_15**2 + relu_15 = None + mul_15 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + * pow_16 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( + pow_16 + ) = None + x_139 = ( + mul_15 + + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + ) + mul_15 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) + x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) + x_139 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_140 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_134 + x_142 + x_134 = x_142 = None + x_144 = x_143.permute(0, 2, 3, 1) + x_145 = torch.nn.functional.layer_norm( + x_144, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_144 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) + x_146 = x_145.permute(0, 3, 1, 2) + x_145 = None + x_147 = torch.conv2d( + x_146, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_146 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_16 = torch.nn.functional.relu(x_147, inplace=False) + x_147 = None + pow_17 = relu_16**2 + relu_16 = None + mul_16 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ + * pow_17 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_17 + ) = None + x_148 = ( + mul_16 + + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_16 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_149 = torch.conv2d( + x_148, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_148 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_150 = torch.conv2d( + x_149, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_149 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_151 = x_143 + x_150 + x_143 = x_150 = None + x_152 = x_151.permute(0, 2, 3, 1) + x_153 = torch.nn.functional.layer_norm( + x_152, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_152 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) + x_154 = x_153.permute(0, 3, 1, 2) + x_153 = None + x_155 = torch.conv2d( + x_154, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_154 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_17 = torch.nn.functional.relu(x_155, inplace=False) + x_155 = None + pow_18 = relu_17**2 + relu_17 = None + mul_17 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + * pow_18 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( + pow_18 + ) = None + x_156 = ( + mul_17 + + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + ) + mul_17 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch.conv2d( + x_157, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_157 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_151 + x_159 + x_151 = x_159 = None + x_161 = x_160.permute(0, 2, 3, 1) + x_162 = torch.nn.functional.layer_norm( + x_161, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_161 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) + x_163 = x_162.permute(0, 3, 1, 2) + x_162 = None + x_164 = torch.conv2d( + x_163, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_163 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_18 = torch.nn.functional.relu(x_164, inplace=False) + x_164 = None + pow_19 = relu_18**2 + relu_18 = None + mul_18 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ + * pow_19 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_19 + ) = None + x_165 = ( + mul_18 + + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_18 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_166 = torch.conv2d( + x_165, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_165 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_167 = torch.conv2d( + x_166, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_166 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_168 = x_160 + x_167 + x_160 = x_167 = None + x_169 = x_168.permute(0, 2, 3, 1) + x_170 = torch.nn.functional.layer_norm( + x_169, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_169 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) + x_171 = x_170.permute(0, 3, 1, 2) + x_170 = None + x_172 = torch.conv2d( + x_171, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_171 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_19 = torch.nn.functional.relu(x_172, inplace=False) + x_172 = None + pow_20 = relu_19**2 + relu_19 = None + mul_19 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + * pow_20 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( + pow_20 + ) = None + x_173 = ( + mul_19 + + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + ) + mul_19 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) + x_174 = torch.nn.functional.dropout(x_173, 0.0, False, False) + x_173 = None + x_175 = torch.conv2d( + x_174, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_174 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = x_168 + x_176 + x_168 = x_176 = None + x_178 = x_177.permute(0, 2, 3, 1) + x_179 = torch.nn.functional.layer_norm( + x_178, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_178 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) + x_180 = x_179.permute(0, 3, 1, 2) + x_179 = None + x_181 = torch.conv2d( + x_180, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_180 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_20 = torch.nn.functional.relu(x_181, inplace=False) + x_181 = None + pow_21 = relu_20**2 + relu_20 = None + mul_20 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ + * pow_21 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_21 + ) = None + x_182 = ( + mul_20 + + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_20 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_183 = torch.conv2d( + x_182, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_182 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_184 = torch.conv2d( + x_183, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_183 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_185 = x_177 + x_184 + x_177 = x_184 = None + x_186 = x_185.permute(0, 2, 3, 1) + x_187 = torch.nn.functional.layer_norm( + x_186, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) + x_188 = x_187.permute(0, 3, 1, 2) + x_187 = None + x_189 = torch.conv2d( + x_188, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_188 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_21 = torch.nn.functional.relu(x_189, inplace=False) + x_189 = None + pow_22 = relu_21**2 + relu_21 = None + mul_21 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + * pow_22 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( + pow_22 + ) = None + x_190 = ( + mul_21 + + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + ) + mul_21 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) + x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) + x_190 = None + x_192 = torch.conv2d( + x_191, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_191 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_185 + x_193 + x_185 = x_193 = None + x_195 = x_194.permute(0, 2, 3, 1) + x_196 = torch.nn.functional.layer_norm( + x_195, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_195 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) + x_197 = x_196.permute(0, 3, 1, 2) + x_196 = None + x_198 = torch.conv2d( + x_197, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_197 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_22 = torch.nn.functional.relu(x_198, inplace=False) + x_198 = None + pow_23 = relu_22**2 + relu_22 = None + mul_22 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ + * pow_23 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_23 + ) = None + x_199 = ( + mul_22 + + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_22 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_200 = torch.conv2d( + x_199, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_199 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_201 = torch.conv2d( + x_200, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_200 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_202 = x_194 + x_201 + x_194 = x_201 = None + x_203 = x_202.permute(0, 2, 3, 1) + x_204 = torch.nn.functional.layer_norm( + x_203, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_203 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) + x_205 = x_204.permute(0, 3, 1, 2) + x_204 = None + x_206 = torch.conv2d( + x_205, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_205 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_23 = torch.nn.functional.relu(x_206, inplace=False) + x_206 = None + pow_24 = relu_23**2 + relu_23 = None + mul_23 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + * pow_24 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( + pow_24 + ) = None + x_207 = ( + mul_23 + + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + ) + mul_23 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) + x_208 = torch.nn.functional.dropout(x_207, 0.0, False, False) + x_207 = None + x_209 = torch.conv2d( + x_208, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_208 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) + x_209 = None + x_211 = x_202 + x_210 + x_202 = x_210 = None + x_212 = x_211.permute(0, 2, 3, 1) + x_213 = torch.nn.functional.layer_norm( + x_212, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_212 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) + x_214 = x_213.permute(0, 3, 1, 2) + x_213 = None + x_215 = torch.conv2d( + x_214, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_214 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_24 = torch.nn.functional.relu(x_215, inplace=False) + x_215 = None + pow_25 = relu_24**2 + relu_24 = None + mul_24 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ + * pow_25 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_25 + ) = None + x_216 = ( + mul_24 + + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_24 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_217 = torch.conv2d( + x_216, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_216 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_217 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_219 = x_211 + x_218 + x_211 = x_218 = None + x_220 = x_219.permute(0, 2, 3, 1) + x_221 = torch.nn.functional.layer_norm( + x_220, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_220 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) + x_222 = x_221.permute(0, 3, 1, 2) + x_221 = None + x_223 = torch.conv2d( + x_222, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_222 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_25 = torch.nn.functional.relu(x_223, inplace=False) + x_223 = None + pow_26 = relu_25**2 + relu_25 = None + mul_25 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + * pow_26 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( + pow_26 + ) = None + x_224 = ( + mul_25 + + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + ) + mul_25 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = torch.conv2d( + x_225, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_225 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) + x_226 = None + x_228 = x_219 + x_227 + x_219 = x_227 = None + x_229 = x_228.permute(0, 2, 3, 1) + x_230 = torch.nn.functional.layer_norm( + x_229, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_229 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) + x_231 = x_230.permute(0, 3, 1, 2) + x_230 = None + x_232 = torch.conv2d( + x_231, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_231 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_26 = torch.nn.functional.relu(x_232, inplace=False) + x_232 = None + pow_27 = relu_26**2 + relu_26 = None + mul_26 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ + * pow_27 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_27 + ) = None + x_233 = ( + mul_26 + + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_26 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_234 = torch.conv2d( + x_233, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_233 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_235 = torch.conv2d( + x_234, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_234 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_236 = x_228 + x_235 + x_228 = x_235 = None + x_237 = x_236.permute(0, 2, 3, 1) + x_238 = torch.nn.functional.layer_norm( + x_237, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_237 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) + x_239 = x_238.permute(0, 3, 1, 2) + x_238 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_239 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_27 = torch.nn.functional.relu(x_240, inplace=False) + x_240 = None + pow_28 = relu_27**2 + relu_27 = None + mul_27 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + * pow_28 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( + pow_28 + ) = None + x_241 = ( + mul_27 + + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + ) + mul_27 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) + x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) + x_241 = None + x_243 = torch.conv2d( + x_242, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_242 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_244 = torch.nn.functional.dropout(x_243, 0.0, False, False) + x_243 = None + x_245 = x_236 + x_244 + x_236 = x_244 = None + x_246 = x_245.permute(0, 2, 3, 1) + x_247 = torch.nn.functional.layer_norm( + x_246, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_246 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) + x_248 = x_247.permute(0, 3, 1, 2) + x_247 = None + x_249 = torch.conv2d( + x_248, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_248 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_28 = torch.nn.functional.relu(x_249, inplace=False) + x_249 = None + pow_29 = relu_28**2 + relu_28 = None + mul_28 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ + * pow_29 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_29 + ) = None + x_250 = ( + mul_28 + + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_28 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_250 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_252 = torch.conv2d( + x_251, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_251 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_253 = x_245 + x_252 + x_245 = x_252 = None + x_254 = x_253.permute(0, 2, 3, 1) + x_255 = torch.nn.functional.layer_norm( + x_254, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_254 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) + x_256 = x_255.permute(0, 3, 1, 2) + x_255 = None + x_257 = torch.conv2d( + x_256, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_256 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_29 = torch.nn.functional.relu(x_257, inplace=False) + x_257 = None + pow_30 = relu_29**2 + relu_29 = None + mul_29 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + * pow_30 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( + pow_30 + ) = None + x_258 = ( + mul_29 + + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + ) + mul_29 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = torch.conv2d( + x_259, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_259 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_261 = torch.nn.functional.dropout(x_260, 0.0, False, False) + x_260 = None + x_262 = x_253 + x_261 + x_253 = x_261 = None + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (256,), + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) + x_265 = x_264.permute(0, 3, 1, 2) + x_264 = None + x_266 = torch.conv2d( + x_265, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_265 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_2 = sym_sum // 16 + sym_sum_3 = torch.sym_sum([1, floordiv_2]) + floordiv_2 = None + reshape = x_266.reshape(1, 512, -1) + x_266 = None + x_267 = reshape.transpose(1, 2) + reshape = None + view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_30 = x_267 * view + view = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_267, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_52 = sym_sum_3**2 + linear = torch._C._nn.linear( + layer_norm_33, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_33 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_1 = linear.reshape(1, pow_52, 3, 16, 32) + linear = None + qkv = reshape_1.permute(2, 0, 3, 1, 4) + reshape_1 = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_268 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) + q = k = v = None + transpose_1 = x_268.transpose(1, 2) + x_268 = None + x_269 = transpose_1.reshape(1, pow_52, 512) + transpose_1 = None + x_270 = torch._C._nn.linear( + x_269, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_269 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_271 = torch.nn.functional.dropout(x_270, 0.0, False, False) + x_270 = None + x_272 = mul_30 + x_271 + mul_30 = x_271 = None + view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_31 = x_272 * view_1 + view_1 = None + layer_norm_34 = torch.nn.functional.layer_norm( + x_272, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_272 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_273 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_34 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_30 = torch.nn.functional.relu(x_273, inplace=False) + x_273 = None + pow_31 = relu_30**2 + relu_30 = None + mul_32 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_31 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_31 + ) = None + x_274 = ( + mul_32 + + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_275 = torch.nn.functional.dropout(x_274, 0.0, False, False) + x_274 = None + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_277 = torch.nn.functional.dropout(x_276, 0.0, False, False) + x_276 = None + x_278 = mul_31 + x_277 + mul_31 = x_277 = None + view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_33 = x_278 * view_2 + view_2 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_278, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_3 = linear_4.reshape(1, pow_52, 3, 16, 32) + linear_4 = None + qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) + reshape_3 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_279 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) + q_1 = k_1 = v_1 = None + transpose_2 = x_279.transpose(1, 2) + x_279 = None + x_280 = transpose_2.reshape(1, pow_52, 512) + transpose_2 = None + x_281 = torch._C._nn.linear( + x_280, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_280 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) + x_281 = None + x_283 = mul_33 + x_282 + mul_33 = x_282 = None + view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_34 = x_283 * view_3 + view_3 = None + layer_norm_36 = torch.nn.functional.layer_norm( + x_283, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_283 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_284 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_36 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_31 = torch.nn.functional.relu(x_284, inplace=False) + x_284 = None + pow_32 = relu_31**2 + relu_31 = None + mul_35 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_32 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_32 + ) = None + x_285 = ( + mul_35 + + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_286 = torch.nn.functional.dropout(x_285, 0.0, False, False) + x_285 = None + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) + x_287 = None + x_289 = mul_34 + x_288 + mul_34 = x_288 = None + view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_36 = x_289 * view_4 + view_4 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_289, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_37 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_5 = linear_8.reshape(1, pow_52, 3, 16, 32) + linear_8 = None + qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) + reshape_5 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_290 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) + q_2 = k_2 = v_2 = None + transpose_3 = x_290.transpose(1, 2) + x_290 = None + x_291 = transpose_3.reshape(1, pow_52, 512) + transpose_3 = None + x_292 = torch._C._nn.linear( + x_291, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_291 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_293 = torch.nn.functional.dropout(x_292, 0.0, False, False) + x_292 = None + x_294 = mul_36 + x_293 + mul_36 = x_293 = None + view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_37 = x_294 * view_5 + view_5 = None + layer_norm_38 = torch.nn.functional.layer_norm( + x_294, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_294 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_295 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_32 = torch.nn.functional.relu(x_295, inplace=False) + x_295 = None + pow_33 = relu_32**2 + relu_32 = None + mul_38 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_33 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_33 + ) = None + x_296 = ( + mul_38 + + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_297 = torch.nn.functional.dropout(x_296, 0.0, False, False) + x_296 = None + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_299 = torch.nn.functional.dropout(x_298, 0.0, False, False) + x_298 = None + x_300 = mul_37 + x_299 + mul_37 = x_299 = None + view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_39 = x_300 * view_6 + view_6 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_300, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_39 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_7 = linear_12.reshape(1, pow_52, 3, 16, 32) + linear_12 = None + qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) + reshape_7 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_301 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) + q_3 = k_3 = v_3 = None + transpose_4 = x_301.transpose(1, 2) + x_301 = None + x_302 = transpose_4.reshape(1, pow_52, 512) + transpose_4 = None + x_303 = torch._C._nn.linear( + x_302, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_302 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_304 = torch.nn.functional.dropout(x_303, 0.0, False, False) + x_303 = None + x_305 = mul_39 + x_304 + mul_39 = x_304 = None + view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_40 = x_305 * view_7 + view_7 = None + layer_norm_40 = torch.nn.functional.layer_norm( + x_305, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_305 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) + x_306 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_40 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_33 = torch.nn.functional.relu(x_306, inplace=False) + x_306 = None + pow_34 = relu_33**2 + relu_33 = None + mul_41 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + * pow_34 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( + pow_34 + ) = None + x_307 = ( + mul_41 + + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + ) + mul_41 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) + x_308 = torch.nn.functional.dropout(x_307, 0.0, False, False) + x_307 = None + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) + x_309 = None + x_311 = mul_40 + x_310 + mul_40 = x_310 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_42 = x_311 * view_8 + view_8 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_311, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_41 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_9 = linear_16.reshape(1, pow_52, 3, 16, 32) + linear_16 = None + qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) + reshape_9 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_312 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) + q_4 = k_4 = v_4 = None + transpose_5 = x_312.transpose(1, 2) + x_312 = None + x_313 = transpose_5.reshape(1, pow_52, 512) + transpose_5 = None + x_314 = torch._C._nn.linear( + x_313, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_313 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) + x_314 = None + x_316 = mul_42 + x_315 + mul_42 = x_315 = None + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_43 = x_316 * view_9 + view_9 = None + layer_norm_42 = torch.nn.functional.layer_norm( + x_316, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_316 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) + x_317 = torch._C._nn.linear( + layer_norm_42, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_42 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_34 = torch.nn.functional.relu(x_317, inplace=False) + x_317 = None + pow_35 = relu_34**2 + relu_34 = None + mul_44 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + * pow_35 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( + pow_35 + ) = None + x_318 = ( + mul_44 + + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + ) + mul_44 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) + x_319 = torch.nn.functional.dropout(x_318, 0.0, False, False) + x_318 = None + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_321 = torch.nn.functional.dropout(x_320, 0.0, False, False) + x_320 = None + x_322 = mul_43 + x_321 + mul_43 = x_321 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_45 = x_322 * view_10 + view_10 = None + layer_norm_43 = torch.nn.functional.layer_norm( + x_322, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) + linear_20 = torch._C._nn.linear( + layer_norm_43, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_43 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_11 = linear_20.reshape(1, pow_52, 3, 16, 32) + linear_20 = None + qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) + reshape_11 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_323 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) + q_5 = k_5 = v_5 = None + transpose_6 = x_323.transpose(1, 2) + x_323 = None + x_324 = transpose_6.reshape(1, pow_52, 512) + transpose_6 = None + x_325 = torch._C._nn.linear( + x_324, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_324 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_326 = torch.nn.functional.dropout(x_325, 0.0, False, False) + x_325 = None + x_327 = mul_45 + x_326 + mul_45 = x_326 = None + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_46 = x_327 * view_11 + view_11 = None + layer_norm_44 = torch.nn.functional.layer_norm( + x_327, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_327 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) + x_328 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_44 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_35 = torch.nn.functional.relu(x_328, inplace=False) + x_328 = None + pow_36 = relu_35**2 + relu_35 = None + mul_47 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + * pow_36 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( + pow_36 + ) = None + x_329 = ( + mul_47 + + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + ) + mul_47 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) + x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) + x_329 = None + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_332 = torch.nn.functional.dropout(x_331, 0.0, False, False) + x_331 = None + x_333 = mul_46 + x_332 + mul_46 = x_332 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_48 = x_333 * view_12 + view_12 = None + layer_norm_45 = torch.nn.functional.layer_norm( + x_333, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) + linear_24 = torch._C._nn.linear( + layer_norm_45, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_45 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_13 = linear_24.reshape(1, pow_52, 3, 16, 32) + linear_24 = None + qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) + reshape_13 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_334 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) + q_6 = k_6 = v_6 = None + transpose_7 = x_334.transpose(1, 2) + x_334 = None + x_335 = transpose_7.reshape(1, pow_52, 512) + transpose_7 = None + x_336 = torch._C._nn.linear( + x_335, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_335 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_337 = torch.nn.functional.dropout(x_336, 0.0, False, False) + x_336 = None + x_338 = mul_48 + x_337 + mul_48 = x_337 = None + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_49 = x_338 * view_13 + view_13 = None + layer_norm_46 = torch.nn.functional.layer_norm( + x_338, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_338 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) + x_339 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_46 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_36 = torch.nn.functional.relu(x_339, inplace=False) + x_339 = None + pow_37 = relu_36**2 + relu_36 = None + mul_50 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + * pow_37 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( + pow_37 + ) = None + x_340 = ( + mul_50 + + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + ) + mul_50 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) + x_341 = torch.nn.functional.dropout(x_340, 0.0, False, False) + x_340 = None + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_343 = torch.nn.functional.dropout(x_342, 0.0, False, False) + x_342 = None + x_344 = mul_49 + x_343 + mul_49 = x_343 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_51 = x_344 * view_14 + view_14 = None + layer_norm_47 = torch.nn.functional.layer_norm( + x_344, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) + linear_28 = torch._C._nn.linear( + layer_norm_47, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_47 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_15 = linear_28.reshape(1, pow_52, 3, 16, 32) + linear_28 = None + qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) + reshape_15 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_345 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) + q_7 = k_7 = v_7 = None + transpose_8 = x_345.transpose(1, 2) + x_345 = None + x_346 = transpose_8.reshape(1, pow_52, 512) + transpose_8 = None + x_347 = torch._C._nn.linear( + x_346, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_346 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) + x_347 = None + x_349 = mul_51 + x_348 + mul_51 = x_348 = None + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_52 = x_349 * view_15 + view_15 = None + layer_norm_48 = torch.nn.functional.layer_norm( + x_349, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_349 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) + x_350 = torch._C._nn.linear( + layer_norm_48, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_48 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_37 = torch.nn.functional.relu(x_350, inplace=False) + x_350 = None + pow_38 = relu_37**2 + relu_37 = None + mul_53 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + * pow_38 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( + pow_38 + ) = None + x_351 = ( + mul_53 + + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + ) + mul_53 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) + x_352 = torch.nn.functional.dropout(x_351, 0.0, False, False) + x_351 = None + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_354 = torch.nn.functional.dropout(x_353, 0.0, False, False) + x_353 = None + x_355 = mul_52 + x_354 + mul_52 = x_354 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_54 = x_355 * view_16 + view_16 = None + layer_norm_49 = torch.nn.functional.layer_norm( + x_355, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) + linear_32 = torch._C._nn.linear( + layer_norm_49, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_49 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_17 = linear_32.reshape(1, pow_52, 3, 16, 32) + linear_32 = None + qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) + reshape_17 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_356 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) + q_8 = k_8 = v_8 = None + transpose_9 = x_356.transpose(1, 2) + x_356 = None + x_357 = transpose_9.reshape(1, pow_52, 512) + transpose_9 = None + x_358 = torch._C._nn.linear( + x_357, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_357 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_359 = torch.nn.functional.dropout(x_358, 0.0, False, False) + x_358 = None + x_360 = mul_54 + x_359 + mul_54 = x_359 = None + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_55 = x_360 * view_17 + view_17 = None + layer_norm_50 = torch.nn.functional.layer_norm( + x_360, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_360 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) + x_361 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_50 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_38 = torch.nn.functional.relu(x_361, inplace=False) + x_361 = None + pow_39 = relu_38**2 + relu_38 = None + mul_56 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + * pow_39 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( + pow_39 + ) = None + x_362 = ( + mul_56 + + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + ) + mul_56 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) + x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) + x_362 = None + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_365 = torch.nn.functional.dropout(x_364, 0.0, False, False) + x_364 = None + x_366 = mul_55 + x_365 + mul_55 = x_365 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_57 = x_366 * view_18 + view_18 = None + layer_norm_51 = torch.nn.functional.layer_norm( + x_366, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_51, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_51 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_19 = linear_36.reshape(1, pow_52, 3, 16, 32) + linear_36 = None + qkv_9 = reshape_19.permute(2, 0, 3, 1, 4) + reshape_19 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_367 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) + q_9 = k_9 = v_9 = None + transpose_10 = x_367.transpose(1, 2) + x_367 = None + x_368 = transpose_10.reshape(1, pow_52, 512) + transpose_10 = None + x_369 = torch._C._nn.linear( + x_368, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_368 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_370 = torch.nn.functional.dropout(x_369, 0.0, False, False) + x_369 = None + x_371 = mul_57 + x_370 + mul_57 = x_370 = None + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_58 = x_371 * view_19 + view_19 = None + layer_norm_52 = torch.nn.functional.layer_norm( + x_371, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_371 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) + x_372 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_52 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_39 = torch.nn.functional.relu(x_372, inplace=False) + x_372 = None + pow_40 = relu_39**2 + relu_39 = None + mul_59 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + * pow_40 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( + pow_40 + ) = None + x_373 = ( + mul_59 + + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + ) + mul_59 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) + x_374 = torch.nn.functional.dropout(x_373, 0.0, False, False) + x_373 = None + x_375 = torch._C._nn.linear( + x_374, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_374 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_376 = torch.nn.functional.dropout(x_375, 0.0, False, False) + x_375 = None + x_377 = mul_58 + x_376 + mul_58 = x_376 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_60 = x_377 * view_20 + view_20 = None + layer_norm_53 = torch.nn.functional.layer_norm( + x_377, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_377 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_53, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_53 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_21 = linear_40.reshape(1, pow_52, 3, 16, 32) + linear_40 = None + qkv_10 = reshape_21.permute(2, 0, 3, 1, 4) + reshape_21 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_378 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_378.transpose(1, 2) + x_378 = None + x_379 = transpose_11.reshape(1, pow_52, 512) + transpose_11 = None + x_380 = torch._C._nn.linear( + x_379, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_379 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_381 = torch.nn.functional.dropout(x_380, 0.0, False, False) + x_380 = None + x_382 = mul_60 + x_381 + mul_60 = x_381 = None + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_61 = x_382 * view_21 + view_21 = None + layer_norm_54 = torch.nn.functional.layer_norm( + x_382, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_382 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) + x_383 = torch._C._nn.linear( + layer_norm_54, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_54 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_40 = torch.nn.functional.relu(x_383, inplace=False) + x_383 = None + pow_41 = relu_40**2 + relu_40 = None + mul_62 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + * pow_41 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( + pow_41 + ) = None + x_384 = ( + mul_62 + + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + ) + mul_62 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) + x_385 = torch.nn.functional.dropout(x_384, 0.0, False, False) + x_384 = None + x_386 = torch._C._nn.linear( + x_385, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_385 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_387 = torch.nn.functional.dropout(x_386, 0.0, False, False) + x_386 = None + x_388 = mul_61 + x_387 + mul_61 = x_387 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_63 = x_388 * view_22 + view_22 = None + layer_norm_55 = torch.nn.functional.layer_norm( + x_388, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_388 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_55, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_55 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_23 = linear_44.reshape(1, pow_52, 3, 16, 32) + linear_44 = None + qkv_11 = reshape_23.permute(2, 0, 3, 1, 4) + reshape_23 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_389 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_389.transpose(1, 2) + x_389 = None + x_390 = transpose_12.reshape(1, pow_52, 512) + transpose_12 = None + x_391 = torch._C._nn.linear( + x_390, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_390 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_392 = torch.nn.functional.dropout(x_391, 0.0, False, False) + x_391 = None + x_393 = mul_63 + x_392 + mul_63 = x_392 = None + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_64 = x_393 * view_23 + view_23 = None + layer_norm_56 = torch.nn.functional.layer_norm( + x_393, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_393 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) + x_394 = torch._C._nn.linear( + layer_norm_56, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_56 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_41 = torch.nn.functional.relu(x_394, inplace=False) + x_394 = None + pow_42 = relu_41**2 + relu_41 = None + mul_65 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + * pow_42 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( + pow_42 + ) = None + x_395 = ( + mul_65 + + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + ) + mul_65 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) + x_396 = torch.nn.functional.dropout(x_395, 0.0, False, False) + x_395 = None + x_397 = torch._C._nn.linear( + x_396, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_396 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_398 = torch.nn.functional.dropout(x_397, 0.0, False, False) + x_397 = None + x_399 = mul_64 + x_398 + mul_64 = x_398 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_66 = x_399 * view_24 + view_24 = None + layer_norm_57 = torch.nn.functional.layer_norm( + x_399, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_399 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_57, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_57 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_25 = linear_48.reshape(1, pow_52, 3, 16, 32) + linear_48 = None + qkv_12 = reshape_25.permute(2, 0, 3, 1, 4) + reshape_25 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_12 = unbind_12[0] + k_12 = unbind_12[1] + v_12 = unbind_12[2] + unbind_12 = None + x_400 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_13 = x_400.transpose(1, 2) + x_400 = None + x_401 = transpose_13.reshape(1, pow_52, 512) + transpose_13 = None + x_402 = torch._C._nn.linear( + x_401, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_401 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_403 = torch.nn.functional.dropout(x_402, 0.0, False, False) + x_402 = None + x_404 = mul_66 + x_403 + mul_66 = x_403 = None + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_67 = x_404 * view_25 + view_25 = None + layer_norm_58 = torch.nn.functional.layer_norm( + x_404, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_404 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = (None) + x_405 = torch._C._nn.linear( + layer_norm_58, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_58 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_42 = torch.nn.functional.relu(x_405, inplace=False) + x_405 = None + pow_43 = relu_42**2 + relu_42 = None + mul_68 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ + * pow_43 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = ( + pow_43 + ) = None + x_406 = ( + mul_68 + + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ + ) + mul_68 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = (None) + x_407 = torch.nn.functional.dropout(x_406, 0.0, False, False) + x_406 = None + x_408 = torch._C._nn.linear( + x_407, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_407 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_409 = torch.nn.functional.dropout(x_408, 0.0, False, False) + x_408 = None + x_410 = mul_67 + x_409 + mul_67 = x_409 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_69 = x_410 * view_26 + view_26 = None + layer_norm_59 = torch.nn.functional.layer_norm( + x_410, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_410 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = (None) + linear_52 = torch._C._nn.linear( + layer_norm_59, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_59 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_27 = linear_52.reshape(1, pow_52, 3, 16, 32) + linear_52 = None + qkv_13 = reshape_27.permute(2, 0, 3, 1, 4) + reshape_27 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_13 = unbind_13[0] + k_13 = unbind_13[1] + v_13 = unbind_13[2] + unbind_13 = None + x_411 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_14 = x_411.transpose(1, 2) + x_411 = None + x_412 = transpose_14.reshape(1, pow_52, 512) + transpose_14 = None + x_413 = torch._C._nn.linear( + x_412, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_412 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_414 = torch.nn.functional.dropout(x_413, 0.0, False, False) + x_413 = None + x_415 = mul_69 + x_414 + mul_69 = x_414 = None + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_70 = x_415 * view_27 + view_27 = None + layer_norm_60 = torch.nn.functional.layer_norm( + x_415, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_415 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = (None) + x_416 = torch._C._nn.linear( + layer_norm_60, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_60 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_43 = torch.nn.functional.relu(x_416, inplace=False) + x_416 = None + pow_44 = relu_43**2 + relu_43 = None + mul_71 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ + * pow_44 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = ( + pow_44 + ) = None + x_417 = ( + mul_71 + + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ + ) + mul_71 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = (None) + x_418 = torch.nn.functional.dropout(x_417, 0.0, False, False) + x_417 = None + x_419 = torch._C._nn.linear( + x_418, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_418 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_420 = torch.nn.functional.dropout(x_419, 0.0, False, False) + x_419 = None + x_421 = mul_70 + x_420 + mul_70 = x_420 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_72 = x_421 * view_28 + view_28 = None + layer_norm_61 = torch.nn.functional.layer_norm( + x_421, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_421 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_61, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_61 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_29 = linear_56.reshape(1, pow_52, 3, 16, 32) + linear_56 = None + qkv_14 = reshape_29.permute(2, 0, 3, 1, 4) + reshape_29 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_14 = unbind_14[0] + k_14 = unbind_14[1] + v_14 = unbind_14[2] + unbind_14 = None + x_422 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_15 = x_422.transpose(1, 2) + x_422 = None + x_423 = transpose_15.reshape(1, pow_52, 512) + transpose_15 = None + x_424 = torch._C._nn.linear( + x_423, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_423 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_425 = torch.nn.functional.dropout(x_424, 0.0, False, False) + x_424 = None + x_426 = mul_72 + x_425 + mul_72 = x_425 = None + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_73 = x_426 * view_29 + view_29 = None + layer_norm_62 = torch.nn.functional.layer_norm( + x_426, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_426 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = (None) + x_427 = torch._C._nn.linear( + layer_norm_62, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_62 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_44 = torch.nn.functional.relu(x_427, inplace=False) + x_427 = None + pow_45 = relu_44**2 + relu_44 = None + mul_74 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ + * pow_45 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = ( + pow_45 + ) = None + x_428 = ( + mul_74 + + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ + ) + mul_74 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = (None) + x_429 = torch.nn.functional.dropout(x_428, 0.0, False, False) + x_428 = None + x_430 = torch._C._nn.linear( + x_429, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_429 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_431 = torch.nn.functional.dropout(x_430, 0.0, False, False) + x_430 = None + x_432 = mul_73 + x_431 + mul_73 = x_431 = None + view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_75 = x_432 * view_30 + view_30 = None + layer_norm_63 = torch.nn.functional.layer_norm( + x_432, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_432 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = (None) + linear_60 = torch._C._nn.linear( + layer_norm_63, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_63 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_31 = linear_60.reshape(1, pow_52, 3, 16, 32) + linear_60 = None + qkv_15 = reshape_31.permute(2, 0, 3, 1, 4) + reshape_31 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_15 = unbind_15[0] + k_15 = unbind_15[1] + v_15 = unbind_15[2] + unbind_15 = None + x_433 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_16 = x_433.transpose(1, 2) + x_433 = None + x_434 = transpose_16.reshape(1, pow_52, 512) + transpose_16 = None + x_435 = torch._C._nn.linear( + x_434, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_434 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_436 = torch.nn.functional.dropout(x_435, 0.0, False, False) + x_435 = None + x_437 = mul_75 + x_436 + mul_75 = x_436 = None + view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_76 = x_437 * view_31 + view_31 = None + layer_norm_64 = torch.nn.functional.layer_norm( + x_437, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_437 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = (None) + x_438 = torch._C._nn.linear( + layer_norm_64, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_64 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_45 = torch.nn.functional.relu(x_438, inplace=False) + x_438 = None + pow_46 = relu_45**2 + relu_45 = None + mul_77 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ + * pow_46 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = ( + pow_46 + ) = None + x_439 = ( + mul_77 + + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ + ) + mul_77 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = (None) + x_440 = torch.nn.functional.dropout(x_439, 0.0, False, False) + x_439 = None + x_441 = torch._C._nn.linear( + x_440, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_440 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_442 = torch.nn.functional.dropout(x_441, 0.0, False, False) + x_441 = None + x_443 = mul_76 + x_442 + mul_76 = x_442 = None + view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_78 = x_443 * view_32 + view_32 = None + layer_norm_65 = torch.nn.functional.layer_norm( + x_443, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_443 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_65, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_65 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_33 = linear_64.reshape(1, pow_52, 3, 16, 32) + linear_64 = None + qkv_16 = reshape_33.permute(2, 0, 3, 1, 4) + reshape_33 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_16 = unbind_16[0] + k_16 = unbind_16[1] + v_16 = unbind_16[2] + unbind_16 = None + x_444 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_17 = x_444.transpose(1, 2) + x_444 = None + x_445 = transpose_17.reshape(1, pow_52, 512) + transpose_17 = None + x_446 = torch._C._nn.linear( + x_445, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_445 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_447 = torch.nn.functional.dropout(x_446, 0.0, False, False) + x_446 = None + x_448 = mul_78 + x_447 + mul_78 = x_447 = None + view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_79 = x_448 * view_33 + view_33 = None + layer_norm_66 = torch.nn.functional.layer_norm( + x_448, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_448 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = (None) + x_449 = torch._C._nn.linear( + layer_norm_66, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_66 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_46 = torch.nn.functional.relu(x_449, inplace=False) + x_449 = None + pow_47 = relu_46**2 + relu_46 = None + mul_80 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ + * pow_47 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = ( + pow_47 + ) = None + x_450 = ( + mul_80 + + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ + ) + mul_80 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = (None) + x_451 = torch.nn.functional.dropout(x_450, 0.0, False, False) + x_450 = None + x_452 = torch._C._nn.linear( + x_451, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_451 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_453 = torch.nn.functional.dropout(x_452, 0.0, False, False) + x_452 = None + x_454 = mul_79 + x_453 + mul_79 = x_453 = None + view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_81 = x_454 * view_34 + view_34 = None + layer_norm_67 = torch.nn.functional.layer_norm( + x_454, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_454 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = (None) + linear_68 = torch._C._nn.linear( + layer_norm_67, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_67 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_35 = linear_68.reshape(1, pow_52, 3, 16, 32) + linear_68 = None + qkv_17 = reshape_35.permute(2, 0, 3, 1, 4) + reshape_35 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_17 = unbind_17[0] + k_17 = unbind_17[1] + v_17 = unbind_17[2] + unbind_17 = None + x_455 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_18 = x_455.transpose(1, 2) + x_455 = None + x_456 = transpose_18.reshape(1, pow_52, 512) + transpose_18 = pow_52 = None + x_457 = torch._C._nn.linear( + x_456, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_456 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_458 = torch.nn.functional.dropout(x_457, 0.0, False, False) + x_457 = None + x_459 = mul_81 + x_458 + mul_81 = x_458 = None + view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_82 = x_459 * view_35 + view_35 = None + layer_norm_68 = torch.nn.functional.layer_norm( + x_459, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_459 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = (None) + x_460 = torch._C._nn.linear( + layer_norm_68, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_68 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_47 = torch.nn.functional.relu(x_460, inplace=False) + x_460 = None + pow_48 = relu_47**2 + relu_47 = None + mul_83 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ + * pow_48 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = ( + pow_48 + ) = None + x_461 = ( + mul_83 + + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ + ) + mul_83 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = (None) + x_462 = torch.nn.functional.dropout(x_461, 0.0, False, False) + x_461 = None + x_463 = torch._C._nn.linear( + x_462, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_462 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_464 = torch.nn.functional.dropout(x_463, 0.0, False, False) + x_463 = None + x_465 = mul_82 + x_464 + mul_82 = x_464 = None + transpose_19 = x_465.transpose(1, 2) + x_465 = None + x_466 = transpose_19.reshape(1, 512, sym_sum_3, sym_sum_3) + transpose_19 = sym_sum_3 = None + x_467 = x_466.permute(0, 2, 3, 1) + x_466 = None + x_468 = torch.nn.functional.layer_norm( + x_467, + (512,), + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_467 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) + x_469 = x_468.permute(0, 3, 1, 2) + x_468 = None + x_470 = torch.conv2d( + x_469, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_469 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_3 = sym_sum // 32 + sym_sum = None + sym_sum_4 = torch.sym_sum([1, floordiv_3]) + floordiv_3 = None + reshape_38 = x_470.reshape(1, 768, -1) + x_470 = None + x_471 = reshape_38.transpose(1, 2) + reshape_38 = None + view_36 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_84 = x_471 * view_36 + view_36 = None + layer_norm_70 = torch.nn.functional.layer_norm( + x_471, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_471 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_53 = sym_sum_4**2 + linear_72 = torch._C._nn.linear( + layer_norm_70, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_70 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_39 = linear_72.reshape(1, pow_53, 3, 24, 32) + linear_72 = None + qkv_18 = reshape_39.permute(2, 0, 3, 1, 4) + reshape_39 = None + unbind_18 = qkv_18.unbind(0) + qkv_18 = None + q_18 = unbind_18[0] + k_18 = unbind_18[1] + v_18 = unbind_18[2] + unbind_18 = None + x_472 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_21 = x_472.transpose(1, 2) + x_472 = None + x_473 = transpose_21.reshape(1, pow_53, 768) + transpose_21 = None + x_474 = torch._C._nn.linear( + x_473, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_473 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_475 = torch.nn.functional.dropout(x_474, 0.0, False, False) + x_474 = None + x_476 = mul_84 + x_475 + mul_84 = x_475 = None + view_37 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_85 = x_476 * view_37 + view_37 = None + layer_norm_71 = torch.nn.functional.layer_norm( + x_476, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_476 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_477 = torch._C._nn.linear( + layer_norm_71, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_71 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_48 = torch.nn.functional.relu(x_477, inplace=False) + x_477 = None + pow_49 = relu_48**2 + relu_48 = None + mul_86 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_49 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_49 + ) = None + x_478 = ( + mul_86 + + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_86 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_479 = torch.nn.functional.dropout(x_478, 0.0, False, False) + x_478 = None + x_480 = torch._C._nn.linear( + x_479, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_479 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_481 = torch.nn.functional.dropout(x_480, 0.0, False, False) + x_480 = None + x_482 = mul_85 + x_481 + mul_85 = x_481 = None + view_38 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_87 = x_482 * view_38 + view_38 = None + layer_norm_72 = torch.nn.functional.layer_norm( + x_482, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_482 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_76 = torch._C._nn.linear( + layer_norm_72, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_72 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_41 = linear_76.reshape(1, pow_53, 3, 24, 32) + linear_76 = None + qkv_19 = reshape_41.permute(2, 0, 3, 1, 4) + reshape_41 = None + unbind_19 = qkv_19.unbind(0) + qkv_19 = None + q_19 = unbind_19[0] + k_19 = unbind_19[1] + v_19 = unbind_19[2] + unbind_19 = None + x_483 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_22 = x_483.transpose(1, 2) + x_483 = None + x_484 = transpose_22.reshape(1, pow_53, 768) + transpose_22 = None + x_485 = torch._C._nn.linear( + x_484, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_484 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_486 = torch.nn.functional.dropout(x_485, 0.0, False, False) + x_485 = None + x_487 = mul_87 + x_486 + mul_87 = x_486 = None + view_39 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_88 = x_487 * view_39 + view_39 = None + layer_norm_73 = torch.nn.functional.layer_norm( + x_487, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_487 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_488 = torch._C._nn.linear( + layer_norm_73, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_73 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_49 = torch.nn.functional.relu(x_488, inplace=False) + x_488 = None + pow_50 = relu_49**2 + relu_49 = None + mul_89 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_50 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_50 + ) = None + x_489 = ( + mul_89 + + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_89 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_490 = torch.nn.functional.dropout(x_489, 0.0, False, False) + x_489 = None + x_491 = torch._C._nn.linear( + x_490, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_490 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_492 = torch.nn.functional.dropout(x_491, 0.0, False, False) + x_491 = None + x_493 = mul_88 + x_492 + mul_88 = x_492 = None + view_40 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_90 = x_493 * view_40 + view_40 = None + layer_norm_74 = torch.nn.functional.layer_norm( + x_493, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_493 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_74, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_74 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_43 = linear_80.reshape(1, pow_53, 3, 24, 32) + linear_80 = None + qkv_20 = reshape_43.permute(2, 0, 3, 1, 4) + reshape_43 = None + unbind_20 = qkv_20.unbind(0) + qkv_20 = None + q_20 = unbind_20[0] + k_20 = unbind_20[1] + v_20 = unbind_20[2] + unbind_20 = None + x_494 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_23 = x_494.transpose(1, 2) + x_494 = None + x_495 = transpose_23.reshape(1, pow_53, 768) + transpose_23 = pow_53 = None + x_496 = torch._C._nn.linear( + x_495, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_495 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_497 = torch.nn.functional.dropout(x_496, 0.0, False, False) + x_496 = None + x_498 = mul_90 + x_497 + mul_90 = x_497 = None + view_41 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_91 = x_498 * view_41 + view_41 = None + layer_norm_75 = torch.nn.functional.layer_norm( + x_498, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_498 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_499 = torch._C._nn.linear( + layer_norm_75, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_75 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_50 = torch.nn.functional.relu(x_499, inplace=False) + x_499 = None + pow_51 = relu_50**2 + relu_50 = None + mul_92 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_51 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_51 + ) = None + x_500 = ( + mul_92 + + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_92 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_501 = torch.nn.functional.dropout(x_500, 0.0, False, False) + x_500 = None + x_502 = torch._C._nn.linear( + x_501, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_501 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_503 = torch.nn.functional.dropout(x_502, 0.0, False, False) + x_502 = None + x_504 = mul_91 + x_503 + mul_91 = x_503 = None + transpose_24 = x_504.transpose(1, 2) + x_504 = None + x_505 = transpose_24.reshape(1, 768, sym_sum_4, sym_sum_4) + transpose_24 = sym_sum_4 = None + x_506 = torch.nn.functional.adaptive_avg_pool2d(x_505, 1) + x_505 = None + x_507 = x_506.permute(0, 2, 3, 1) + x_506 = None + x_508 = torch.nn.functional.layer_norm( + x_507, + (768,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_507 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_509 = x_508.permute(0, 3, 1, 2) + x_508 = None + x_510 = x_509.flatten(1, -1) + x_509 = None + x_511 = torch.nn.functional.dropout(x_510, 0.0, False, False) + x_510 = None + x_512 = torch._C._nn.linear( + x_511, + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, + ) + x_511 = ( + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None + relu_51 = torch.nn.functional.relu(x_512, inplace=False) + x_512 = None + x_513 = torch.square(relu_51) + relu_51 = None + x_514 = torch.nn.functional.layer_norm( + x_513, + (3072,), + l_self_modules_head_modules_fc_modules_norm_parameters_weight_, + l_self_modules_head_modules_fc_modules_norm_parameters_bias_, + 1e-06, + ) + x_513 = ( + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None + x_515 = torch.nn.functional.dropout(x_514, 0.0, False, False) + x_514 = None + x_516 = torch._C._nn.linear( + x_515, + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, + ) + x_515 = ( + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None + return (x_516,) diff --git a/samples/timm/caformer_b36.sail_in1k_384/weight_meta.py b/samples/timm/caformer_b36.sail_in1k_384/weight_meta.py new file mode 100644 index 0000000000..827d4a9370 --- /dev/null +++ b/samples/timm/caformer_b36.sail_in1k_384/weight_meta.py @@ -0,0 +1,3968 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv_parameters_weight_" + shape = [128, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: + name = "L_self_modules_stem_modules_conv_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.226 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: + name = "L_self_modules_stem_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" + shape = [768, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" + shape = [1000, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/caformer_b36.sail_in22k/graph_net.json b/samples/timm/caformer_b36.sail_in22k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/caformer_b36.sail_in22k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/caformer_b36.sail_in22k/input_meta.py b/samples/timm/caformer_b36.sail_in22k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_b36.sail_in22k/input_tensor_constraints.py b/samples/timm/caformer_b36.sail_in22k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_b36.sail_in22k/model.py b/samples/timm/caformer_b36.sail_in22k/model.py new file mode 100644 index 0000000000..08d3de35cc --- /dev/null +++ b/samples/timm/caformer_b36.sail_in22k/model.py @@ -0,0 +1,4758 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv_parameters_bias_ = ( + L_self_modules_stem_modules_conv_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_norm_parameters_weight_ = ( + L_self_modules_stem_modules_norm_parameters_weight_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv_parameters_weight_, + l_self_modules_stem_modules_conv_parameters_bias_, + (4, 4), + (2, 2), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_conv_parameters_weight_ + ) = l_self_modules_stem_modules_conv_parameters_bias_ = None + x_1 = x.permute(0, 2, 3, 1) + x = None + x_2 = torch.nn.functional.layer_norm( + x_1, + (128,), + l_self_modules_stem_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None + x_3 = x_2.permute(0, 3, 1, 2) + x_2 = None + sym_sum = torch.sym_sum([-3, s1]) + s1 = None + floordiv = sym_sum // 4 + sym_sum_1 = torch.sym_sum([1, floordiv]) + floordiv = sym_sum_1 = None + x_4 = x_3.permute(0, 2, 3, 1) + x_5 = torch.nn.functional.layer_norm( + x_4, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu = torch.nn.functional.relu(x_7, inplace=False) + x_7 = None + pow_1 = relu**2 + relu = None + mul = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_1 + ) = None + x_8 = ( + mul + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_11 = x_3 + x_10 + x_3 = x_10 = None + x_12 = x_11.permute(0, 2, 3, 1) + x_13 = torch.nn.functional.layer_norm( + x_12, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_14 = x_13.permute(0, 3, 1, 2) + x_13 = None + x_15 = torch.conv2d( + x_14, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_1 = torch.nn.functional.relu(x_15, inplace=False) + x_15 = None + pow_2 = relu_1**2 + relu_1 = None + mul_1 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_2 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_2 + ) = None + x_16 = ( + mul_1 + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = x_11 + x_19 + x_11 = x_19 = None + x_21 = x_20.permute(0, 2, 3, 1) + x_22 = torch.nn.functional.layer_norm( + x_21, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = torch.conv2d( + x_23, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_2 = torch.nn.functional.relu(x_24, inplace=False) + x_24 = None + pow_3 = relu_2**2 + relu_2 = None + mul_2 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_3 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_3 + ) = None + x_25 = ( + mul_2 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_27 = torch.conv2d( + x_26, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_28 = x_20 + x_27 + x_20 = x_27 = None + x_29 = x_28.permute(0, 2, 3, 1) + x_30 = torch.nn.functional.layer_norm( + x_29, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_3 = torch.nn.functional.relu(x_32, inplace=False) + x_32 = None + pow_4 = relu_3**2 + relu_3 = None + mul_3 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_4 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_4 + ) = None + x_33 = ( + mul_3 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + x_37 = x_28 + x_36 + x_28 = x_36 = None + x_38 = x_37.permute(0, 2, 3, 1) + x_39 = torch.nn.functional.layer_norm( + x_38, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_40 = x_39.permute(0, 3, 1, 2) + x_39 = None + x_41 = torch.conv2d( + x_40, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_4 = torch.nn.functional.relu(x_41, inplace=False) + x_41 = None + pow_5 = relu_4**2 + relu_4 = None + mul_4 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_5 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_5 + ) = None + x_42 = ( + mul_4 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_44 = torch.conv2d( + x_43, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_45 = x_37 + x_44 + x_37 = x_44 = None + x_46 = x_45.permute(0, 2, 3, 1) + x_47 = torch.nn.functional.layer_norm( + x_46, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_5 = torch.nn.functional.relu(x_49, inplace=False) + x_49 = None + pow_6 = relu_5**2 + relu_5 = None + mul_5 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_6 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_6 + ) = None + x_50 = ( + mul_5 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = x_45 + x_53 + x_45 = x_53 = None + x_55 = x_54.permute(0, 2, 3, 1) + x_54 = None + x_56 = torch.nn.functional.layer_norm( + x_55, + (128,), + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) + x_57 = x_56.permute(0, 3, 1, 2) + x_56 = None + x_58 = torch.conv2d( + x_57, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_1 = sym_sum // 8 + sym_sum_2 = torch.sym_sum([1, floordiv_1]) + floordiv_1 = sym_sum_2 = None + x_59 = x_58.permute(0, 2, 3, 1) + x_60 = torch.nn.functional.layer_norm( + x_59, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_61 = x_60.permute(0, 3, 1, 2) + x_60 = None + x_62 = torch.conv2d( + x_61, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_6 = torch.nn.functional.relu(x_62, inplace=False) + x_62 = None + pow_7 = relu_6**2 + relu_6 = None + mul_6 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_7 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_7 + ) = None + x_63 = ( + mul_6 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_66 = x_58 + x_65 + x_58 = x_65 = None + x_67 = x_66.permute(0, 2, 3, 1) + x_68 = torch.nn.functional.layer_norm( + x_67, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + x_70 = torch.conv2d( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_7 = torch.nn.functional.relu(x_70, inplace=False) + x_70 = None + pow_8 = relu_7**2 + relu_7 = None + mul_7 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_8 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_8 + ) = None + x_71 = ( + mul_7 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + x_73 = torch.conv2d( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + x_75 = x_66 + x_74 + x_66 = x_74 = None + x_76 = x_75.permute(0, 2, 3, 1) + x_77 = torch.nn.functional.layer_norm( + x_76, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_8 = torch.nn.functional.relu(x_79, inplace=False) + x_79 = None + pow_9 = relu_8**2 + relu_8 = None + mul_8 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_9 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_9 + ) = None + x_80 = ( + mul_8 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_81 = torch.conv2d( + x_80, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_83 = x_75 + x_82 + x_75 = x_82 = None + x_84 = x_83.permute(0, 2, 3, 1) + x_85 = torch.nn.functional.layer_norm( + x_84, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_86 = x_85.permute(0, 3, 1, 2) + x_85 = None + x_87 = torch.conv2d( + x_86, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_9 = torch.nn.functional.relu(x_87, inplace=False) + x_87 = None + pow_10 = relu_9**2 + relu_9 = None + mul_9 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_10 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_10 + ) = None + x_88 = ( + mul_9 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = x_83 + x_91 + x_83 = x_91 = None + x_93 = x_92.permute(0, 2, 3, 1) + x_94 = torch.nn.functional.layer_norm( + x_93, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = torch.conv2d( + x_95, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_10 = torch.nn.functional.relu(x_96, inplace=False) + x_96 = None + pow_11 = relu_10**2 + relu_10 = None + mul_10 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_11 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_11 + ) = None + x_97 = ( + mul_10 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_98 = torch.conv2d( + x_97, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_99 = torch.conv2d( + x_98, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_100 = x_92 + x_99 + x_92 = x_99 = None + x_101 = x_100.permute(0, 2, 3, 1) + x_102 = torch.nn.functional.layer_norm( + x_101, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_103 = x_102.permute(0, 3, 1, 2) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_11 = torch.nn.functional.relu(x_104, inplace=False) + x_104 = None + pow_12 = relu_11**2 + relu_11 = None + mul_11 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_12 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_12 + ) = None + x_105 = ( + mul_11 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_100 + x_108 + x_100 = x_108 = None + x_110 = x_109.permute(0, 2, 3, 1) + x_111 = torch.nn.functional.layer_norm( + x_110, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_110 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) + x_112 = x_111.permute(0, 3, 1, 2) + x_111 = None + x_113 = torch.conv2d( + x_112, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_112 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_12 = torch.nn.functional.relu(x_113, inplace=False) + x_113 = None + pow_13 = relu_12**2 + relu_12 = None + mul_12 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ + * pow_13 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_13 + ) = None + x_114 = ( + mul_12 + + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_12 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_115 = torch.conv2d( + x_114, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_114 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_117 = x_109 + x_116 + x_109 = x_116 = None + x_118 = x_117.permute(0, 2, 3, 1) + x_119 = torch.nn.functional.layer_norm( + x_118, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_118 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) + x_120 = x_119.permute(0, 3, 1, 2) + x_119 = None + x_121 = torch.conv2d( + x_120, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_120 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_13 = torch.nn.functional.relu(x_121, inplace=False) + x_121 = None + pow_14 = relu_13**2 + relu_13 = None + mul_13 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + * pow_14 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( + pow_14 + ) = None + x_122 = ( + mul_13 + + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + ) + mul_13 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) + x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) + x_122 = None + x_124 = torch.conv2d( + x_123, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_123 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_126 = x_117 + x_125 + x_117 = x_125 = None + x_127 = x_126.permute(0, 2, 3, 1) + x_128 = torch.nn.functional.layer_norm( + x_127, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_127 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) + x_129 = x_128.permute(0, 3, 1, 2) + x_128 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_14 = torch.nn.functional.relu(x_130, inplace=False) + x_130 = None + pow_15 = relu_14**2 + relu_14 = None + mul_14 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ + * pow_15 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_15 + ) = None + x_131 = ( + mul_14 + + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_14 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_132 = torch.conv2d( + x_131, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_131 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_132 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_134 = x_126 + x_133 + x_126 = x_133 = None + x_135 = x_134.permute(0, 2, 3, 1) + x_136 = torch.nn.functional.layer_norm( + x_135, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_135 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) + x_137 = x_136.permute(0, 3, 1, 2) + x_136 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_15 = torch.nn.functional.relu(x_138, inplace=False) + x_138 = None + pow_16 = relu_15**2 + relu_15 = None + mul_15 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + * pow_16 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( + pow_16 + ) = None + x_139 = ( + mul_15 + + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + ) + mul_15 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) + x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) + x_139 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_140 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_134 + x_142 + x_134 = x_142 = None + x_144 = x_143.permute(0, 2, 3, 1) + x_145 = torch.nn.functional.layer_norm( + x_144, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_144 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) + x_146 = x_145.permute(0, 3, 1, 2) + x_145 = None + x_147 = torch.conv2d( + x_146, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_146 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_16 = torch.nn.functional.relu(x_147, inplace=False) + x_147 = None + pow_17 = relu_16**2 + relu_16 = None + mul_16 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ + * pow_17 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_17 + ) = None + x_148 = ( + mul_16 + + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_16 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_149 = torch.conv2d( + x_148, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_148 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_150 = torch.conv2d( + x_149, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_149 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_151 = x_143 + x_150 + x_143 = x_150 = None + x_152 = x_151.permute(0, 2, 3, 1) + x_153 = torch.nn.functional.layer_norm( + x_152, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_152 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) + x_154 = x_153.permute(0, 3, 1, 2) + x_153 = None + x_155 = torch.conv2d( + x_154, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_154 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_17 = torch.nn.functional.relu(x_155, inplace=False) + x_155 = None + pow_18 = relu_17**2 + relu_17 = None + mul_17 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + * pow_18 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( + pow_18 + ) = None + x_156 = ( + mul_17 + + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + ) + mul_17 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch.conv2d( + x_157, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_157 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_151 + x_159 + x_151 = x_159 = None + x_161 = x_160.permute(0, 2, 3, 1) + x_162 = torch.nn.functional.layer_norm( + x_161, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_161 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) + x_163 = x_162.permute(0, 3, 1, 2) + x_162 = None + x_164 = torch.conv2d( + x_163, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_163 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_18 = torch.nn.functional.relu(x_164, inplace=False) + x_164 = None + pow_19 = relu_18**2 + relu_18 = None + mul_18 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ + * pow_19 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_19 + ) = None + x_165 = ( + mul_18 + + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_18 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_166 = torch.conv2d( + x_165, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_165 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_167 = torch.conv2d( + x_166, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_166 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_168 = x_160 + x_167 + x_160 = x_167 = None + x_169 = x_168.permute(0, 2, 3, 1) + x_170 = torch.nn.functional.layer_norm( + x_169, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_169 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) + x_171 = x_170.permute(0, 3, 1, 2) + x_170 = None + x_172 = torch.conv2d( + x_171, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_171 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_19 = torch.nn.functional.relu(x_172, inplace=False) + x_172 = None + pow_20 = relu_19**2 + relu_19 = None + mul_19 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + * pow_20 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( + pow_20 + ) = None + x_173 = ( + mul_19 + + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + ) + mul_19 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) + x_174 = torch.nn.functional.dropout(x_173, 0.0, False, False) + x_173 = None + x_175 = torch.conv2d( + x_174, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_174 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = x_168 + x_176 + x_168 = x_176 = None + x_178 = x_177.permute(0, 2, 3, 1) + x_179 = torch.nn.functional.layer_norm( + x_178, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_178 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) + x_180 = x_179.permute(0, 3, 1, 2) + x_179 = None + x_181 = torch.conv2d( + x_180, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_180 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_20 = torch.nn.functional.relu(x_181, inplace=False) + x_181 = None + pow_21 = relu_20**2 + relu_20 = None + mul_20 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ + * pow_21 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_21 + ) = None + x_182 = ( + mul_20 + + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_20 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_183 = torch.conv2d( + x_182, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_182 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_184 = torch.conv2d( + x_183, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_183 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_185 = x_177 + x_184 + x_177 = x_184 = None + x_186 = x_185.permute(0, 2, 3, 1) + x_187 = torch.nn.functional.layer_norm( + x_186, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) + x_188 = x_187.permute(0, 3, 1, 2) + x_187 = None + x_189 = torch.conv2d( + x_188, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_188 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_21 = torch.nn.functional.relu(x_189, inplace=False) + x_189 = None + pow_22 = relu_21**2 + relu_21 = None + mul_21 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + * pow_22 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( + pow_22 + ) = None + x_190 = ( + mul_21 + + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + ) + mul_21 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) + x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) + x_190 = None + x_192 = torch.conv2d( + x_191, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_191 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_185 + x_193 + x_185 = x_193 = None + x_195 = x_194.permute(0, 2, 3, 1) + x_196 = torch.nn.functional.layer_norm( + x_195, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_195 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) + x_197 = x_196.permute(0, 3, 1, 2) + x_196 = None + x_198 = torch.conv2d( + x_197, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_197 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_22 = torch.nn.functional.relu(x_198, inplace=False) + x_198 = None + pow_23 = relu_22**2 + relu_22 = None + mul_22 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ + * pow_23 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_23 + ) = None + x_199 = ( + mul_22 + + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_22 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_200 = torch.conv2d( + x_199, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_199 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_201 = torch.conv2d( + x_200, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_200 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_202 = x_194 + x_201 + x_194 = x_201 = None + x_203 = x_202.permute(0, 2, 3, 1) + x_204 = torch.nn.functional.layer_norm( + x_203, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_203 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) + x_205 = x_204.permute(0, 3, 1, 2) + x_204 = None + x_206 = torch.conv2d( + x_205, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_205 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_23 = torch.nn.functional.relu(x_206, inplace=False) + x_206 = None + pow_24 = relu_23**2 + relu_23 = None + mul_23 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + * pow_24 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( + pow_24 + ) = None + x_207 = ( + mul_23 + + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + ) + mul_23 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) + x_208 = torch.nn.functional.dropout(x_207, 0.0, False, False) + x_207 = None + x_209 = torch.conv2d( + x_208, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_208 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) + x_209 = None + x_211 = x_202 + x_210 + x_202 = x_210 = None + x_212 = x_211.permute(0, 2, 3, 1) + x_213 = torch.nn.functional.layer_norm( + x_212, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_212 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) + x_214 = x_213.permute(0, 3, 1, 2) + x_213 = None + x_215 = torch.conv2d( + x_214, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_214 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_24 = torch.nn.functional.relu(x_215, inplace=False) + x_215 = None + pow_25 = relu_24**2 + relu_24 = None + mul_24 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ + * pow_25 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_25 + ) = None + x_216 = ( + mul_24 + + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_24 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_217 = torch.conv2d( + x_216, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_216 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_217 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_219 = x_211 + x_218 + x_211 = x_218 = None + x_220 = x_219.permute(0, 2, 3, 1) + x_221 = torch.nn.functional.layer_norm( + x_220, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_220 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) + x_222 = x_221.permute(0, 3, 1, 2) + x_221 = None + x_223 = torch.conv2d( + x_222, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_222 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_25 = torch.nn.functional.relu(x_223, inplace=False) + x_223 = None + pow_26 = relu_25**2 + relu_25 = None + mul_25 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + * pow_26 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( + pow_26 + ) = None + x_224 = ( + mul_25 + + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + ) + mul_25 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = torch.conv2d( + x_225, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_225 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) + x_226 = None + x_228 = x_219 + x_227 + x_219 = x_227 = None + x_229 = x_228.permute(0, 2, 3, 1) + x_230 = torch.nn.functional.layer_norm( + x_229, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_229 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) + x_231 = x_230.permute(0, 3, 1, 2) + x_230 = None + x_232 = torch.conv2d( + x_231, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_231 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_26 = torch.nn.functional.relu(x_232, inplace=False) + x_232 = None + pow_27 = relu_26**2 + relu_26 = None + mul_26 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ + * pow_27 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_27 + ) = None + x_233 = ( + mul_26 + + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_26 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_234 = torch.conv2d( + x_233, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_233 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_235 = torch.conv2d( + x_234, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_234 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_236 = x_228 + x_235 + x_228 = x_235 = None + x_237 = x_236.permute(0, 2, 3, 1) + x_238 = torch.nn.functional.layer_norm( + x_237, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_237 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) + x_239 = x_238.permute(0, 3, 1, 2) + x_238 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_239 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_27 = torch.nn.functional.relu(x_240, inplace=False) + x_240 = None + pow_28 = relu_27**2 + relu_27 = None + mul_27 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + * pow_28 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( + pow_28 + ) = None + x_241 = ( + mul_27 + + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + ) + mul_27 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) + x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) + x_241 = None + x_243 = torch.conv2d( + x_242, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_242 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_244 = torch.nn.functional.dropout(x_243, 0.0, False, False) + x_243 = None + x_245 = x_236 + x_244 + x_236 = x_244 = None + x_246 = x_245.permute(0, 2, 3, 1) + x_247 = torch.nn.functional.layer_norm( + x_246, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_246 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) + x_248 = x_247.permute(0, 3, 1, 2) + x_247 = None + x_249 = torch.conv2d( + x_248, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_248 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_28 = torch.nn.functional.relu(x_249, inplace=False) + x_249 = None + pow_29 = relu_28**2 + relu_28 = None + mul_28 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ + * pow_29 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_29 + ) = None + x_250 = ( + mul_28 + + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_28 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_250 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_252 = torch.conv2d( + x_251, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_251 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_253 = x_245 + x_252 + x_245 = x_252 = None + x_254 = x_253.permute(0, 2, 3, 1) + x_255 = torch.nn.functional.layer_norm( + x_254, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_254 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) + x_256 = x_255.permute(0, 3, 1, 2) + x_255 = None + x_257 = torch.conv2d( + x_256, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_256 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_29 = torch.nn.functional.relu(x_257, inplace=False) + x_257 = None + pow_30 = relu_29**2 + relu_29 = None + mul_29 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + * pow_30 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( + pow_30 + ) = None + x_258 = ( + mul_29 + + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + ) + mul_29 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = torch.conv2d( + x_259, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_259 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_261 = torch.nn.functional.dropout(x_260, 0.0, False, False) + x_260 = None + x_262 = x_253 + x_261 + x_253 = x_261 = None + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (256,), + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) + x_265 = x_264.permute(0, 3, 1, 2) + x_264 = None + x_266 = torch.conv2d( + x_265, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_265 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_2 = sym_sum // 16 + sym_sum_3 = torch.sym_sum([1, floordiv_2]) + floordiv_2 = None + reshape = x_266.reshape(1, 512, -1) + x_266 = None + x_267 = reshape.transpose(1, 2) + reshape = None + view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_30 = x_267 * view + view = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_267, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_52 = sym_sum_3**2 + linear = torch._C._nn.linear( + layer_norm_33, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_33 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_1 = linear.reshape(1, pow_52, 3, 16, 32) + linear = None + qkv = reshape_1.permute(2, 0, 3, 1, 4) + reshape_1 = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_268 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) + q = k = v = None + transpose_1 = x_268.transpose(1, 2) + x_268 = None + x_269 = transpose_1.reshape(1, pow_52, 512) + transpose_1 = None + x_270 = torch._C._nn.linear( + x_269, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_269 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_271 = torch.nn.functional.dropout(x_270, 0.0, False, False) + x_270 = None + x_272 = mul_30 + x_271 + mul_30 = x_271 = None + view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_31 = x_272 * view_1 + view_1 = None + layer_norm_34 = torch.nn.functional.layer_norm( + x_272, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_272 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_273 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_34 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_30 = torch.nn.functional.relu(x_273, inplace=False) + x_273 = None + pow_31 = relu_30**2 + relu_30 = None + mul_32 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_31 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_31 + ) = None + x_274 = ( + mul_32 + + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_275 = torch.nn.functional.dropout(x_274, 0.0, False, False) + x_274 = None + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_277 = torch.nn.functional.dropout(x_276, 0.0, False, False) + x_276 = None + x_278 = mul_31 + x_277 + mul_31 = x_277 = None + view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_33 = x_278 * view_2 + view_2 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_278, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_3 = linear_4.reshape(1, pow_52, 3, 16, 32) + linear_4 = None + qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) + reshape_3 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_279 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) + q_1 = k_1 = v_1 = None + transpose_2 = x_279.transpose(1, 2) + x_279 = None + x_280 = transpose_2.reshape(1, pow_52, 512) + transpose_2 = None + x_281 = torch._C._nn.linear( + x_280, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_280 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) + x_281 = None + x_283 = mul_33 + x_282 + mul_33 = x_282 = None + view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_34 = x_283 * view_3 + view_3 = None + layer_norm_36 = torch.nn.functional.layer_norm( + x_283, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_283 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_284 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_36 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_31 = torch.nn.functional.relu(x_284, inplace=False) + x_284 = None + pow_32 = relu_31**2 + relu_31 = None + mul_35 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_32 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_32 + ) = None + x_285 = ( + mul_35 + + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_286 = torch.nn.functional.dropout(x_285, 0.0, False, False) + x_285 = None + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) + x_287 = None + x_289 = mul_34 + x_288 + mul_34 = x_288 = None + view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_36 = x_289 * view_4 + view_4 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_289, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_37 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_5 = linear_8.reshape(1, pow_52, 3, 16, 32) + linear_8 = None + qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) + reshape_5 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_290 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) + q_2 = k_2 = v_2 = None + transpose_3 = x_290.transpose(1, 2) + x_290 = None + x_291 = transpose_3.reshape(1, pow_52, 512) + transpose_3 = None + x_292 = torch._C._nn.linear( + x_291, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_291 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_293 = torch.nn.functional.dropout(x_292, 0.0, False, False) + x_292 = None + x_294 = mul_36 + x_293 + mul_36 = x_293 = None + view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_37 = x_294 * view_5 + view_5 = None + layer_norm_38 = torch.nn.functional.layer_norm( + x_294, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_294 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_295 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_32 = torch.nn.functional.relu(x_295, inplace=False) + x_295 = None + pow_33 = relu_32**2 + relu_32 = None + mul_38 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_33 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_33 + ) = None + x_296 = ( + mul_38 + + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_297 = torch.nn.functional.dropout(x_296, 0.0, False, False) + x_296 = None + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_299 = torch.nn.functional.dropout(x_298, 0.0, False, False) + x_298 = None + x_300 = mul_37 + x_299 + mul_37 = x_299 = None + view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_39 = x_300 * view_6 + view_6 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_300, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_39 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_7 = linear_12.reshape(1, pow_52, 3, 16, 32) + linear_12 = None + qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) + reshape_7 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_301 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) + q_3 = k_3 = v_3 = None + transpose_4 = x_301.transpose(1, 2) + x_301 = None + x_302 = transpose_4.reshape(1, pow_52, 512) + transpose_4 = None + x_303 = torch._C._nn.linear( + x_302, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_302 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_304 = torch.nn.functional.dropout(x_303, 0.0, False, False) + x_303 = None + x_305 = mul_39 + x_304 + mul_39 = x_304 = None + view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_40 = x_305 * view_7 + view_7 = None + layer_norm_40 = torch.nn.functional.layer_norm( + x_305, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_305 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) + x_306 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_40 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_33 = torch.nn.functional.relu(x_306, inplace=False) + x_306 = None + pow_34 = relu_33**2 + relu_33 = None + mul_41 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + * pow_34 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( + pow_34 + ) = None + x_307 = ( + mul_41 + + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + ) + mul_41 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) + x_308 = torch.nn.functional.dropout(x_307, 0.0, False, False) + x_307 = None + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) + x_309 = None + x_311 = mul_40 + x_310 + mul_40 = x_310 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_42 = x_311 * view_8 + view_8 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_311, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_41 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_9 = linear_16.reshape(1, pow_52, 3, 16, 32) + linear_16 = None + qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) + reshape_9 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_312 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) + q_4 = k_4 = v_4 = None + transpose_5 = x_312.transpose(1, 2) + x_312 = None + x_313 = transpose_5.reshape(1, pow_52, 512) + transpose_5 = None + x_314 = torch._C._nn.linear( + x_313, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_313 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) + x_314 = None + x_316 = mul_42 + x_315 + mul_42 = x_315 = None + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_43 = x_316 * view_9 + view_9 = None + layer_norm_42 = torch.nn.functional.layer_norm( + x_316, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_316 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) + x_317 = torch._C._nn.linear( + layer_norm_42, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_42 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_34 = torch.nn.functional.relu(x_317, inplace=False) + x_317 = None + pow_35 = relu_34**2 + relu_34 = None + mul_44 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + * pow_35 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( + pow_35 + ) = None + x_318 = ( + mul_44 + + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + ) + mul_44 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) + x_319 = torch.nn.functional.dropout(x_318, 0.0, False, False) + x_318 = None + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_321 = torch.nn.functional.dropout(x_320, 0.0, False, False) + x_320 = None + x_322 = mul_43 + x_321 + mul_43 = x_321 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_45 = x_322 * view_10 + view_10 = None + layer_norm_43 = torch.nn.functional.layer_norm( + x_322, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) + linear_20 = torch._C._nn.linear( + layer_norm_43, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_43 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_11 = linear_20.reshape(1, pow_52, 3, 16, 32) + linear_20 = None + qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) + reshape_11 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_323 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) + q_5 = k_5 = v_5 = None + transpose_6 = x_323.transpose(1, 2) + x_323 = None + x_324 = transpose_6.reshape(1, pow_52, 512) + transpose_6 = None + x_325 = torch._C._nn.linear( + x_324, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_324 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_326 = torch.nn.functional.dropout(x_325, 0.0, False, False) + x_325 = None + x_327 = mul_45 + x_326 + mul_45 = x_326 = None + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_46 = x_327 * view_11 + view_11 = None + layer_norm_44 = torch.nn.functional.layer_norm( + x_327, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_327 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) + x_328 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_44 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_35 = torch.nn.functional.relu(x_328, inplace=False) + x_328 = None + pow_36 = relu_35**2 + relu_35 = None + mul_47 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + * pow_36 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( + pow_36 + ) = None + x_329 = ( + mul_47 + + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + ) + mul_47 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) + x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) + x_329 = None + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_332 = torch.nn.functional.dropout(x_331, 0.0, False, False) + x_331 = None + x_333 = mul_46 + x_332 + mul_46 = x_332 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_48 = x_333 * view_12 + view_12 = None + layer_norm_45 = torch.nn.functional.layer_norm( + x_333, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) + linear_24 = torch._C._nn.linear( + layer_norm_45, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_45 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_13 = linear_24.reshape(1, pow_52, 3, 16, 32) + linear_24 = None + qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) + reshape_13 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_334 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) + q_6 = k_6 = v_6 = None + transpose_7 = x_334.transpose(1, 2) + x_334 = None + x_335 = transpose_7.reshape(1, pow_52, 512) + transpose_7 = None + x_336 = torch._C._nn.linear( + x_335, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_335 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_337 = torch.nn.functional.dropout(x_336, 0.0, False, False) + x_336 = None + x_338 = mul_48 + x_337 + mul_48 = x_337 = None + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_49 = x_338 * view_13 + view_13 = None + layer_norm_46 = torch.nn.functional.layer_norm( + x_338, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_338 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) + x_339 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_46 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_36 = torch.nn.functional.relu(x_339, inplace=False) + x_339 = None + pow_37 = relu_36**2 + relu_36 = None + mul_50 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + * pow_37 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( + pow_37 + ) = None + x_340 = ( + mul_50 + + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + ) + mul_50 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) + x_341 = torch.nn.functional.dropout(x_340, 0.0, False, False) + x_340 = None + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_343 = torch.nn.functional.dropout(x_342, 0.0, False, False) + x_342 = None + x_344 = mul_49 + x_343 + mul_49 = x_343 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_51 = x_344 * view_14 + view_14 = None + layer_norm_47 = torch.nn.functional.layer_norm( + x_344, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) + linear_28 = torch._C._nn.linear( + layer_norm_47, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_47 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_15 = linear_28.reshape(1, pow_52, 3, 16, 32) + linear_28 = None + qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) + reshape_15 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_345 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) + q_7 = k_7 = v_7 = None + transpose_8 = x_345.transpose(1, 2) + x_345 = None + x_346 = transpose_8.reshape(1, pow_52, 512) + transpose_8 = None + x_347 = torch._C._nn.linear( + x_346, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_346 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) + x_347 = None + x_349 = mul_51 + x_348 + mul_51 = x_348 = None + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_52 = x_349 * view_15 + view_15 = None + layer_norm_48 = torch.nn.functional.layer_norm( + x_349, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_349 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) + x_350 = torch._C._nn.linear( + layer_norm_48, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_48 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_37 = torch.nn.functional.relu(x_350, inplace=False) + x_350 = None + pow_38 = relu_37**2 + relu_37 = None + mul_53 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + * pow_38 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( + pow_38 + ) = None + x_351 = ( + mul_53 + + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + ) + mul_53 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) + x_352 = torch.nn.functional.dropout(x_351, 0.0, False, False) + x_351 = None + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_354 = torch.nn.functional.dropout(x_353, 0.0, False, False) + x_353 = None + x_355 = mul_52 + x_354 + mul_52 = x_354 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_54 = x_355 * view_16 + view_16 = None + layer_norm_49 = torch.nn.functional.layer_norm( + x_355, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) + linear_32 = torch._C._nn.linear( + layer_norm_49, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_49 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_17 = linear_32.reshape(1, pow_52, 3, 16, 32) + linear_32 = None + qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) + reshape_17 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_356 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) + q_8 = k_8 = v_8 = None + transpose_9 = x_356.transpose(1, 2) + x_356 = None + x_357 = transpose_9.reshape(1, pow_52, 512) + transpose_9 = None + x_358 = torch._C._nn.linear( + x_357, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_357 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_359 = torch.nn.functional.dropout(x_358, 0.0, False, False) + x_358 = None + x_360 = mul_54 + x_359 + mul_54 = x_359 = None + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_55 = x_360 * view_17 + view_17 = None + layer_norm_50 = torch.nn.functional.layer_norm( + x_360, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_360 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) + x_361 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_50 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_38 = torch.nn.functional.relu(x_361, inplace=False) + x_361 = None + pow_39 = relu_38**2 + relu_38 = None + mul_56 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + * pow_39 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( + pow_39 + ) = None + x_362 = ( + mul_56 + + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + ) + mul_56 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) + x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) + x_362 = None + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_365 = torch.nn.functional.dropout(x_364, 0.0, False, False) + x_364 = None + x_366 = mul_55 + x_365 + mul_55 = x_365 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_57 = x_366 * view_18 + view_18 = None + layer_norm_51 = torch.nn.functional.layer_norm( + x_366, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_51, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_51 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_19 = linear_36.reshape(1, pow_52, 3, 16, 32) + linear_36 = None + qkv_9 = reshape_19.permute(2, 0, 3, 1, 4) + reshape_19 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_367 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) + q_9 = k_9 = v_9 = None + transpose_10 = x_367.transpose(1, 2) + x_367 = None + x_368 = transpose_10.reshape(1, pow_52, 512) + transpose_10 = None + x_369 = torch._C._nn.linear( + x_368, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_368 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_370 = torch.nn.functional.dropout(x_369, 0.0, False, False) + x_369 = None + x_371 = mul_57 + x_370 + mul_57 = x_370 = None + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_58 = x_371 * view_19 + view_19 = None + layer_norm_52 = torch.nn.functional.layer_norm( + x_371, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_371 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) + x_372 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_52 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_39 = torch.nn.functional.relu(x_372, inplace=False) + x_372 = None + pow_40 = relu_39**2 + relu_39 = None + mul_59 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + * pow_40 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( + pow_40 + ) = None + x_373 = ( + mul_59 + + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + ) + mul_59 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) + x_374 = torch.nn.functional.dropout(x_373, 0.0, False, False) + x_373 = None + x_375 = torch._C._nn.linear( + x_374, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_374 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_376 = torch.nn.functional.dropout(x_375, 0.0, False, False) + x_375 = None + x_377 = mul_58 + x_376 + mul_58 = x_376 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_60 = x_377 * view_20 + view_20 = None + layer_norm_53 = torch.nn.functional.layer_norm( + x_377, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_377 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_53, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_53 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_21 = linear_40.reshape(1, pow_52, 3, 16, 32) + linear_40 = None + qkv_10 = reshape_21.permute(2, 0, 3, 1, 4) + reshape_21 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_378 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_378.transpose(1, 2) + x_378 = None + x_379 = transpose_11.reshape(1, pow_52, 512) + transpose_11 = None + x_380 = torch._C._nn.linear( + x_379, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_379 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_381 = torch.nn.functional.dropout(x_380, 0.0, False, False) + x_380 = None + x_382 = mul_60 + x_381 + mul_60 = x_381 = None + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_61 = x_382 * view_21 + view_21 = None + layer_norm_54 = torch.nn.functional.layer_norm( + x_382, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_382 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) + x_383 = torch._C._nn.linear( + layer_norm_54, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_54 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_40 = torch.nn.functional.relu(x_383, inplace=False) + x_383 = None + pow_41 = relu_40**2 + relu_40 = None + mul_62 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + * pow_41 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( + pow_41 + ) = None + x_384 = ( + mul_62 + + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + ) + mul_62 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) + x_385 = torch.nn.functional.dropout(x_384, 0.0, False, False) + x_384 = None + x_386 = torch._C._nn.linear( + x_385, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_385 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_387 = torch.nn.functional.dropout(x_386, 0.0, False, False) + x_386 = None + x_388 = mul_61 + x_387 + mul_61 = x_387 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_63 = x_388 * view_22 + view_22 = None + layer_norm_55 = torch.nn.functional.layer_norm( + x_388, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_388 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_55, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_55 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_23 = linear_44.reshape(1, pow_52, 3, 16, 32) + linear_44 = None + qkv_11 = reshape_23.permute(2, 0, 3, 1, 4) + reshape_23 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_389 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_389.transpose(1, 2) + x_389 = None + x_390 = transpose_12.reshape(1, pow_52, 512) + transpose_12 = None + x_391 = torch._C._nn.linear( + x_390, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_390 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_392 = torch.nn.functional.dropout(x_391, 0.0, False, False) + x_391 = None + x_393 = mul_63 + x_392 + mul_63 = x_392 = None + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_64 = x_393 * view_23 + view_23 = None + layer_norm_56 = torch.nn.functional.layer_norm( + x_393, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_393 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) + x_394 = torch._C._nn.linear( + layer_norm_56, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_56 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_41 = torch.nn.functional.relu(x_394, inplace=False) + x_394 = None + pow_42 = relu_41**2 + relu_41 = None + mul_65 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + * pow_42 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( + pow_42 + ) = None + x_395 = ( + mul_65 + + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + ) + mul_65 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) + x_396 = torch.nn.functional.dropout(x_395, 0.0, False, False) + x_395 = None + x_397 = torch._C._nn.linear( + x_396, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_396 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_398 = torch.nn.functional.dropout(x_397, 0.0, False, False) + x_397 = None + x_399 = mul_64 + x_398 + mul_64 = x_398 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_66 = x_399 * view_24 + view_24 = None + layer_norm_57 = torch.nn.functional.layer_norm( + x_399, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_399 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_57, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_57 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_25 = linear_48.reshape(1, pow_52, 3, 16, 32) + linear_48 = None + qkv_12 = reshape_25.permute(2, 0, 3, 1, 4) + reshape_25 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_12 = unbind_12[0] + k_12 = unbind_12[1] + v_12 = unbind_12[2] + unbind_12 = None + x_400 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_13 = x_400.transpose(1, 2) + x_400 = None + x_401 = transpose_13.reshape(1, pow_52, 512) + transpose_13 = None + x_402 = torch._C._nn.linear( + x_401, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_401 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_403 = torch.nn.functional.dropout(x_402, 0.0, False, False) + x_402 = None + x_404 = mul_66 + x_403 + mul_66 = x_403 = None + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_67 = x_404 * view_25 + view_25 = None + layer_norm_58 = torch.nn.functional.layer_norm( + x_404, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_404 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = (None) + x_405 = torch._C._nn.linear( + layer_norm_58, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_58 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_42 = torch.nn.functional.relu(x_405, inplace=False) + x_405 = None + pow_43 = relu_42**2 + relu_42 = None + mul_68 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ + * pow_43 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = ( + pow_43 + ) = None + x_406 = ( + mul_68 + + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ + ) + mul_68 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = (None) + x_407 = torch.nn.functional.dropout(x_406, 0.0, False, False) + x_406 = None + x_408 = torch._C._nn.linear( + x_407, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_407 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_409 = torch.nn.functional.dropout(x_408, 0.0, False, False) + x_408 = None + x_410 = mul_67 + x_409 + mul_67 = x_409 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_69 = x_410 * view_26 + view_26 = None + layer_norm_59 = torch.nn.functional.layer_norm( + x_410, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_410 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = (None) + linear_52 = torch._C._nn.linear( + layer_norm_59, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_59 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_27 = linear_52.reshape(1, pow_52, 3, 16, 32) + linear_52 = None + qkv_13 = reshape_27.permute(2, 0, 3, 1, 4) + reshape_27 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_13 = unbind_13[0] + k_13 = unbind_13[1] + v_13 = unbind_13[2] + unbind_13 = None + x_411 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_14 = x_411.transpose(1, 2) + x_411 = None + x_412 = transpose_14.reshape(1, pow_52, 512) + transpose_14 = None + x_413 = torch._C._nn.linear( + x_412, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_412 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_414 = torch.nn.functional.dropout(x_413, 0.0, False, False) + x_413 = None + x_415 = mul_69 + x_414 + mul_69 = x_414 = None + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_70 = x_415 * view_27 + view_27 = None + layer_norm_60 = torch.nn.functional.layer_norm( + x_415, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_415 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = (None) + x_416 = torch._C._nn.linear( + layer_norm_60, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_60 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_43 = torch.nn.functional.relu(x_416, inplace=False) + x_416 = None + pow_44 = relu_43**2 + relu_43 = None + mul_71 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ + * pow_44 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = ( + pow_44 + ) = None + x_417 = ( + mul_71 + + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ + ) + mul_71 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = (None) + x_418 = torch.nn.functional.dropout(x_417, 0.0, False, False) + x_417 = None + x_419 = torch._C._nn.linear( + x_418, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_418 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_420 = torch.nn.functional.dropout(x_419, 0.0, False, False) + x_419 = None + x_421 = mul_70 + x_420 + mul_70 = x_420 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_72 = x_421 * view_28 + view_28 = None + layer_norm_61 = torch.nn.functional.layer_norm( + x_421, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_421 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_61, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_61 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_29 = linear_56.reshape(1, pow_52, 3, 16, 32) + linear_56 = None + qkv_14 = reshape_29.permute(2, 0, 3, 1, 4) + reshape_29 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_14 = unbind_14[0] + k_14 = unbind_14[1] + v_14 = unbind_14[2] + unbind_14 = None + x_422 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_15 = x_422.transpose(1, 2) + x_422 = None + x_423 = transpose_15.reshape(1, pow_52, 512) + transpose_15 = None + x_424 = torch._C._nn.linear( + x_423, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_423 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_425 = torch.nn.functional.dropout(x_424, 0.0, False, False) + x_424 = None + x_426 = mul_72 + x_425 + mul_72 = x_425 = None + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_73 = x_426 * view_29 + view_29 = None + layer_norm_62 = torch.nn.functional.layer_norm( + x_426, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_426 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = (None) + x_427 = torch._C._nn.linear( + layer_norm_62, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_62 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_44 = torch.nn.functional.relu(x_427, inplace=False) + x_427 = None + pow_45 = relu_44**2 + relu_44 = None + mul_74 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ + * pow_45 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = ( + pow_45 + ) = None + x_428 = ( + mul_74 + + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ + ) + mul_74 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = (None) + x_429 = torch.nn.functional.dropout(x_428, 0.0, False, False) + x_428 = None + x_430 = torch._C._nn.linear( + x_429, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_429 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_431 = torch.nn.functional.dropout(x_430, 0.0, False, False) + x_430 = None + x_432 = mul_73 + x_431 + mul_73 = x_431 = None + view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_75 = x_432 * view_30 + view_30 = None + layer_norm_63 = torch.nn.functional.layer_norm( + x_432, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_432 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = (None) + linear_60 = torch._C._nn.linear( + layer_norm_63, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_63 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_31 = linear_60.reshape(1, pow_52, 3, 16, 32) + linear_60 = None + qkv_15 = reshape_31.permute(2, 0, 3, 1, 4) + reshape_31 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_15 = unbind_15[0] + k_15 = unbind_15[1] + v_15 = unbind_15[2] + unbind_15 = None + x_433 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_16 = x_433.transpose(1, 2) + x_433 = None + x_434 = transpose_16.reshape(1, pow_52, 512) + transpose_16 = None + x_435 = torch._C._nn.linear( + x_434, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_434 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_436 = torch.nn.functional.dropout(x_435, 0.0, False, False) + x_435 = None + x_437 = mul_75 + x_436 + mul_75 = x_436 = None + view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_76 = x_437 * view_31 + view_31 = None + layer_norm_64 = torch.nn.functional.layer_norm( + x_437, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_437 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = (None) + x_438 = torch._C._nn.linear( + layer_norm_64, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_64 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_45 = torch.nn.functional.relu(x_438, inplace=False) + x_438 = None + pow_46 = relu_45**2 + relu_45 = None + mul_77 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ + * pow_46 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = ( + pow_46 + ) = None + x_439 = ( + mul_77 + + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ + ) + mul_77 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = (None) + x_440 = torch.nn.functional.dropout(x_439, 0.0, False, False) + x_439 = None + x_441 = torch._C._nn.linear( + x_440, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_440 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_442 = torch.nn.functional.dropout(x_441, 0.0, False, False) + x_441 = None + x_443 = mul_76 + x_442 + mul_76 = x_442 = None + view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_78 = x_443 * view_32 + view_32 = None + layer_norm_65 = torch.nn.functional.layer_norm( + x_443, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_443 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_65, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_65 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_33 = linear_64.reshape(1, pow_52, 3, 16, 32) + linear_64 = None + qkv_16 = reshape_33.permute(2, 0, 3, 1, 4) + reshape_33 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_16 = unbind_16[0] + k_16 = unbind_16[1] + v_16 = unbind_16[2] + unbind_16 = None + x_444 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_17 = x_444.transpose(1, 2) + x_444 = None + x_445 = transpose_17.reshape(1, pow_52, 512) + transpose_17 = None + x_446 = torch._C._nn.linear( + x_445, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_445 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_447 = torch.nn.functional.dropout(x_446, 0.0, False, False) + x_446 = None + x_448 = mul_78 + x_447 + mul_78 = x_447 = None + view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_79 = x_448 * view_33 + view_33 = None + layer_norm_66 = torch.nn.functional.layer_norm( + x_448, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_448 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = (None) + x_449 = torch._C._nn.linear( + layer_norm_66, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_66 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_46 = torch.nn.functional.relu(x_449, inplace=False) + x_449 = None + pow_47 = relu_46**2 + relu_46 = None + mul_80 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ + * pow_47 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = ( + pow_47 + ) = None + x_450 = ( + mul_80 + + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ + ) + mul_80 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = (None) + x_451 = torch.nn.functional.dropout(x_450, 0.0, False, False) + x_450 = None + x_452 = torch._C._nn.linear( + x_451, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_451 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_453 = torch.nn.functional.dropout(x_452, 0.0, False, False) + x_452 = None + x_454 = mul_79 + x_453 + mul_79 = x_453 = None + view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_81 = x_454 * view_34 + view_34 = None + layer_norm_67 = torch.nn.functional.layer_norm( + x_454, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_454 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = (None) + linear_68 = torch._C._nn.linear( + layer_norm_67, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_67 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_35 = linear_68.reshape(1, pow_52, 3, 16, 32) + linear_68 = None + qkv_17 = reshape_35.permute(2, 0, 3, 1, 4) + reshape_35 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_17 = unbind_17[0] + k_17 = unbind_17[1] + v_17 = unbind_17[2] + unbind_17 = None + x_455 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_18 = x_455.transpose(1, 2) + x_455 = None + x_456 = transpose_18.reshape(1, pow_52, 512) + transpose_18 = pow_52 = None + x_457 = torch._C._nn.linear( + x_456, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_456 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_458 = torch.nn.functional.dropout(x_457, 0.0, False, False) + x_457 = None + x_459 = mul_81 + x_458 + mul_81 = x_458 = None + view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_82 = x_459 * view_35 + view_35 = None + layer_norm_68 = torch.nn.functional.layer_norm( + x_459, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_459 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = (None) + x_460 = torch._C._nn.linear( + layer_norm_68, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_68 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_47 = torch.nn.functional.relu(x_460, inplace=False) + x_460 = None + pow_48 = relu_47**2 + relu_47 = None + mul_83 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ + * pow_48 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = ( + pow_48 + ) = None + x_461 = ( + mul_83 + + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ + ) + mul_83 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = (None) + x_462 = torch.nn.functional.dropout(x_461, 0.0, False, False) + x_461 = None + x_463 = torch._C._nn.linear( + x_462, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_462 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_464 = torch.nn.functional.dropout(x_463, 0.0, False, False) + x_463 = None + x_465 = mul_82 + x_464 + mul_82 = x_464 = None + transpose_19 = x_465.transpose(1, 2) + x_465 = None + x_466 = transpose_19.reshape(1, 512, sym_sum_3, sym_sum_3) + transpose_19 = sym_sum_3 = None + x_467 = x_466.permute(0, 2, 3, 1) + x_466 = None + x_468 = torch.nn.functional.layer_norm( + x_467, + (512,), + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_467 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) + x_469 = x_468.permute(0, 3, 1, 2) + x_468 = None + x_470 = torch.conv2d( + x_469, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_469 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_3 = sym_sum // 32 + sym_sum = None + sym_sum_4 = torch.sym_sum([1, floordiv_3]) + floordiv_3 = None + reshape_38 = x_470.reshape(1, 768, -1) + x_470 = None + x_471 = reshape_38.transpose(1, 2) + reshape_38 = None + view_36 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_84 = x_471 * view_36 + view_36 = None + layer_norm_70 = torch.nn.functional.layer_norm( + x_471, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_471 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_53 = sym_sum_4**2 + linear_72 = torch._C._nn.linear( + layer_norm_70, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_70 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_39 = linear_72.reshape(1, pow_53, 3, 24, 32) + linear_72 = None + qkv_18 = reshape_39.permute(2, 0, 3, 1, 4) + reshape_39 = None + unbind_18 = qkv_18.unbind(0) + qkv_18 = None + q_18 = unbind_18[0] + k_18 = unbind_18[1] + v_18 = unbind_18[2] + unbind_18 = None + x_472 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_21 = x_472.transpose(1, 2) + x_472 = None + x_473 = transpose_21.reshape(1, pow_53, 768) + transpose_21 = None + x_474 = torch._C._nn.linear( + x_473, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_473 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_475 = torch.nn.functional.dropout(x_474, 0.0, False, False) + x_474 = None + x_476 = mul_84 + x_475 + mul_84 = x_475 = None + view_37 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_85 = x_476 * view_37 + view_37 = None + layer_norm_71 = torch.nn.functional.layer_norm( + x_476, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_476 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_477 = torch._C._nn.linear( + layer_norm_71, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_71 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_48 = torch.nn.functional.relu(x_477, inplace=False) + x_477 = None + pow_49 = relu_48**2 + relu_48 = None + mul_86 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_49 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_49 + ) = None + x_478 = ( + mul_86 + + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_86 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_479 = torch.nn.functional.dropout(x_478, 0.0, False, False) + x_478 = None + x_480 = torch._C._nn.linear( + x_479, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_479 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_481 = torch.nn.functional.dropout(x_480, 0.0, False, False) + x_480 = None + x_482 = mul_85 + x_481 + mul_85 = x_481 = None + view_38 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_87 = x_482 * view_38 + view_38 = None + layer_norm_72 = torch.nn.functional.layer_norm( + x_482, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_482 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_76 = torch._C._nn.linear( + layer_norm_72, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_72 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_41 = linear_76.reshape(1, pow_53, 3, 24, 32) + linear_76 = None + qkv_19 = reshape_41.permute(2, 0, 3, 1, 4) + reshape_41 = None + unbind_19 = qkv_19.unbind(0) + qkv_19 = None + q_19 = unbind_19[0] + k_19 = unbind_19[1] + v_19 = unbind_19[2] + unbind_19 = None + x_483 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_22 = x_483.transpose(1, 2) + x_483 = None + x_484 = transpose_22.reshape(1, pow_53, 768) + transpose_22 = None + x_485 = torch._C._nn.linear( + x_484, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_484 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_486 = torch.nn.functional.dropout(x_485, 0.0, False, False) + x_485 = None + x_487 = mul_87 + x_486 + mul_87 = x_486 = None + view_39 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_88 = x_487 * view_39 + view_39 = None + layer_norm_73 = torch.nn.functional.layer_norm( + x_487, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_487 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_488 = torch._C._nn.linear( + layer_norm_73, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_73 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_49 = torch.nn.functional.relu(x_488, inplace=False) + x_488 = None + pow_50 = relu_49**2 + relu_49 = None + mul_89 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_50 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_50 + ) = None + x_489 = ( + mul_89 + + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_89 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_490 = torch.nn.functional.dropout(x_489, 0.0, False, False) + x_489 = None + x_491 = torch._C._nn.linear( + x_490, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_490 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_492 = torch.nn.functional.dropout(x_491, 0.0, False, False) + x_491 = None + x_493 = mul_88 + x_492 + mul_88 = x_492 = None + view_40 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_90 = x_493 * view_40 + view_40 = None + layer_norm_74 = torch.nn.functional.layer_norm( + x_493, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_493 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_74, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_74 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_43 = linear_80.reshape(1, pow_53, 3, 24, 32) + linear_80 = None + qkv_20 = reshape_43.permute(2, 0, 3, 1, 4) + reshape_43 = None + unbind_20 = qkv_20.unbind(0) + qkv_20 = None + q_20 = unbind_20[0] + k_20 = unbind_20[1] + v_20 = unbind_20[2] + unbind_20 = None + x_494 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_23 = x_494.transpose(1, 2) + x_494 = None + x_495 = transpose_23.reshape(1, pow_53, 768) + transpose_23 = pow_53 = None + x_496 = torch._C._nn.linear( + x_495, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_495 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_497 = torch.nn.functional.dropout(x_496, 0.0, False, False) + x_496 = None + x_498 = mul_90 + x_497 + mul_90 = x_497 = None + view_41 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_91 = x_498 * view_41 + view_41 = None + layer_norm_75 = torch.nn.functional.layer_norm( + x_498, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_498 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_499 = torch._C._nn.linear( + layer_norm_75, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_75 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_50 = torch.nn.functional.relu(x_499, inplace=False) + x_499 = None + pow_51 = relu_50**2 + relu_50 = None + mul_92 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_51 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_51 + ) = None + x_500 = ( + mul_92 + + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_92 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_501 = torch.nn.functional.dropout(x_500, 0.0, False, False) + x_500 = None + x_502 = torch._C._nn.linear( + x_501, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_501 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_503 = torch.nn.functional.dropout(x_502, 0.0, False, False) + x_502 = None + x_504 = mul_91 + x_503 + mul_91 = x_503 = None + transpose_24 = x_504.transpose(1, 2) + x_504 = None + x_505 = transpose_24.reshape(1, 768, sym_sum_4, sym_sum_4) + transpose_24 = sym_sum_4 = None + x_506 = torch.nn.functional.adaptive_avg_pool2d(x_505, 1) + x_505 = None + x_507 = x_506.permute(0, 2, 3, 1) + x_506 = None + x_508 = torch.nn.functional.layer_norm( + x_507, + (768,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_507 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_509 = x_508.permute(0, 3, 1, 2) + x_508 = None + x_510 = x_509.flatten(1, -1) + x_509 = None + x_511 = torch.nn.functional.dropout(x_510, 0.0, False, False) + x_510 = None + x_512 = torch._C._nn.linear( + x_511, + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, + ) + x_511 = ( + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None + relu_51 = torch.nn.functional.relu(x_512, inplace=False) + x_512 = None + x_513 = torch.square(relu_51) + relu_51 = None + x_514 = torch.nn.functional.layer_norm( + x_513, + (3072,), + l_self_modules_head_modules_fc_modules_norm_parameters_weight_, + l_self_modules_head_modules_fc_modules_norm_parameters_bias_, + 1e-06, + ) + x_513 = ( + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None + x_515 = torch.nn.functional.dropout(x_514, 0.0, False, False) + x_514 = None + x_516 = torch._C._nn.linear( + x_515, + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, + ) + x_515 = ( + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None + return (x_516,) diff --git a/samples/timm/caformer_b36.sail_in22k/weight_meta.py b/samples/timm/caformer_b36.sail_in22k/weight_meta.py new file mode 100644 index 0000000000..d7c23666a9 --- /dev/null +++ b/samples/timm/caformer_b36.sail_in22k/weight_meta.py @@ -0,0 +1,3968 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv_parameters_weight_" + shape = [128, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: + name = "L_self_modules_stem_modules_conv_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.227 + std = 1.283 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: + name = "L_self_modules_stem_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" + shape = [768, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" + shape = [21841, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" + shape = [21841] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/caformer_b36.sail_in22k_ft_in1k/graph_net.json b/samples/timm/caformer_b36.sail_in22k_ft_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/caformer_b36.sail_in22k_ft_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/caformer_b36.sail_in22k_ft_in1k/input_meta.py b/samples/timm/caformer_b36.sail_in22k_ft_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_b36.sail_in22k_ft_in1k/input_tensor_constraints.py b/samples/timm/caformer_b36.sail_in22k_ft_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_b36.sail_in22k_ft_in1k/model.py b/samples/timm/caformer_b36.sail_in22k_ft_in1k/model.py new file mode 100644 index 0000000000..08d3de35cc --- /dev/null +++ b/samples/timm/caformer_b36.sail_in22k_ft_in1k/model.py @@ -0,0 +1,4758 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv_parameters_bias_ = ( + L_self_modules_stem_modules_conv_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_norm_parameters_weight_ = ( + L_self_modules_stem_modules_norm_parameters_weight_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv_parameters_weight_, + l_self_modules_stem_modules_conv_parameters_bias_, + (4, 4), + (2, 2), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_conv_parameters_weight_ + ) = l_self_modules_stem_modules_conv_parameters_bias_ = None + x_1 = x.permute(0, 2, 3, 1) + x = None + x_2 = torch.nn.functional.layer_norm( + x_1, + (128,), + l_self_modules_stem_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None + x_3 = x_2.permute(0, 3, 1, 2) + x_2 = None + sym_sum = torch.sym_sum([-3, s1]) + s1 = None + floordiv = sym_sum // 4 + sym_sum_1 = torch.sym_sum([1, floordiv]) + floordiv = sym_sum_1 = None + x_4 = x_3.permute(0, 2, 3, 1) + x_5 = torch.nn.functional.layer_norm( + x_4, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu = torch.nn.functional.relu(x_7, inplace=False) + x_7 = None + pow_1 = relu**2 + relu = None + mul = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_1 + ) = None + x_8 = ( + mul + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_11 = x_3 + x_10 + x_3 = x_10 = None + x_12 = x_11.permute(0, 2, 3, 1) + x_13 = torch.nn.functional.layer_norm( + x_12, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_14 = x_13.permute(0, 3, 1, 2) + x_13 = None + x_15 = torch.conv2d( + x_14, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_1 = torch.nn.functional.relu(x_15, inplace=False) + x_15 = None + pow_2 = relu_1**2 + relu_1 = None + mul_1 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_2 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_2 + ) = None + x_16 = ( + mul_1 + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = x_11 + x_19 + x_11 = x_19 = None + x_21 = x_20.permute(0, 2, 3, 1) + x_22 = torch.nn.functional.layer_norm( + x_21, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = torch.conv2d( + x_23, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_2 = torch.nn.functional.relu(x_24, inplace=False) + x_24 = None + pow_3 = relu_2**2 + relu_2 = None + mul_2 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_3 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_3 + ) = None + x_25 = ( + mul_2 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_27 = torch.conv2d( + x_26, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_28 = x_20 + x_27 + x_20 = x_27 = None + x_29 = x_28.permute(0, 2, 3, 1) + x_30 = torch.nn.functional.layer_norm( + x_29, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_3 = torch.nn.functional.relu(x_32, inplace=False) + x_32 = None + pow_4 = relu_3**2 + relu_3 = None + mul_3 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_4 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_4 + ) = None + x_33 = ( + mul_3 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + x_37 = x_28 + x_36 + x_28 = x_36 = None + x_38 = x_37.permute(0, 2, 3, 1) + x_39 = torch.nn.functional.layer_norm( + x_38, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_40 = x_39.permute(0, 3, 1, 2) + x_39 = None + x_41 = torch.conv2d( + x_40, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_4 = torch.nn.functional.relu(x_41, inplace=False) + x_41 = None + pow_5 = relu_4**2 + relu_4 = None + mul_4 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_5 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_5 + ) = None + x_42 = ( + mul_4 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_44 = torch.conv2d( + x_43, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_45 = x_37 + x_44 + x_37 = x_44 = None + x_46 = x_45.permute(0, 2, 3, 1) + x_47 = torch.nn.functional.layer_norm( + x_46, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_5 = torch.nn.functional.relu(x_49, inplace=False) + x_49 = None + pow_6 = relu_5**2 + relu_5 = None + mul_5 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_6 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_6 + ) = None + x_50 = ( + mul_5 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = x_45 + x_53 + x_45 = x_53 = None + x_55 = x_54.permute(0, 2, 3, 1) + x_54 = None + x_56 = torch.nn.functional.layer_norm( + x_55, + (128,), + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) + x_57 = x_56.permute(0, 3, 1, 2) + x_56 = None + x_58 = torch.conv2d( + x_57, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_1 = sym_sum // 8 + sym_sum_2 = torch.sym_sum([1, floordiv_1]) + floordiv_1 = sym_sum_2 = None + x_59 = x_58.permute(0, 2, 3, 1) + x_60 = torch.nn.functional.layer_norm( + x_59, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_61 = x_60.permute(0, 3, 1, 2) + x_60 = None + x_62 = torch.conv2d( + x_61, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_6 = torch.nn.functional.relu(x_62, inplace=False) + x_62 = None + pow_7 = relu_6**2 + relu_6 = None + mul_6 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_7 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_7 + ) = None + x_63 = ( + mul_6 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_66 = x_58 + x_65 + x_58 = x_65 = None + x_67 = x_66.permute(0, 2, 3, 1) + x_68 = torch.nn.functional.layer_norm( + x_67, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + x_70 = torch.conv2d( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_7 = torch.nn.functional.relu(x_70, inplace=False) + x_70 = None + pow_8 = relu_7**2 + relu_7 = None + mul_7 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_8 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_8 + ) = None + x_71 = ( + mul_7 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + x_73 = torch.conv2d( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + x_75 = x_66 + x_74 + x_66 = x_74 = None + x_76 = x_75.permute(0, 2, 3, 1) + x_77 = torch.nn.functional.layer_norm( + x_76, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_8 = torch.nn.functional.relu(x_79, inplace=False) + x_79 = None + pow_9 = relu_8**2 + relu_8 = None + mul_8 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_9 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_9 + ) = None + x_80 = ( + mul_8 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_81 = torch.conv2d( + x_80, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_83 = x_75 + x_82 + x_75 = x_82 = None + x_84 = x_83.permute(0, 2, 3, 1) + x_85 = torch.nn.functional.layer_norm( + x_84, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_86 = x_85.permute(0, 3, 1, 2) + x_85 = None + x_87 = torch.conv2d( + x_86, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_9 = torch.nn.functional.relu(x_87, inplace=False) + x_87 = None + pow_10 = relu_9**2 + relu_9 = None + mul_9 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_10 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_10 + ) = None + x_88 = ( + mul_9 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = x_83 + x_91 + x_83 = x_91 = None + x_93 = x_92.permute(0, 2, 3, 1) + x_94 = torch.nn.functional.layer_norm( + x_93, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = torch.conv2d( + x_95, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_10 = torch.nn.functional.relu(x_96, inplace=False) + x_96 = None + pow_11 = relu_10**2 + relu_10 = None + mul_10 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_11 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_11 + ) = None + x_97 = ( + mul_10 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_98 = torch.conv2d( + x_97, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_99 = torch.conv2d( + x_98, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_100 = x_92 + x_99 + x_92 = x_99 = None + x_101 = x_100.permute(0, 2, 3, 1) + x_102 = torch.nn.functional.layer_norm( + x_101, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_103 = x_102.permute(0, 3, 1, 2) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_11 = torch.nn.functional.relu(x_104, inplace=False) + x_104 = None + pow_12 = relu_11**2 + relu_11 = None + mul_11 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_12 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_12 + ) = None + x_105 = ( + mul_11 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_100 + x_108 + x_100 = x_108 = None + x_110 = x_109.permute(0, 2, 3, 1) + x_111 = torch.nn.functional.layer_norm( + x_110, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_110 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) + x_112 = x_111.permute(0, 3, 1, 2) + x_111 = None + x_113 = torch.conv2d( + x_112, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_112 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_12 = torch.nn.functional.relu(x_113, inplace=False) + x_113 = None + pow_13 = relu_12**2 + relu_12 = None + mul_12 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ + * pow_13 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_13 + ) = None + x_114 = ( + mul_12 + + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_12 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_115 = torch.conv2d( + x_114, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_114 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_117 = x_109 + x_116 + x_109 = x_116 = None + x_118 = x_117.permute(0, 2, 3, 1) + x_119 = torch.nn.functional.layer_norm( + x_118, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_118 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) + x_120 = x_119.permute(0, 3, 1, 2) + x_119 = None + x_121 = torch.conv2d( + x_120, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_120 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_13 = torch.nn.functional.relu(x_121, inplace=False) + x_121 = None + pow_14 = relu_13**2 + relu_13 = None + mul_13 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + * pow_14 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( + pow_14 + ) = None + x_122 = ( + mul_13 + + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + ) + mul_13 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) + x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) + x_122 = None + x_124 = torch.conv2d( + x_123, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_123 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_126 = x_117 + x_125 + x_117 = x_125 = None + x_127 = x_126.permute(0, 2, 3, 1) + x_128 = torch.nn.functional.layer_norm( + x_127, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_127 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) + x_129 = x_128.permute(0, 3, 1, 2) + x_128 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_14 = torch.nn.functional.relu(x_130, inplace=False) + x_130 = None + pow_15 = relu_14**2 + relu_14 = None + mul_14 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ + * pow_15 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_15 + ) = None + x_131 = ( + mul_14 + + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_14 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_132 = torch.conv2d( + x_131, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_131 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_132 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_134 = x_126 + x_133 + x_126 = x_133 = None + x_135 = x_134.permute(0, 2, 3, 1) + x_136 = torch.nn.functional.layer_norm( + x_135, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_135 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) + x_137 = x_136.permute(0, 3, 1, 2) + x_136 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_15 = torch.nn.functional.relu(x_138, inplace=False) + x_138 = None + pow_16 = relu_15**2 + relu_15 = None + mul_15 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + * pow_16 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( + pow_16 + ) = None + x_139 = ( + mul_15 + + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + ) + mul_15 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) + x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) + x_139 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_140 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_134 + x_142 + x_134 = x_142 = None + x_144 = x_143.permute(0, 2, 3, 1) + x_145 = torch.nn.functional.layer_norm( + x_144, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_144 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) + x_146 = x_145.permute(0, 3, 1, 2) + x_145 = None + x_147 = torch.conv2d( + x_146, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_146 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_16 = torch.nn.functional.relu(x_147, inplace=False) + x_147 = None + pow_17 = relu_16**2 + relu_16 = None + mul_16 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ + * pow_17 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_17 + ) = None + x_148 = ( + mul_16 + + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_16 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_149 = torch.conv2d( + x_148, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_148 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_150 = torch.conv2d( + x_149, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_149 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_151 = x_143 + x_150 + x_143 = x_150 = None + x_152 = x_151.permute(0, 2, 3, 1) + x_153 = torch.nn.functional.layer_norm( + x_152, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_152 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) + x_154 = x_153.permute(0, 3, 1, 2) + x_153 = None + x_155 = torch.conv2d( + x_154, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_154 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_17 = torch.nn.functional.relu(x_155, inplace=False) + x_155 = None + pow_18 = relu_17**2 + relu_17 = None + mul_17 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + * pow_18 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( + pow_18 + ) = None + x_156 = ( + mul_17 + + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + ) + mul_17 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch.conv2d( + x_157, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_157 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_151 + x_159 + x_151 = x_159 = None + x_161 = x_160.permute(0, 2, 3, 1) + x_162 = torch.nn.functional.layer_norm( + x_161, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_161 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) + x_163 = x_162.permute(0, 3, 1, 2) + x_162 = None + x_164 = torch.conv2d( + x_163, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_163 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_18 = torch.nn.functional.relu(x_164, inplace=False) + x_164 = None + pow_19 = relu_18**2 + relu_18 = None + mul_18 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ + * pow_19 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_19 + ) = None + x_165 = ( + mul_18 + + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_18 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_166 = torch.conv2d( + x_165, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_165 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_167 = torch.conv2d( + x_166, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_166 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_168 = x_160 + x_167 + x_160 = x_167 = None + x_169 = x_168.permute(0, 2, 3, 1) + x_170 = torch.nn.functional.layer_norm( + x_169, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_169 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) + x_171 = x_170.permute(0, 3, 1, 2) + x_170 = None + x_172 = torch.conv2d( + x_171, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_171 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_19 = torch.nn.functional.relu(x_172, inplace=False) + x_172 = None + pow_20 = relu_19**2 + relu_19 = None + mul_19 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + * pow_20 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( + pow_20 + ) = None + x_173 = ( + mul_19 + + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + ) + mul_19 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) + x_174 = torch.nn.functional.dropout(x_173, 0.0, False, False) + x_173 = None + x_175 = torch.conv2d( + x_174, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_174 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = x_168 + x_176 + x_168 = x_176 = None + x_178 = x_177.permute(0, 2, 3, 1) + x_179 = torch.nn.functional.layer_norm( + x_178, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_178 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) + x_180 = x_179.permute(0, 3, 1, 2) + x_179 = None + x_181 = torch.conv2d( + x_180, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_180 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_20 = torch.nn.functional.relu(x_181, inplace=False) + x_181 = None + pow_21 = relu_20**2 + relu_20 = None + mul_20 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ + * pow_21 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_21 + ) = None + x_182 = ( + mul_20 + + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_20 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_183 = torch.conv2d( + x_182, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_182 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_184 = torch.conv2d( + x_183, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_183 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_185 = x_177 + x_184 + x_177 = x_184 = None + x_186 = x_185.permute(0, 2, 3, 1) + x_187 = torch.nn.functional.layer_norm( + x_186, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) + x_188 = x_187.permute(0, 3, 1, 2) + x_187 = None + x_189 = torch.conv2d( + x_188, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_188 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_21 = torch.nn.functional.relu(x_189, inplace=False) + x_189 = None + pow_22 = relu_21**2 + relu_21 = None + mul_21 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + * pow_22 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( + pow_22 + ) = None + x_190 = ( + mul_21 + + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + ) + mul_21 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) + x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) + x_190 = None + x_192 = torch.conv2d( + x_191, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_191 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_185 + x_193 + x_185 = x_193 = None + x_195 = x_194.permute(0, 2, 3, 1) + x_196 = torch.nn.functional.layer_norm( + x_195, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_195 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) + x_197 = x_196.permute(0, 3, 1, 2) + x_196 = None + x_198 = torch.conv2d( + x_197, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_197 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_22 = torch.nn.functional.relu(x_198, inplace=False) + x_198 = None + pow_23 = relu_22**2 + relu_22 = None + mul_22 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ + * pow_23 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_23 + ) = None + x_199 = ( + mul_22 + + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_22 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_200 = torch.conv2d( + x_199, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_199 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_201 = torch.conv2d( + x_200, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_200 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_202 = x_194 + x_201 + x_194 = x_201 = None + x_203 = x_202.permute(0, 2, 3, 1) + x_204 = torch.nn.functional.layer_norm( + x_203, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_203 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) + x_205 = x_204.permute(0, 3, 1, 2) + x_204 = None + x_206 = torch.conv2d( + x_205, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_205 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_23 = torch.nn.functional.relu(x_206, inplace=False) + x_206 = None + pow_24 = relu_23**2 + relu_23 = None + mul_23 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + * pow_24 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( + pow_24 + ) = None + x_207 = ( + mul_23 + + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + ) + mul_23 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) + x_208 = torch.nn.functional.dropout(x_207, 0.0, False, False) + x_207 = None + x_209 = torch.conv2d( + x_208, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_208 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) + x_209 = None + x_211 = x_202 + x_210 + x_202 = x_210 = None + x_212 = x_211.permute(0, 2, 3, 1) + x_213 = torch.nn.functional.layer_norm( + x_212, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_212 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) + x_214 = x_213.permute(0, 3, 1, 2) + x_213 = None + x_215 = torch.conv2d( + x_214, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_214 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_24 = torch.nn.functional.relu(x_215, inplace=False) + x_215 = None + pow_25 = relu_24**2 + relu_24 = None + mul_24 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ + * pow_25 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_25 + ) = None + x_216 = ( + mul_24 + + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_24 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_217 = torch.conv2d( + x_216, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_216 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_217 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_219 = x_211 + x_218 + x_211 = x_218 = None + x_220 = x_219.permute(0, 2, 3, 1) + x_221 = torch.nn.functional.layer_norm( + x_220, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_220 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) + x_222 = x_221.permute(0, 3, 1, 2) + x_221 = None + x_223 = torch.conv2d( + x_222, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_222 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_25 = torch.nn.functional.relu(x_223, inplace=False) + x_223 = None + pow_26 = relu_25**2 + relu_25 = None + mul_25 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + * pow_26 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( + pow_26 + ) = None + x_224 = ( + mul_25 + + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + ) + mul_25 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = torch.conv2d( + x_225, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_225 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) + x_226 = None + x_228 = x_219 + x_227 + x_219 = x_227 = None + x_229 = x_228.permute(0, 2, 3, 1) + x_230 = torch.nn.functional.layer_norm( + x_229, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_229 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) + x_231 = x_230.permute(0, 3, 1, 2) + x_230 = None + x_232 = torch.conv2d( + x_231, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_231 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_26 = torch.nn.functional.relu(x_232, inplace=False) + x_232 = None + pow_27 = relu_26**2 + relu_26 = None + mul_26 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ + * pow_27 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_27 + ) = None + x_233 = ( + mul_26 + + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_26 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_234 = torch.conv2d( + x_233, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_233 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_235 = torch.conv2d( + x_234, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_234 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_236 = x_228 + x_235 + x_228 = x_235 = None + x_237 = x_236.permute(0, 2, 3, 1) + x_238 = torch.nn.functional.layer_norm( + x_237, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_237 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) + x_239 = x_238.permute(0, 3, 1, 2) + x_238 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_239 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_27 = torch.nn.functional.relu(x_240, inplace=False) + x_240 = None + pow_28 = relu_27**2 + relu_27 = None + mul_27 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + * pow_28 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( + pow_28 + ) = None + x_241 = ( + mul_27 + + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + ) + mul_27 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) + x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) + x_241 = None + x_243 = torch.conv2d( + x_242, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_242 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_244 = torch.nn.functional.dropout(x_243, 0.0, False, False) + x_243 = None + x_245 = x_236 + x_244 + x_236 = x_244 = None + x_246 = x_245.permute(0, 2, 3, 1) + x_247 = torch.nn.functional.layer_norm( + x_246, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_246 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) + x_248 = x_247.permute(0, 3, 1, 2) + x_247 = None + x_249 = torch.conv2d( + x_248, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_248 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_28 = torch.nn.functional.relu(x_249, inplace=False) + x_249 = None + pow_29 = relu_28**2 + relu_28 = None + mul_28 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ + * pow_29 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_29 + ) = None + x_250 = ( + mul_28 + + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_28 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_250 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_252 = torch.conv2d( + x_251, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_251 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_253 = x_245 + x_252 + x_245 = x_252 = None + x_254 = x_253.permute(0, 2, 3, 1) + x_255 = torch.nn.functional.layer_norm( + x_254, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_254 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) + x_256 = x_255.permute(0, 3, 1, 2) + x_255 = None + x_257 = torch.conv2d( + x_256, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_256 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_29 = torch.nn.functional.relu(x_257, inplace=False) + x_257 = None + pow_30 = relu_29**2 + relu_29 = None + mul_29 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + * pow_30 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( + pow_30 + ) = None + x_258 = ( + mul_29 + + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + ) + mul_29 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = torch.conv2d( + x_259, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_259 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_261 = torch.nn.functional.dropout(x_260, 0.0, False, False) + x_260 = None + x_262 = x_253 + x_261 + x_253 = x_261 = None + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (256,), + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) + x_265 = x_264.permute(0, 3, 1, 2) + x_264 = None + x_266 = torch.conv2d( + x_265, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_265 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_2 = sym_sum // 16 + sym_sum_3 = torch.sym_sum([1, floordiv_2]) + floordiv_2 = None + reshape = x_266.reshape(1, 512, -1) + x_266 = None + x_267 = reshape.transpose(1, 2) + reshape = None + view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_30 = x_267 * view + view = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_267, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_52 = sym_sum_3**2 + linear = torch._C._nn.linear( + layer_norm_33, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_33 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_1 = linear.reshape(1, pow_52, 3, 16, 32) + linear = None + qkv = reshape_1.permute(2, 0, 3, 1, 4) + reshape_1 = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_268 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) + q = k = v = None + transpose_1 = x_268.transpose(1, 2) + x_268 = None + x_269 = transpose_1.reshape(1, pow_52, 512) + transpose_1 = None + x_270 = torch._C._nn.linear( + x_269, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_269 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_271 = torch.nn.functional.dropout(x_270, 0.0, False, False) + x_270 = None + x_272 = mul_30 + x_271 + mul_30 = x_271 = None + view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_31 = x_272 * view_1 + view_1 = None + layer_norm_34 = torch.nn.functional.layer_norm( + x_272, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_272 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_273 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_34 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_30 = torch.nn.functional.relu(x_273, inplace=False) + x_273 = None + pow_31 = relu_30**2 + relu_30 = None + mul_32 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_31 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_31 + ) = None + x_274 = ( + mul_32 + + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_275 = torch.nn.functional.dropout(x_274, 0.0, False, False) + x_274 = None + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_277 = torch.nn.functional.dropout(x_276, 0.0, False, False) + x_276 = None + x_278 = mul_31 + x_277 + mul_31 = x_277 = None + view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_33 = x_278 * view_2 + view_2 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_278, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_3 = linear_4.reshape(1, pow_52, 3, 16, 32) + linear_4 = None + qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) + reshape_3 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_279 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) + q_1 = k_1 = v_1 = None + transpose_2 = x_279.transpose(1, 2) + x_279 = None + x_280 = transpose_2.reshape(1, pow_52, 512) + transpose_2 = None + x_281 = torch._C._nn.linear( + x_280, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_280 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) + x_281 = None + x_283 = mul_33 + x_282 + mul_33 = x_282 = None + view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_34 = x_283 * view_3 + view_3 = None + layer_norm_36 = torch.nn.functional.layer_norm( + x_283, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_283 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_284 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_36 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_31 = torch.nn.functional.relu(x_284, inplace=False) + x_284 = None + pow_32 = relu_31**2 + relu_31 = None + mul_35 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_32 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_32 + ) = None + x_285 = ( + mul_35 + + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_286 = torch.nn.functional.dropout(x_285, 0.0, False, False) + x_285 = None + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) + x_287 = None + x_289 = mul_34 + x_288 + mul_34 = x_288 = None + view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_36 = x_289 * view_4 + view_4 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_289, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_37 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_5 = linear_8.reshape(1, pow_52, 3, 16, 32) + linear_8 = None + qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) + reshape_5 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_290 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) + q_2 = k_2 = v_2 = None + transpose_3 = x_290.transpose(1, 2) + x_290 = None + x_291 = transpose_3.reshape(1, pow_52, 512) + transpose_3 = None + x_292 = torch._C._nn.linear( + x_291, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_291 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_293 = torch.nn.functional.dropout(x_292, 0.0, False, False) + x_292 = None + x_294 = mul_36 + x_293 + mul_36 = x_293 = None + view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_37 = x_294 * view_5 + view_5 = None + layer_norm_38 = torch.nn.functional.layer_norm( + x_294, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_294 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_295 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_32 = torch.nn.functional.relu(x_295, inplace=False) + x_295 = None + pow_33 = relu_32**2 + relu_32 = None + mul_38 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_33 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_33 + ) = None + x_296 = ( + mul_38 + + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_297 = torch.nn.functional.dropout(x_296, 0.0, False, False) + x_296 = None + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_299 = torch.nn.functional.dropout(x_298, 0.0, False, False) + x_298 = None + x_300 = mul_37 + x_299 + mul_37 = x_299 = None + view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_39 = x_300 * view_6 + view_6 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_300, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_39 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_7 = linear_12.reshape(1, pow_52, 3, 16, 32) + linear_12 = None + qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) + reshape_7 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_301 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) + q_3 = k_3 = v_3 = None + transpose_4 = x_301.transpose(1, 2) + x_301 = None + x_302 = transpose_4.reshape(1, pow_52, 512) + transpose_4 = None + x_303 = torch._C._nn.linear( + x_302, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_302 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_304 = torch.nn.functional.dropout(x_303, 0.0, False, False) + x_303 = None + x_305 = mul_39 + x_304 + mul_39 = x_304 = None + view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_40 = x_305 * view_7 + view_7 = None + layer_norm_40 = torch.nn.functional.layer_norm( + x_305, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_305 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) + x_306 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_40 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_33 = torch.nn.functional.relu(x_306, inplace=False) + x_306 = None + pow_34 = relu_33**2 + relu_33 = None + mul_41 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + * pow_34 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( + pow_34 + ) = None + x_307 = ( + mul_41 + + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + ) + mul_41 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) + x_308 = torch.nn.functional.dropout(x_307, 0.0, False, False) + x_307 = None + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) + x_309 = None + x_311 = mul_40 + x_310 + mul_40 = x_310 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_42 = x_311 * view_8 + view_8 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_311, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_41 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_9 = linear_16.reshape(1, pow_52, 3, 16, 32) + linear_16 = None + qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) + reshape_9 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_312 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) + q_4 = k_4 = v_4 = None + transpose_5 = x_312.transpose(1, 2) + x_312 = None + x_313 = transpose_5.reshape(1, pow_52, 512) + transpose_5 = None + x_314 = torch._C._nn.linear( + x_313, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_313 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) + x_314 = None + x_316 = mul_42 + x_315 + mul_42 = x_315 = None + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_43 = x_316 * view_9 + view_9 = None + layer_norm_42 = torch.nn.functional.layer_norm( + x_316, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_316 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) + x_317 = torch._C._nn.linear( + layer_norm_42, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_42 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_34 = torch.nn.functional.relu(x_317, inplace=False) + x_317 = None + pow_35 = relu_34**2 + relu_34 = None + mul_44 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + * pow_35 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( + pow_35 + ) = None + x_318 = ( + mul_44 + + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + ) + mul_44 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) + x_319 = torch.nn.functional.dropout(x_318, 0.0, False, False) + x_318 = None + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_321 = torch.nn.functional.dropout(x_320, 0.0, False, False) + x_320 = None + x_322 = mul_43 + x_321 + mul_43 = x_321 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_45 = x_322 * view_10 + view_10 = None + layer_norm_43 = torch.nn.functional.layer_norm( + x_322, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) + linear_20 = torch._C._nn.linear( + layer_norm_43, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_43 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_11 = linear_20.reshape(1, pow_52, 3, 16, 32) + linear_20 = None + qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) + reshape_11 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_323 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) + q_5 = k_5 = v_5 = None + transpose_6 = x_323.transpose(1, 2) + x_323 = None + x_324 = transpose_6.reshape(1, pow_52, 512) + transpose_6 = None + x_325 = torch._C._nn.linear( + x_324, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_324 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_326 = torch.nn.functional.dropout(x_325, 0.0, False, False) + x_325 = None + x_327 = mul_45 + x_326 + mul_45 = x_326 = None + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_46 = x_327 * view_11 + view_11 = None + layer_norm_44 = torch.nn.functional.layer_norm( + x_327, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_327 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) + x_328 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_44 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_35 = torch.nn.functional.relu(x_328, inplace=False) + x_328 = None + pow_36 = relu_35**2 + relu_35 = None + mul_47 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + * pow_36 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( + pow_36 + ) = None + x_329 = ( + mul_47 + + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + ) + mul_47 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) + x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) + x_329 = None + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_332 = torch.nn.functional.dropout(x_331, 0.0, False, False) + x_331 = None + x_333 = mul_46 + x_332 + mul_46 = x_332 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_48 = x_333 * view_12 + view_12 = None + layer_norm_45 = torch.nn.functional.layer_norm( + x_333, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) + linear_24 = torch._C._nn.linear( + layer_norm_45, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_45 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_13 = linear_24.reshape(1, pow_52, 3, 16, 32) + linear_24 = None + qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) + reshape_13 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_334 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) + q_6 = k_6 = v_6 = None + transpose_7 = x_334.transpose(1, 2) + x_334 = None + x_335 = transpose_7.reshape(1, pow_52, 512) + transpose_7 = None + x_336 = torch._C._nn.linear( + x_335, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_335 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_337 = torch.nn.functional.dropout(x_336, 0.0, False, False) + x_336 = None + x_338 = mul_48 + x_337 + mul_48 = x_337 = None + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_49 = x_338 * view_13 + view_13 = None + layer_norm_46 = torch.nn.functional.layer_norm( + x_338, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_338 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) + x_339 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_46 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_36 = torch.nn.functional.relu(x_339, inplace=False) + x_339 = None + pow_37 = relu_36**2 + relu_36 = None + mul_50 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + * pow_37 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( + pow_37 + ) = None + x_340 = ( + mul_50 + + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + ) + mul_50 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) + x_341 = torch.nn.functional.dropout(x_340, 0.0, False, False) + x_340 = None + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_343 = torch.nn.functional.dropout(x_342, 0.0, False, False) + x_342 = None + x_344 = mul_49 + x_343 + mul_49 = x_343 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_51 = x_344 * view_14 + view_14 = None + layer_norm_47 = torch.nn.functional.layer_norm( + x_344, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) + linear_28 = torch._C._nn.linear( + layer_norm_47, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_47 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_15 = linear_28.reshape(1, pow_52, 3, 16, 32) + linear_28 = None + qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) + reshape_15 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_345 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) + q_7 = k_7 = v_7 = None + transpose_8 = x_345.transpose(1, 2) + x_345 = None + x_346 = transpose_8.reshape(1, pow_52, 512) + transpose_8 = None + x_347 = torch._C._nn.linear( + x_346, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_346 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) + x_347 = None + x_349 = mul_51 + x_348 + mul_51 = x_348 = None + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_52 = x_349 * view_15 + view_15 = None + layer_norm_48 = torch.nn.functional.layer_norm( + x_349, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_349 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) + x_350 = torch._C._nn.linear( + layer_norm_48, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_48 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_37 = torch.nn.functional.relu(x_350, inplace=False) + x_350 = None + pow_38 = relu_37**2 + relu_37 = None + mul_53 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + * pow_38 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( + pow_38 + ) = None + x_351 = ( + mul_53 + + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + ) + mul_53 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) + x_352 = torch.nn.functional.dropout(x_351, 0.0, False, False) + x_351 = None + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_354 = torch.nn.functional.dropout(x_353, 0.0, False, False) + x_353 = None + x_355 = mul_52 + x_354 + mul_52 = x_354 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_54 = x_355 * view_16 + view_16 = None + layer_norm_49 = torch.nn.functional.layer_norm( + x_355, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) + linear_32 = torch._C._nn.linear( + layer_norm_49, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_49 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_17 = linear_32.reshape(1, pow_52, 3, 16, 32) + linear_32 = None + qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) + reshape_17 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_356 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) + q_8 = k_8 = v_8 = None + transpose_9 = x_356.transpose(1, 2) + x_356 = None + x_357 = transpose_9.reshape(1, pow_52, 512) + transpose_9 = None + x_358 = torch._C._nn.linear( + x_357, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_357 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_359 = torch.nn.functional.dropout(x_358, 0.0, False, False) + x_358 = None + x_360 = mul_54 + x_359 + mul_54 = x_359 = None + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_55 = x_360 * view_17 + view_17 = None + layer_norm_50 = torch.nn.functional.layer_norm( + x_360, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_360 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) + x_361 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_50 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_38 = torch.nn.functional.relu(x_361, inplace=False) + x_361 = None + pow_39 = relu_38**2 + relu_38 = None + mul_56 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + * pow_39 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( + pow_39 + ) = None + x_362 = ( + mul_56 + + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + ) + mul_56 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) + x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) + x_362 = None + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_365 = torch.nn.functional.dropout(x_364, 0.0, False, False) + x_364 = None + x_366 = mul_55 + x_365 + mul_55 = x_365 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_57 = x_366 * view_18 + view_18 = None + layer_norm_51 = torch.nn.functional.layer_norm( + x_366, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_51, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_51 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_19 = linear_36.reshape(1, pow_52, 3, 16, 32) + linear_36 = None + qkv_9 = reshape_19.permute(2, 0, 3, 1, 4) + reshape_19 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_367 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) + q_9 = k_9 = v_9 = None + transpose_10 = x_367.transpose(1, 2) + x_367 = None + x_368 = transpose_10.reshape(1, pow_52, 512) + transpose_10 = None + x_369 = torch._C._nn.linear( + x_368, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_368 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_370 = torch.nn.functional.dropout(x_369, 0.0, False, False) + x_369 = None + x_371 = mul_57 + x_370 + mul_57 = x_370 = None + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_58 = x_371 * view_19 + view_19 = None + layer_norm_52 = torch.nn.functional.layer_norm( + x_371, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_371 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) + x_372 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_52 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_39 = torch.nn.functional.relu(x_372, inplace=False) + x_372 = None + pow_40 = relu_39**2 + relu_39 = None + mul_59 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + * pow_40 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( + pow_40 + ) = None + x_373 = ( + mul_59 + + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + ) + mul_59 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) + x_374 = torch.nn.functional.dropout(x_373, 0.0, False, False) + x_373 = None + x_375 = torch._C._nn.linear( + x_374, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_374 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_376 = torch.nn.functional.dropout(x_375, 0.0, False, False) + x_375 = None + x_377 = mul_58 + x_376 + mul_58 = x_376 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_60 = x_377 * view_20 + view_20 = None + layer_norm_53 = torch.nn.functional.layer_norm( + x_377, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_377 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_53, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_53 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_21 = linear_40.reshape(1, pow_52, 3, 16, 32) + linear_40 = None + qkv_10 = reshape_21.permute(2, 0, 3, 1, 4) + reshape_21 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_378 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_378.transpose(1, 2) + x_378 = None + x_379 = transpose_11.reshape(1, pow_52, 512) + transpose_11 = None + x_380 = torch._C._nn.linear( + x_379, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_379 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_381 = torch.nn.functional.dropout(x_380, 0.0, False, False) + x_380 = None + x_382 = mul_60 + x_381 + mul_60 = x_381 = None + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_61 = x_382 * view_21 + view_21 = None + layer_norm_54 = torch.nn.functional.layer_norm( + x_382, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_382 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) + x_383 = torch._C._nn.linear( + layer_norm_54, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_54 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_40 = torch.nn.functional.relu(x_383, inplace=False) + x_383 = None + pow_41 = relu_40**2 + relu_40 = None + mul_62 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + * pow_41 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( + pow_41 + ) = None + x_384 = ( + mul_62 + + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + ) + mul_62 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) + x_385 = torch.nn.functional.dropout(x_384, 0.0, False, False) + x_384 = None + x_386 = torch._C._nn.linear( + x_385, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_385 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_387 = torch.nn.functional.dropout(x_386, 0.0, False, False) + x_386 = None + x_388 = mul_61 + x_387 + mul_61 = x_387 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_63 = x_388 * view_22 + view_22 = None + layer_norm_55 = torch.nn.functional.layer_norm( + x_388, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_388 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_55, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_55 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_23 = linear_44.reshape(1, pow_52, 3, 16, 32) + linear_44 = None + qkv_11 = reshape_23.permute(2, 0, 3, 1, 4) + reshape_23 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_389 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_389.transpose(1, 2) + x_389 = None + x_390 = transpose_12.reshape(1, pow_52, 512) + transpose_12 = None + x_391 = torch._C._nn.linear( + x_390, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_390 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_392 = torch.nn.functional.dropout(x_391, 0.0, False, False) + x_391 = None + x_393 = mul_63 + x_392 + mul_63 = x_392 = None + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_64 = x_393 * view_23 + view_23 = None + layer_norm_56 = torch.nn.functional.layer_norm( + x_393, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_393 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) + x_394 = torch._C._nn.linear( + layer_norm_56, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_56 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_41 = torch.nn.functional.relu(x_394, inplace=False) + x_394 = None + pow_42 = relu_41**2 + relu_41 = None + mul_65 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + * pow_42 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( + pow_42 + ) = None + x_395 = ( + mul_65 + + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + ) + mul_65 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) + x_396 = torch.nn.functional.dropout(x_395, 0.0, False, False) + x_395 = None + x_397 = torch._C._nn.linear( + x_396, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_396 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_398 = torch.nn.functional.dropout(x_397, 0.0, False, False) + x_397 = None + x_399 = mul_64 + x_398 + mul_64 = x_398 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_66 = x_399 * view_24 + view_24 = None + layer_norm_57 = torch.nn.functional.layer_norm( + x_399, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_399 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_57, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_57 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_25 = linear_48.reshape(1, pow_52, 3, 16, 32) + linear_48 = None + qkv_12 = reshape_25.permute(2, 0, 3, 1, 4) + reshape_25 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_12 = unbind_12[0] + k_12 = unbind_12[1] + v_12 = unbind_12[2] + unbind_12 = None + x_400 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_13 = x_400.transpose(1, 2) + x_400 = None + x_401 = transpose_13.reshape(1, pow_52, 512) + transpose_13 = None + x_402 = torch._C._nn.linear( + x_401, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_401 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_403 = torch.nn.functional.dropout(x_402, 0.0, False, False) + x_402 = None + x_404 = mul_66 + x_403 + mul_66 = x_403 = None + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_67 = x_404 * view_25 + view_25 = None + layer_norm_58 = torch.nn.functional.layer_norm( + x_404, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_404 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = (None) + x_405 = torch._C._nn.linear( + layer_norm_58, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_58 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_42 = torch.nn.functional.relu(x_405, inplace=False) + x_405 = None + pow_43 = relu_42**2 + relu_42 = None + mul_68 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ + * pow_43 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = ( + pow_43 + ) = None + x_406 = ( + mul_68 + + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ + ) + mul_68 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = (None) + x_407 = torch.nn.functional.dropout(x_406, 0.0, False, False) + x_406 = None + x_408 = torch._C._nn.linear( + x_407, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_407 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_409 = torch.nn.functional.dropout(x_408, 0.0, False, False) + x_408 = None + x_410 = mul_67 + x_409 + mul_67 = x_409 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_69 = x_410 * view_26 + view_26 = None + layer_norm_59 = torch.nn.functional.layer_norm( + x_410, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_410 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = (None) + linear_52 = torch._C._nn.linear( + layer_norm_59, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_59 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_27 = linear_52.reshape(1, pow_52, 3, 16, 32) + linear_52 = None + qkv_13 = reshape_27.permute(2, 0, 3, 1, 4) + reshape_27 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_13 = unbind_13[0] + k_13 = unbind_13[1] + v_13 = unbind_13[2] + unbind_13 = None + x_411 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_14 = x_411.transpose(1, 2) + x_411 = None + x_412 = transpose_14.reshape(1, pow_52, 512) + transpose_14 = None + x_413 = torch._C._nn.linear( + x_412, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_412 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_414 = torch.nn.functional.dropout(x_413, 0.0, False, False) + x_413 = None + x_415 = mul_69 + x_414 + mul_69 = x_414 = None + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_70 = x_415 * view_27 + view_27 = None + layer_norm_60 = torch.nn.functional.layer_norm( + x_415, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_415 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = (None) + x_416 = torch._C._nn.linear( + layer_norm_60, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_60 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_43 = torch.nn.functional.relu(x_416, inplace=False) + x_416 = None + pow_44 = relu_43**2 + relu_43 = None + mul_71 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ + * pow_44 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = ( + pow_44 + ) = None + x_417 = ( + mul_71 + + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ + ) + mul_71 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = (None) + x_418 = torch.nn.functional.dropout(x_417, 0.0, False, False) + x_417 = None + x_419 = torch._C._nn.linear( + x_418, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_418 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_420 = torch.nn.functional.dropout(x_419, 0.0, False, False) + x_419 = None + x_421 = mul_70 + x_420 + mul_70 = x_420 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_72 = x_421 * view_28 + view_28 = None + layer_norm_61 = torch.nn.functional.layer_norm( + x_421, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_421 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_61, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_61 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_29 = linear_56.reshape(1, pow_52, 3, 16, 32) + linear_56 = None + qkv_14 = reshape_29.permute(2, 0, 3, 1, 4) + reshape_29 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_14 = unbind_14[0] + k_14 = unbind_14[1] + v_14 = unbind_14[2] + unbind_14 = None + x_422 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_15 = x_422.transpose(1, 2) + x_422 = None + x_423 = transpose_15.reshape(1, pow_52, 512) + transpose_15 = None + x_424 = torch._C._nn.linear( + x_423, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_423 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_425 = torch.nn.functional.dropout(x_424, 0.0, False, False) + x_424 = None + x_426 = mul_72 + x_425 + mul_72 = x_425 = None + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_73 = x_426 * view_29 + view_29 = None + layer_norm_62 = torch.nn.functional.layer_norm( + x_426, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_426 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = (None) + x_427 = torch._C._nn.linear( + layer_norm_62, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_62 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_44 = torch.nn.functional.relu(x_427, inplace=False) + x_427 = None + pow_45 = relu_44**2 + relu_44 = None + mul_74 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ + * pow_45 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = ( + pow_45 + ) = None + x_428 = ( + mul_74 + + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ + ) + mul_74 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = (None) + x_429 = torch.nn.functional.dropout(x_428, 0.0, False, False) + x_428 = None + x_430 = torch._C._nn.linear( + x_429, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_429 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_431 = torch.nn.functional.dropout(x_430, 0.0, False, False) + x_430 = None + x_432 = mul_73 + x_431 + mul_73 = x_431 = None + view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_75 = x_432 * view_30 + view_30 = None + layer_norm_63 = torch.nn.functional.layer_norm( + x_432, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_432 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = (None) + linear_60 = torch._C._nn.linear( + layer_norm_63, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_63 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_31 = linear_60.reshape(1, pow_52, 3, 16, 32) + linear_60 = None + qkv_15 = reshape_31.permute(2, 0, 3, 1, 4) + reshape_31 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_15 = unbind_15[0] + k_15 = unbind_15[1] + v_15 = unbind_15[2] + unbind_15 = None + x_433 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_16 = x_433.transpose(1, 2) + x_433 = None + x_434 = transpose_16.reshape(1, pow_52, 512) + transpose_16 = None + x_435 = torch._C._nn.linear( + x_434, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_434 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_436 = torch.nn.functional.dropout(x_435, 0.0, False, False) + x_435 = None + x_437 = mul_75 + x_436 + mul_75 = x_436 = None + view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_76 = x_437 * view_31 + view_31 = None + layer_norm_64 = torch.nn.functional.layer_norm( + x_437, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_437 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = (None) + x_438 = torch._C._nn.linear( + layer_norm_64, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_64 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_45 = torch.nn.functional.relu(x_438, inplace=False) + x_438 = None + pow_46 = relu_45**2 + relu_45 = None + mul_77 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ + * pow_46 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = ( + pow_46 + ) = None + x_439 = ( + mul_77 + + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ + ) + mul_77 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = (None) + x_440 = torch.nn.functional.dropout(x_439, 0.0, False, False) + x_439 = None + x_441 = torch._C._nn.linear( + x_440, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_440 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_442 = torch.nn.functional.dropout(x_441, 0.0, False, False) + x_441 = None + x_443 = mul_76 + x_442 + mul_76 = x_442 = None + view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_78 = x_443 * view_32 + view_32 = None + layer_norm_65 = torch.nn.functional.layer_norm( + x_443, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_443 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_65, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_65 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_33 = linear_64.reshape(1, pow_52, 3, 16, 32) + linear_64 = None + qkv_16 = reshape_33.permute(2, 0, 3, 1, 4) + reshape_33 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_16 = unbind_16[0] + k_16 = unbind_16[1] + v_16 = unbind_16[2] + unbind_16 = None + x_444 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_17 = x_444.transpose(1, 2) + x_444 = None + x_445 = transpose_17.reshape(1, pow_52, 512) + transpose_17 = None + x_446 = torch._C._nn.linear( + x_445, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_445 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_447 = torch.nn.functional.dropout(x_446, 0.0, False, False) + x_446 = None + x_448 = mul_78 + x_447 + mul_78 = x_447 = None + view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_79 = x_448 * view_33 + view_33 = None + layer_norm_66 = torch.nn.functional.layer_norm( + x_448, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_448 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = (None) + x_449 = torch._C._nn.linear( + layer_norm_66, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_66 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_46 = torch.nn.functional.relu(x_449, inplace=False) + x_449 = None + pow_47 = relu_46**2 + relu_46 = None + mul_80 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ + * pow_47 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = ( + pow_47 + ) = None + x_450 = ( + mul_80 + + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ + ) + mul_80 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = (None) + x_451 = torch.nn.functional.dropout(x_450, 0.0, False, False) + x_450 = None + x_452 = torch._C._nn.linear( + x_451, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_451 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_453 = torch.nn.functional.dropout(x_452, 0.0, False, False) + x_452 = None + x_454 = mul_79 + x_453 + mul_79 = x_453 = None + view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_81 = x_454 * view_34 + view_34 = None + layer_norm_67 = torch.nn.functional.layer_norm( + x_454, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_454 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = (None) + linear_68 = torch._C._nn.linear( + layer_norm_67, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_67 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_35 = linear_68.reshape(1, pow_52, 3, 16, 32) + linear_68 = None + qkv_17 = reshape_35.permute(2, 0, 3, 1, 4) + reshape_35 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_17 = unbind_17[0] + k_17 = unbind_17[1] + v_17 = unbind_17[2] + unbind_17 = None + x_455 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_18 = x_455.transpose(1, 2) + x_455 = None + x_456 = transpose_18.reshape(1, pow_52, 512) + transpose_18 = pow_52 = None + x_457 = torch._C._nn.linear( + x_456, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_456 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_458 = torch.nn.functional.dropout(x_457, 0.0, False, False) + x_457 = None + x_459 = mul_81 + x_458 + mul_81 = x_458 = None + view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_82 = x_459 * view_35 + view_35 = None + layer_norm_68 = torch.nn.functional.layer_norm( + x_459, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_459 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = (None) + x_460 = torch._C._nn.linear( + layer_norm_68, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_68 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_47 = torch.nn.functional.relu(x_460, inplace=False) + x_460 = None + pow_48 = relu_47**2 + relu_47 = None + mul_83 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ + * pow_48 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = ( + pow_48 + ) = None + x_461 = ( + mul_83 + + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ + ) + mul_83 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = (None) + x_462 = torch.nn.functional.dropout(x_461, 0.0, False, False) + x_461 = None + x_463 = torch._C._nn.linear( + x_462, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_462 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_464 = torch.nn.functional.dropout(x_463, 0.0, False, False) + x_463 = None + x_465 = mul_82 + x_464 + mul_82 = x_464 = None + transpose_19 = x_465.transpose(1, 2) + x_465 = None + x_466 = transpose_19.reshape(1, 512, sym_sum_3, sym_sum_3) + transpose_19 = sym_sum_3 = None + x_467 = x_466.permute(0, 2, 3, 1) + x_466 = None + x_468 = torch.nn.functional.layer_norm( + x_467, + (512,), + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_467 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) + x_469 = x_468.permute(0, 3, 1, 2) + x_468 = None + x_470 = torch.conv2d( + x_469, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_469 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_3 = sym_sum // 32 + sym_sum = None + sym_sum_4 = torch.sym_sum([1, floordiv_3]) + floordiv_3 = None + reshape_38 = x_470.reshape(1, 768, -1) + x_470 = None + x_471 = reshape_38.transpose(1, 2) + reshape_38 = None + view_36 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_84 = x_471 * view_36 + view_36 = None + layer_norm_70 = torch.nn.functional.layer_norm( + x_471, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_471 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_53 = sym_sum_4**2 + linear_72 = torch._C._nn.linear( + layer_norm_70, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_70 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_39 = linear_72.reshape(1, pow_53, 3, 24, 32) + linear_72 = None + qkv_18 = reshape_39.permute(2, 0, 3, 1, 4) + reshape_39 = None + unbind_18 = qkv_18.unbind(0) + qkv_18 = None + q_18 = unbind_18[0] + k_18 = unbind_18[1] + v_18 = unbind_18[2] + unbind_18 = None + x_472 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_21 = x_472.transpose(1, 2) + x_472 = None + x_473 = transpose_21.reshape(1, pow_53, 768) + transpose_21 = None + x_474 = torch._C._nn.linear( + x_473, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_473 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_475 = torch.nn.functional.dropout(x_474, 0.0, False, False) + x_474 = None + x_476 = mul_84 + x_475 + mul_84 = x_475 = None + view_37 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_85 = x_476 * view_37 + view_37 = None + layer_norm_71 = torch.nn.functional.layer_norm( + x_476, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_476 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_477 = torch._C._nn.linear( + layer_norm_71, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_71 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_48 = torch.nn.functional.relu(x_477, inplace=False) + x_477 = None + pow_49 = relu_48**2 + relu_48 = None + mul_86 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_49 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_49 + ) = None + x_478 = ( + mul_86 + + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_86 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_479 = torch.nn.functional.dropout(x_478, 0.0, False, False) + x_478 = None + x_480 = torch._C._nn.linear( + x_479, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_479 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_481 = torch.nn.functional.dropout(x_480, 0.0, False, False) + x_480 = None + x_482 = mul_85 + x_481 + mul_85 = x_481 = None + view_38 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_87 = x_482 * view_38 + view_38 = None + layer_norm_72 = torch.nn.functional.layer_norm( + x_482, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_482 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_76 = torch._C._nn.linear( + layer_norm_72, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_72 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_41 = linear_76.reshape(1, pow_53, 3, 24, 32) + linear_76 = None + qkv_19 = reshape_41.permute(2, 0, 3, 1, 4) + reshape_41 = None + unbind_19 = qkv_19.unbind(0) + qkv_19 = None + q_19 = unbind_19[0] + k_19 = unbind_19[1] + v_19 = unbind_19[2] + unbind_19 = None + x_483 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_22 = x_483.transpose(1, 2) + x_483 = None + x_484 = transpose_22.reshape(1, pow_53, 768) + transpose_22 = None + x_485 = torch._C._nn.linear( + x_484, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_484 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_486 = torch.nn.functional.dropout(x_485, 0.0, False, False) + x_485 = None + x_487 = mul_87 + x_486 + mul_87 = x_486 = None + view_39 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_88 = x_487 * view_39 + view_39 = None + layer_norm_73 = torch.nn.functional.layer_norm( + x_487, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_487 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_488 = torch._C._nn.linear( + layer_norm_73, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_73 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_49 = torch.nn.functional.relu(x_488, inplace=False) + x_488 = None + pow_50 = relu_49**2 + relu_49 = None + mul_89 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_50 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_50 + ) = None + x_489 = ( + mul_89 + + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_89 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_490 = torch.nn.functional.dropout(x_489, 0.0, False, False) + x_489 = None + x_491 = torch._C._nn.linear( + x_490, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_490 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_492 = torch.nn.functional.dropout(x_491, 0.0, False, False) + x_491 = None + x_493 = mul_88 + x_492 + mul_88 = x_492 = None + view_40 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_90 = x_493 * view_40 + view_40 = None + layer_norm_74 = torch.nn.functional.layer_norm( + x_493, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_493 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_74, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_74 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_43 = linear_80.reshape(1, pow_53, 3, 24, 32) + linear_80 = None + qkv_20 = reshape_43.permute(2, 0, 3, 1, 4) + reshape_43 = None + unbind_20 = qkv_20.unbind(0) + qkv_20 = None + q_20 = unbind_20[0] + k_20 = unbind_20[1] + v_20 = unbind_20[2] + unbind_20 = None + x_494 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_23 = x_494.transpose(1, 2) + x_494 = None + x_495 = transpose_23.reshape(1, pow_53, 768) + transpose_23 = pow_53 = None + x_496 = torch._C._nn.linear( + x_495, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_495 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_497 = torch.nn.functional.dropout(x_496, 0.0, False, False) + x_496 = None + x_498 = mul_90 + x_497 + mul_90 = x_497 = None + view_41 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_91 = x_498 * view_41 + view_41 = None + layer_norm_75 = torch.nn.functional.layer_norm( + x_498, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_498 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_499 = torch._C._nn.linear( + layer_norm_75, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_75 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_50 = torch.nn.functional.relu(x_499, inplace=False) + x_499 = None + pow_51 = relu_50**2 + relu_50 = None + mul_92 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_51 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_51 + ) = None + x_500 = ( + mul_92 + + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_92 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_501 = torch.nn.functional.dropout(x_500, 0.0, False, False) + x_500 = None + x_502 = torch._C._nn.linear( + x_501, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_501 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_503 = torch.nn.functional.dropout(x_502, 0.0, False, False) + x_502 = None + x_504 = mul_91 + x_503 + mul_91 = x_503 = None + transpose_24 = x_504.transpose(1, 2) + x_504 = None + x_505 = transpose_24.reshape(1, 768, sym_sum_4, sym_sum_4) + transpose_24 = sym_sum_4 = None + x_506 = torch.nn.functional.adaptive_avg_pool2d(x_505, 1) + x_505 = None + x_507 = x_506.permute(0, 2, 3, 1) + x_506 = None + x_508 = torch.nn.functional.layer_norm( + x_507, + (768,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_507 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_509 = x_508.permute(0, 3, 1, 2) + x_508 = None + x_510 = x_509.flatten(1, -1) + x_509 = None + x_511 = torch.nn.functional.dropout(x_510, 0.0, False, False) + x_510 = None + x_512 = torch._C._nn.linear( + x_511, + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, + ) + x_511 = ( + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None + relu_51 = torch.nn.functional.relu(x_512, inplace=False) + x_512 = None + x_513 = torch.square(relu_51) + relu_51 = None + x_514 = torch.nn.functional.layer_norm( + x_513, + (3072,), + l_self_modules_head_modules_fc_modules_norm_parameters_weight_, + l_self_modules_head_modules_fc_modules_norm_parameters_bias_, + 1e-06, + ) + x_513 = ( + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None + x_515 = torch.nn.functional.dropout(x_514, 0.0, False, False) + x_514 = None + x_516 = torch._C._nn.linear( + x_515, + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, + ) + x_515 = ( + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None + return (x_516,) diff --git a/samples/timm/caformer_b36.sail_in22k_ft_in1k/weight_meta.py b/samples/timm/caformer_b36.sail_in22k_ft_in1k/weight_meta.py new file mode 100644 index 0000000000..279c30e841 --- /dev/null +++ b/samples/timm/caformer_b36.sail_in22k_ft_in1k/weight_meta.py @@ -0,0 +1,3968 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv_parameters_weight_" + shape = [128, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: + name = "L_self_modules_stem_modules_conv_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.231 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: + name = "L_self_modules_stem_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" + shape = [768, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" + shape = [1000, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/graph_net.json b/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/input_meta.py b/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/input_tensor_constraints.py b/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/model.py b/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/model.py new file mode 100644 index 0000000000..08d3de35cc --- /dev/null +++ b/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/model.py @@ -0,0 +1,4758 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv_parameters_bias_ = ( + L_self_modules_stem_modules_conv_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_norm_parameters_weight_ = ( + L_self_modules_stem_modules_norm_parameters_weight_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv_parameters_weight_, + l_self_modules_stem_modules_conv_parameters_bias_, + (4, 4), + (2, 2), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_conv_parameters_weight_ + ) = l_self_modules_stem_modules_conv_parameters_bias_ = None + x_1 = x.permute(0, 2, 3, 1) + x = None + x_2 = torch.nn.functional.layer_norm( + x_1, + (128,), + l_self_modules_stem_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None + x_3 = x_2.permute(0, 3, 1, 2) + x_2 = None + sym_sum = torch.sym_sum([-3, s1]) + s1 = None + floordiv = sym_sum // 4 + sym_sum_1 = torch.sym_sum([1, floordiv]) + floordiv = sym_sum_1 = None + x_4 = x_3.permute(0, 2, 3, 1) + x_5 = torch.nn.functional.layer_norm( + x_4, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu = torch.nn.functional.relu(x_7, inplace=False) + x_7 = None + pow_1 = relu**2 + relu = None + mul = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_1 + ) = None + x_8 = ( + mul + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_11 = x_3 + x_10 + x_3 = x_10 = None + x_12 = x_11.permute(0, 2, 3, 1) + x_13 = torch.nn.functional.layer_norm( + x_12, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_14 = x_13.permute(0, 3, 1, 2) + x_13 = None + x_15 = torch.conv2d( + x_14, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_1 = torch.nn.functional.relu(x_15, inplace=False) + x_15 = None + pow_2 = relu_1**2 + relu_1 = None + mul_1 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_2 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_2 + ) = None + x_16 = ( + mul_1 + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = x_11 + x_19 + x_11 = x_19 = None + x_21 = x_20.permute(0, 2, 3, 1) + x_22 = torch.nn.functional.layer_norm( + x_21, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = torch.conv2d( + x_23, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_2 = torch.nn.functional.relu(x_24, inplace=False) + x_24 = None + pow_3 = relu_2**2 + relu_2 = None + mul_2 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_3 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_3 + ) = None + x_25 = ( + mul_2 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_27 = torch.conv2d( + x_26, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_28 = x_20 + x_27 + x_20 = x_27 = None + x_29 = x_28.permute(0, 2, 3, 1) + x_30 = torch.nn.functional.layer_norm( + x_29, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_3 = torch.nn.functional.relu(x_32, inplace=False) + x_32 = None + pow_4 = relu_3**2 + relu_3 = None + mul_3 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_4 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_4 + ) = None + x_33 = ( + mul_3 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + x_37 = x_28 + x_36 + x_28 = x_36 = None + x_38 = x_37.permute(0, 2, 3, 1) + x_39 = torch.nn.functional.layer_norm( + x_38, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_40 = x_39.permute(0, 3, 1, 2) + x_39 = None + x_41 = torch.conv2d( + x_40, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_4 = torch.nn.functional.relu(x_41, inplace=False) + x_41 = None + pow_5 = relu_4**2 + relu_4 = None + mul_4 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_5 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_5 + ) = None + x_42 = ( + mul_4 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_44 = torch.conv2d( + x_43, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_45 = x_37 + x_44 + x_37 = x_44 = None + x_46 = x_45.permute(0, 2, 3, 1) + x_47 = torch.nn.functional.layer_norm( + x_46, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_5 = torch.nn.functional.relu(x_49, inplace=False) + x_49 = None + pow_6 = relu_5**2 + relu_5 = None + mul_5 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_6 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_6 + ) = None + x_50 = ( + mul_5 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = x_45 + x_53 + x_45 = x_53 = None + x_55 = x_54.permute(0, 2, 3, 1) + x_54 = None + x_56 = torch.nn.functional.layer_norm( + x_55, + (128,), + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) + x_57 = x_56.permute(0, 3, 1, 2) + x_56 = None + x_58 = torch.conv2d( + x_57, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_1 = sym_sum // 8 + sym_sum_2 = torch.sym_sum([1, floordiv_1]) + floordiv_1 = sym_sum_2 = None + x_59 = x_58.permute(0, 2, 3, 1) + x_60 = torch.nn.functional.layer_norm( + x_59, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_61 = x_60.permute(0, 3, 1, 2) + x_60 = None + x_62 = torch.conv2d( + x_61, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_6 = torch.nn.functional.relu(x_62, inplace=False) + x_62 = None + pow_7 = relu_6**2 + relu_6 = None + mul_6 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_7 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_7 + ) = None + x_63 = ( + mul_6 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_66 = x_58 + x_65 + x_58 = x_65 = None + x_67 = x_66.permute(0, 2, 3, 1) + x_68 = torch.nn.functional.layer_norm( + x_67, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + x_70 = torch.conv2d( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_7 = torch.nn.functional.relu(x_70, inplace=False) + x_70 = None + pow_8 = relu_7**2 + relu_7 = None + mul_7 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_8 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_8 + ) = None + x_71 = ( + mul_7 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + x_73 = torch.conv2d( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + x_75 = x_66 + x_74 + x_66 = x_74 = None + x_76 = x_75.permute(0, 2, 3, 1) + x_77 = torch.nn.functional.layer_norm( + x_76, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_8 = torch.nn.functional.relu(x_79, inplace=False) + x_79 = None + pow_9 = relu_8**2 + relu_8 = None + mul_8 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_9 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_9 + ) = None + x_80 = ( + mul_8 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_81 = torch.conv2d( + x_80, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_83 = x_75 + x_82 + x_75 = x_82 = None + x_84 = x_83.permute(0, 2, 3, 1) + x_85 = torch.nn.functional.layer_norm( + x_84, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_86 = x_85.permute(0, 3, 1, 2) + x_85 = None + x_87 = torch.conv2d( + x_86, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_9 = torch.nn.functional.relu(x_87, inplace=False) + x_87 = None + pow_10 = relu_9**2 + relu_9 = None + mul_9 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_10 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_10 + ) = None + x_88 = ( + mul_9 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = x_83 + x_91 + x_83 = x_91 = None + x_93 = x_92.permute(0, 2, 3, 1) + x_94 = torch.nn.functional.layer_norm( + x_93, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = torch.conv2d( + x_95, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_10 = torch.nn.functional.relu(x_96, inplace=False) + x_96 = None + pow_11 = relu_10**2 + relu_10 = None + mul_10 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_11 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_11 + ) = None + x_97 = ( + mul_10 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_98 = torch.conv2d( + x_97, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_99 = torch.conv2d( + x_98, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_100 = x_92 + x_99 + x_92 = x_99 = None + x_101 = x_100.permute(0, 2, 3, 1) + x_102 = torch.nn.functional.layer_norm( + x_101, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_103 = x_102.permute(0, 3, 1, 2) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_11 = torch.nn.functional.relu(x_104, inplace=False) + x_104 = None + pow_12 = relu_11**2 + relu_11 = None + mul_11 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_12 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_12 + ) = None + x_105 = ( + mul_11 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_100 + x_108 + x_100 = x_108 = None + x_110 = x_109.permute(0, 2, 3, 1) + x_111 = torch.nn.functional.layer_norm( + x_110, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_110 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) + x_112 = x_111.permute(0, 3, 1, 2) + x_111 = None + x_113 = torch.conv2d( + x_112, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_112 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_12 = torch.nn.functional.relu(x_113, inplace=False) + x_113 = None + pow_13 = relu_12**2 + relu_12 = None + mul_12 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ + * pow_13 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_13 + ) = None + x_114 = ( + mul_12 + + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_12 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_115 = torch.conv2d( + x_114, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_114 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_117 = x_109 + x_116 + x_109 = x_116 = None + x_118 = x_117.permute(0, 2, 3, 1) + x_119 = torch.nn.functional.layer_norm( + x_118, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_118 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) + x_120 = x_119.permute(0, 3, 1, 2) + x_119 = None + x_121 = torch.conv2d( + x_120, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_120 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_13 = torch.nn.functional.relu(x_121, inplace=False) + x_121 = None + pow_14 = relu_13**2 + relu_13 = None + mul_13 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + * pow_14 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( + pow_14 + ) = None + x_122 = ( + mul_13 + + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + ) + mul_13 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) + x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) + x_122 = None + x_124 = torch.conv2d( + x_123, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_123 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_126 = x_117 + x_125 + x_117 = x_125 = None + x_127 = x_126.permute(0, 2, 3, 1) + x_128 = torch.nn.functional.layer_norm( + x_127, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_127 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) + x_129 = x_128.permute(0, 3, 1, 2) + x_128 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_14 = torch.nn.functional.relu(x_130, inplace=False) + x_130 = None + pow_15 = relu_14**2 + relu_14 = None + mul_14 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ + * pow_15 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_15 + ) = None + x_131 = ( + mul_14 + + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_14 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_132 = torch.conv2d( + x_131, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_131 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_132 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_134 = x_126 + x_133 + x_126 = x_133 = None + x_135 = x_134.permute(0, 2, 3, 1) + x_136 = torch.nn.functional.layer_norm( + x_135, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_135 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) + x_137 = x_136.permute(0, 3, 1, 2) + x_136 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_15 = torch.nn.functional.relu(x_138, inplace=False) + x_138 = None + pow_16 = relu_15**2 + relu_15 = None + mul_15 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + * pow_16 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( + pow_16 + ) = None + x_139 = ( + mul_15 + + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + ) + mul_15 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) + x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) + x_139 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_140 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_134 + x_142 + x_134 = x_142 = None + x_144 = x_143.permute(0, 2, 3, 1) + x_145 = torch.nn.functional.layer_norm( + x_144, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_144 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) + x_146 = x_145.permute(0, 3, 1, 2) + x_145 = None + x_147 = torch.conv2d( + x_146, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_146 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_16 = torch.nn.functional.relu(x_147, inplace=False) + x_147 = None + pow_17 = relu_16**2 + relu_16 = None + mul_16 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ + * pow_17 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_17 + ) = None + x_148 = ( + mul_16 + + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_16 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_149 = torch.conv2d( + x_148, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_148 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_150 = torch.conv2d( + x_149, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_149 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_151 = x_143 + x_150 + x_143 = x_150 = None + x_152 = x_151.permute(0, 2, 3, 1) + x_153 = torch.nn.functional.layer_norm( + x_152, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_152 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) + x_154 = x_153.permute(0, 3, 1, 2) + x_153 = None + x_155 = torch.conv2d( + x_154, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_154 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_17 = torch.nn.functional.relu(x_155, inplace=False) + x_155 = None + pow_18 = relu_17**2 + relu_17 = None + mul_17 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + * pow_18 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( + pow_18 + ) = None + x_156 = ( + mul_17 + + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + ) + mul_17 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch.conv2d( + x_157, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_157 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_151 + x_159 + x_151 = x_159 = None + x_161 = x_160.permute(0, 2, 3, 1) + x_162 = torch.nn.functional.layer_norm( + x_161, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_161 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) + x_163 = x_162.permute(0, 3, 1, 2) + x_162 = None + x_164 = torch.conv2d( + x_163, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_163 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_18 = torch.nn.functional.relu(x_164, inplace=False) + x_164 = None + pow_19 = relu_18**2 + relu_18 = None + mul_18 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ + * pow_19 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_19 + ) = None + x_165 = ( + mul_18 + + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_18 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_166 = torch.conv2d( + x_165, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_165 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_167 = torch.conv2d( + x_166, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_166 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_168 = x_160 + x_167 + x_160 = x_167 = None + x_169 = x_168.permute(0, 2, 3, 1) + x_170 = torch.nn.functional.layer_norm( + x_169, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_169 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) + x_171 = x_170.permute(0, 3, 1, 2) + x_170 = None + x_172 = torch.conv2d( + x_171, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_171 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_19 = torch.nn.functional.relu(x_172, inplace=False) + x_172 = None + pow_20 = relu_19**2 + relu_19 = None + mul_19 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + * pow_20 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( + pow_20 + ) = None + x_173 = ( + mul_19 + + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + ) + mul_19 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) + x_174 = torch.nn.functional.dropout(x_173, 0.0, False, False) + x_173 = None + x_175 = torch.conv2d( + x_174, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_174 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = x_168 + x_176 + x_168 = x_176 = None + x_178 = x_177.permute(0, 2, 3, 1) + x_179 = torch.nn.functional.layer_norm( + x_178, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_178 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) + x_180 = x_179.permute(0, 3, 1, 2) + x_179 = None + x_181 = torch.conv2d( + x_180, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_180 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_20 = torch.nn.functional.relu(x_181, inplace=False) + x_181 = None + pow_21 = relu_20**2 + relu_20 = None + mul_20 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ + * pow_21 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_21 + ) = None + x_182 = ( + mul_20 + + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_20 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_183 = torch.conv2d( + x_182, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_182 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_184 = torch.conv2d( + x_183, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_183 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_185 = x_177 + x_184 + x_177 = x_184 = None + x_186 = x_185.permute(0, 2, 3, 1) + x_187 = torch.nn.functional.layer_norm( + x_186, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) + x_188 = x_187.permute(0, 3, 1, 2) + x_187 = None + x_189 = torch.conv2d( + x_188, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_188 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_21 = torch.nn.functional.relu(x_189, inplace=False) + x_189 = None + pow_22 = relu_21**2 + relu_21 = None + mul_21 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + * pow_22 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( + pow_22 + ) = None + x_190 = ( + mul_21 + + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + ) + mul_21 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) + x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) + x_190 = None + x_192 = torch.conv2d( + x_191, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_191 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_185 + x_193 + x_185 = x_193 = None + x_195 = x_194.permute(0, 2, 3, 1) + x_196 = torch.nn.functional.layer_norm( + x_195, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_195 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) + x_197 = x_196.permute(0, 3, 1, 2) + x_196 = None + x_198 = torch.conv2d( + x_197, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_197 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_22 = torch.nn.functional.relu(x_198, inplace=False) + x_198 = None + pow_23 = relu_22**2 + relu_22 = None + mul_22 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ + * pow_23 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_23 + ) = None + x_199 = ( + mul_22 + + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_22 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_200 = torch.conv2d( + x_199, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_199 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_201 = torch.conv2d( + x_200, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_200 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_202 = x_194 + x_201 + x_194 = x_201 = None + x_203 = x_202.permute(0, 2, 3, 1) + x_204 = torch.nn.functional.layer_norm( + x_203, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_203 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) + x_205 = x_204.permute(0, 3, 1, 2) + x_204 = None + x_206 = torch.conv2d( + x_205, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_205 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_23 = torch.nn.functional.relu(x_206, inplace=False) + x_206 = None + pow_24 = relu_23**2 + relu_23 = None + mul_23 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + * pow_24 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( + pow_24 + ) = None + x_207 = ( + mul_23 + + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + ) + mul_23 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) + x_208 = torch.nn.functional.dropout(x_207, 0.0, False, False) + x_207 = None + x_209 = torch.conv2d( + x_208, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_208 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) + x_209 = None + x_211 = x_202 + x_210 + x_202 = x_210 = None + x_212 = x_211.permute(0, 2, 3, 1) + x_213 = torch.nn.functional.layer_norm( + x_212, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_212 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) + x_214 = x_213.permute(0, 3, 1, 2) + x_213 = None + x_215 = torch.conv2d( + x_214, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_214 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_24 = torch.nn.functional.relu(x_215, inplace=False) + x_215 = None + pow_25 = relu_24**2 + relu_24 = None + mul_24 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ + * pow_25 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_25 + ) = None + x_216 = ( + mul_24 + + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_24 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_217 = torch.conv2d( + x_216, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_216 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_217 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_219 = x_211 + x_218 + x_211 = x_218 = None + x_220 = x_219.permute(0, 2, 3, 1) + x_221 = torch.nn.functional.layer_norm( + x_220, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_220 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) + x_222 = x_221.permute(0, 3, 1, 2) + x_221 = None + x_223 = torch.conv2d( + x_222, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_222 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_25 = torch.nn.functional.relu(x_223, inplace=False) + x_223 = None + pow_26 = relu_25**2 + relu_25 = None + mul_25 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + * pow_26 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( + pow_26 + ) = None + x_224 = ( + mul_25 + + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + ) + mul_25 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = torch.conv2d( + x_225, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_225 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) + x_226 = None + x_228 = x_219 + x_227 + x_219 = x_227 = None + x_229 = x_228.permute(0, 2, 3, 1) + x_230 = torch.nn.functional.layer_norm( + x_229, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_229 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) + x_231 = x_230.permute(0, 3, 1, 2) + x_230 = None + x_232 = torch.conv2d( + x_231, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_231 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_26 = torch.nn.functional.relu(x_232, inplace=False) + x_232 = None + pow_27 = relu_26**2 + relu_26 = None + mul_26 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ + * pow_27 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_27 + ) = None + x_233 = ( + mul_26 + + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_26 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_234 = torch.conv2d( + x_233, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_233 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_235 = torch.conv2d( + x_234, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_234 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_236 = x_228 + x_235 + x_228 = x_235 = None + x_237 = x_236.permute(0, 2, 3, 1) + x_238 = torch.nn.functional.layer_norm( + x_237, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_237 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) + x_239 = x_238.permute(0, 3, 1, 2) + x_238 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_239 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_27 = torch.nn.functional.relu(x_240, inplace=False) + x_240 = None + pow_28 = relu_27**2 + relu_27 = None + mul_27 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + * pow_28 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( + pow_28 + ) = None + x_241 = ( + mul_27 + + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + ) + mul_27 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) + x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) + x_241 = None + x_243 = torch.conv2d( + x_242, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_242 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_244 = torch.nn.functional.dropout(x_243, 0.0, False, False) + x_243 = None + x_245 = x_236 + x_244 + x_236 = x_244 = None + x_246 = x_245.permute(0, 2, 3, 1) + x_247 = torch.nn.functional.layer_norm( + x_246, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_246 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) + x_248 = x_247.permute(0, 3, 1, 2) + x_247 = None + x_249 = torch.conv2d( + x_248, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_248 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_28 = torch.nn.functional.relu(x_249, inplace=False) + x_249 = None + pow_29 = relu_28**2 + relu_28 = None + mul_28 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ + * pow_29 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_29 + ) = None + x_250 = ( + mul_28 + + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_28 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + x_250 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_252 = torch.conv2d( + x_251, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_251 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_253 = x_245 + x_252 + x_245 = x_252 = None + x_254 = x_253.permute(0, 2, 3, 1) + x_255 = torch.nn.functional.layer_norm( + x_254, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_254 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) + x_256 = x_255.permute(0, 3, 1, 2) + x_255 = None + x_257 = torch.conv2d( + x_256, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_256 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_29 = torch.nn.functional.relu(x_257, inplace=False) + x_257 = None + pow_30 = relu_29**2 + relu_29 = None + mul_29 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + * pow_30 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( + pow_30 + ) = None + x_258 = ( + mul_29 + + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + ) + mul_29 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = torch.conv2d( + x_259, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_259 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_261 = torch.nn.functional.dropout(x_260, 0.0, False, False) + x_260 = None + x_262 = x_253 + x_261 + x_253 = x_261 = None + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (256,), + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) + x_265 = x_264.permute(0, 3, 1, 2) + x_264 = None + x_266 = torch.conv2d( + x_265, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_265 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_2 = sym_sum // 16 + sym_sum_3 = torch.sym_sum([1, floordiv_2]) + floordiv_2 = None + reshape = x_266.reshape(1, 512, -1) + x_266 = None + x_267 = reshape.transpose(1, 2) + reshape = None + view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_30 = x_267 * view + view = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_267, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_52 = sym_sum_3**2 + linear = torch._C._nn.linear( + layer_norm_33, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_33 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_1 = linear.reshape(1, pow_52, 3, 16, 32) + linear = None + qkv = reshape_1.permute(2, 0, 3, 1, 4) + reshape_1 = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_268 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) + q = k = v = None + transpose_1 = x_268.transpose(1, 2) + x_268 = None + x_269 = transpose_1.reshape(1, pow_52, 512) + transpose_1 = None + x_270 = torch._C._nn.linear( + x_269, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_269 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_271 = torch.nn.functional.dropout(x_270, 0.0, False, False) + x_270 = None + x_272 = mul_30 + x_271 + mul_30 = x_271 = None + view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_31 = x_272 * view_1 + view_1 = None + layer_norm_34 = torch.nn.functional.layer_norm( + x_272, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_272 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_273 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_34 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_30 = torch.nn.functional.relu(x_273, inplace=False) + x_273 = None + pow_31 = relu_30**2 + relu_30 = None + mul_32 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_31 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_31 + ) = None + x_274 = ( + mul_32 + + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_275 = torch.nn.functional.dropout(x_274, 0.0, False, False) + x_274 = None + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_277 = torch.nn.functional.dropout(x_276, 0.0, False, False) + x_276 = None + x_278 = mul_31 + x_277 + mul_31 = x_277 = None + view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_33 = x_278 * view_2 + view_2 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_278, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_3 = linear_4.reshape(1, pow_52, 3, 16, 32) + linear_4 = None + qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) + reshape_3 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_279 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) + q_1 = k_1 = v_1 = None + transpose_2 = x_279.transpose(1, 2) + x_279 = None + x_280 = transpose_2.reshape(1, pow_52, 512) + transpose_2 = None + x_281 = torch._C._nn.linear( + x_280, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_280 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) + x_281 = None + x_283 = mul_33 + x_282 + mul_33 = x_282 = None + view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_34 = x_283 * view_3 + view_3 = None + layer_norm_36 = torch.nn.functional.layer_norm( + x_283, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_283 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_284 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_36 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_31 = torch.nn.functional.relu(x_284, inplace=False) + x_284 = None + pow_32 = relu_31**2 + relu_31 = None + mul_35 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_32 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_32 + ) = None + x_285 = ( + mul_35 + + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_286 = torch.nn.functional.dropout(x_285, 0.0, False, False) + x_285 = None + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) + x_287 = None + x_289 = mul_34 + x_288 + mul_34 = x_288 = None + view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_36 = x_289 * view_4 + view_4 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_289, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_37 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_5 = linear_8.reshape(1, pow_52, 3, 16, 32) + linear_8 = None + qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) + reshape_5 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_290 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) + q_2 = k_2 = v_2 = None + transpose_3 = x_290.transpose(1, 2) + x_290 = None + x_291 = transpose_3.reshape(1, pow_52, 512) + transpose_3 = None + x_292 = torch._C._nn.linear( + x_291, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_291 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_293 = torch.nn.functional.dropout(x_292, 0.0, False, False) + x_292 = None + x_294 = mul_36 + x_293 + mul_36 = x_293 = None + view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_37 = x_294 * view_5 + view_5 = None + layer_norm_38 = torch.nn.functional.layer_norm( + x_294, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_294 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_295 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_32 = torch.nn.functional.relu(x_295, inplace=False) + x_295 = None + pow_33 = relu_32**2 + relu_32 = None + mul_38 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_33 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_33 + ) = None + x_296 = ( + mul_38 + + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_297 = torch.nn.functional.dropout(x_296, 0.0, False, False) + x_296 = None + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_299 = torch.nn.functional.dropout(x_298, 0.0, False, False) + x_298 = None + x_300 = mul_37 + x_299 + mul_37 = x_299 = None + view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_39 = x_300 * view_6 + view_6 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_300, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_39 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_7 = linear_12.reshape(1, pow_52, 3, 16, 32) + linear_12 = None + qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) + reshape_7 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_301 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) + q_3 = k_3 = v_3 = None + transpose_4 = x_301.transpose(1, 2) + x_301 = None + x_302 = transpose_4.reshape(1, pow_52, 512) + transpose_4 = None + x_303 = torch._C._nn.linear( + x_302, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_302 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_304 = torch.nn.functional.dropout(x_303, 0.0, False, False) + x_303 = None + x_305 = mul_39 + x_304 + mul_39 = x_304 = None + view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_40 = x_305 * view_7 + view_7 = None + layer_norm_40 = torch.nn.functional.layer_norm( + x_305, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_305 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) + x_306 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_40 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_33 = torch.nn.functional.relu(x_306, inplace=False) + x_306 = None + pow_34 = relu_33**2 + relu_33 = None + mul_41 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + * pow_34 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( + pow_34 + ) = None + x_307 = ( + mul_41 + + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + ) + mul_41 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) + x_308 = torch.nn.functional.dropout(x_307, 0.0, False, False) + x_307 = None + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) + x_309 = None + x_311 = mul_40 + x_310 + mul_40 = x_310 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_42 = x_311 * view_8 + view_8 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_311, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_41 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_9 = linear_16.reshape(1, pow_52, 3, 16, 32) + linear_16 = None + qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) + reshape_9 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_312 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) + q_4 = k_4 = v_4 = None + transpose_5 = x_312.transpose(1, 2) + x_312 = None + x_313 = transpose_5.reshape(1, pow_52, 512) + transpose_5 = None + x_314 = torch._C._nn.linear( + x_313, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_313 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) + x_314 = None + x_316 = mul_42 + x_315 + mul_42 = x_315 = None + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_43 = x_316 * view_9 + view_9 = None + layer_norm_42 = torch.nn.functional.layer_norm( + x_316, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_316 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) + x_317 = torch._C._nn.linear( + layer_norm_42, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_42 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_34 = torch.nn.functional.relu(x_317, inplace=False) + x_317 = None + pow_35 = relu_34**2 + relu_34 = None + mul_44 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + * pow_35 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( + pow_35 + ) = None + x_318 = ( + mul_44 + + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + ) + mul_44 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) + x_319 = torch.nn.functional.dropout(x_318, 0.0, False, False) + x_318 = None + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_321 = torch.nn.functional.dropout(x_320, 0.0, False, False) + x_320 = None + x_322 = mul_43 + x_321 + mul_43 = x_321 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_45 = x_322 * view_10 + view_10 = None + layer_norm_43 = torch.nn.functional.layer_norm( + x_322, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) + linear_20 = torch._C._nn.linear( + layer_norm_43, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_43 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_11 = linear_20.reshape(1, pow_52, 3, 16, 32) + linear_20 = None + qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) + reshape_11 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_323 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) + q_5 = k_5 = v_5 = None + transpose_6 = x_323.transpose(1, 2) + x_323 = None + x_324 = transpose_6.reshape(1, pow_52, 512) + transpose_6 = None + x_325 = torch._C._nn.linear( + x_324, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_324 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_326 = torch.nn.functional.dropout(x_325, 0.0, False, False) + x_325 = None + x_327 = mul_45 + x_326 + mul_45 = x_326 = None + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_46 = x_327 * view_11 + view_11 = None + layer_norm_44 = torch.nn.functional.layer_norm( + x_327, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_327 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) + x_328 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_44 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_35 = torch.nn.functional.relu(x_328, inplace=False) + x_328 = None + pow_36 = relu_35**2 + relu_35 = None + mul_47 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + * pow_36 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( + pow_36 + ) = None + x_329 = ( + mul_47 + + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + ) + mul_47 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) + x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) + x_329 = None + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_332 = torch.nn.functional.dropout(x_331, 0.0, False, False) + x_331 = None + x_333 = mul_46 + x_332 + mul_46 = x_332 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_48 = x_333 * view_12 + view_12 = None + layer_norm_45 = torch.nn.functional.layer_norm( + x_333, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) + linear_24 = torch._C._nn.linear( + layer_norm_45, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_45 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_13 = linear_24.reshape(1, pow_52, 3, 16, 32) + linear_24 = None + qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) + reshape_13 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_334 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) + q_6 = k_6 = v_6 = None + transpose_7 = x_334.transpose(1, 2) + x_334 = None + x_335 = transpose_7.reshape(1, pow_52, 512) + transpose_7 = None + x_336 = torch._C._nn.linear( + x_335, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_335 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_337 = torch.nn.functional.dropout(x_336, 0.0, False, False) + x_336 = None + x_338 = mul_48 + x_337 + mul_48 = x_337 = None + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_49 = x_338 * view_13 + view_13 = None + layer_norm_46 = torch.nn.functional.layer_norm( + x_338, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_338 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) + x_339 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_46 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_36 = torch.nn.functional.relu(x_339, inplace=False) + x_339 = None + pow_37 = relu_36**2 + relu_36 = None + mul_50 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + * pow_37 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( + pow_37 + ) = None + x_340 = ( + mul_50 + + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + ) + mul_50 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) + x_341 = torch.nn.functional.dropout(x_340, 0.0, False, False) + x_340 = None + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_343 = torch.nn.functional.dropout(x_342, 0.0, False, False) + x_342 = None + x_344 = mul_49 + x_343 + mul_49 = x_343 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_51 = x_344 * view_14 + view_14 = None + layer_norm_47 = torch.nn.functional.layer_norm( + x_344, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) + linear_28 = torch._C._nn.linear( + layer_norm_47, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_47 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_15 = linear_28.reshape(1, pow_52, 3, 16, 32) + linear_28 = None + qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) + reshape_15 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_345 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) + q_7 = k_7 = v_7 = None + transpose_8 = x_345.transpose(1, 2) + x_345 = None + x_346 = transpose_8.reshape(1, pow_52, 512) + transpose_8 = None + x_347 = torch._C._nn.linear( + x_346, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_346 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) + x_347 = None + x_349 = mul_51 + x_348 + mul_51 = x_348 = None + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_52 = x_349 * view_15 + view_15 = None + layer_norm_48 = torch.nn.functional.layer_norm( + x_349, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_349 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) + x_350 = torch._C._nn.linear( + layer_norm_48, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_48 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_37 = torch.nn.functional.relu(x_350, inplace=False) + x_350 = None + pow_38 = relu_37**2 + relu_37 = None + mul_53 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + * pow_38 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( + pow_38 + ) = None + x_351 = ( + mul_53 + + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + ) + mul_53 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) + x_352 = torch.nn.functional.dropout(x_351, 0.0, False, False) + x_351 = None + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_354 = torch.nn.functional.dropout(x_353, 0.0, False, False) + x_353 = None + x_355 = mul_52 + x_354 + mul_52 = x_354 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_54 = x_355 * view_16 + view_16 = None + layer_norm_49 = torch.nn.functional.layer_norm( + x_355, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) + linear_32 = torch._C._nn.linear( + layer_norm_49, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_49 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_17 = linear_32.reshape(1, pow_52, 3, 16, 32) + linear_32 = None + qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) + reshape_17 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_356 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) + q_8 = k_8 = v_8 = None + transpose_9 = x_356.transpose(1, 2) + x_356 = None + x_357 = transpose_9.reshape(1, pow_52, 512) + transpose_9 = None + x_358 = torch._C._nn.linear( + x_357, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_357 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_359 = torch.nn.functional.dropout(x_358, 0.0, False, False) + x_358 = None + x_360 = mul_54 + x_359 + mul_54 = x_359 = None + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_55 = x_360 * view_17 + view_17 = None + layer_norm_50 = torch.nn.functional.layer_norm( + x_360, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_360 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) + x_361 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_50 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_38 = torch.nn.functional.relu(x_361, inplace=False) + x_361 = None + pow_39 = relu_38**2 + relu_38 = None + mul_56 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + * pow_39 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( + pow_39 + ) = None + x_362 = ( + mul_56 + + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + ) + mul_56 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) + x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) + x_362 = None + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_365 = torch.nn.functional.dropout(x_364, 0.0, False, False) + x_364 = None + x_366 = mul_55 + x_365 + mul_55 = x_365 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_57 = x_366 * view_18 + view_18 = None + layer_norm_51 = torch.nn.functional.layer_norm( + x_366, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_51, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_51 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_19 = linear_36.reshape(1, pow_52, 3, 16, 32) + linear_36 = None + qkv_9 = reshape_19.permute(2, 0, 3, 1, 4) + reshape_19 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_367 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) + q_9 = k_9 = v_9 = None + transpose_10 = x_367.transpose(1, 2) + x_367 = None + x_368 = transpose_10.reshape(1, pow_52, 512) + transpose_10 = None + x_369 = torch._C._nn.linear( + x_368, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_368 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_370 = torch.nn.functional.dropout(x_369, 0.0, False, False) + x_369 = None + x_371 = mul_57 + x_370 + mul_57 = x_370 = None + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_58 = x_371 * view_19 + view_19 = None + layer_norm_52 = torch.nn.functional.layer_norm( + x_371, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_371 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) + x_372 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_52 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_39 = torch.nn.functional.relu(x_372, inplace=False) + x_372 = None + pow_40 = relu_39**2 + relu_39 = None + mul_59 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + * pow_40 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( + pow_40 + ) = None + x_373 = ( + mul_59 + + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + ) + mul_59 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) + x_374 = torch.nn.functional.dropout(x_373, 0.0, False, False) + x_373 = None + x_375 = torch._C._nn.linear( + x_374, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_374 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_376 = torch.nn.functional.dropout(x_375, 0.0, False, False) + x_375 = None + x_377 = mul_58 + x_376 + mul_58 = x_376 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_60 = x_377 * view_20 + view_20 = None + layer_norm_53 = torch.nn.functional.layer_norm( + x_377, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_377 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_53, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_53 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_21 = linear_40.reshape(1, pow_52, 3, 16, 32) + linear_40 = None + qkv_10 = reshape_21.permute(2, 0, 3, 1, 4) + reshape_21 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_378 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_378.transpose(1, 2) + x_378 = None + x_379 = transpose_11.reshape(1, pow_52, 512) + transpose_11 = None + x_380 = torch._C._nn.linear( + x_379, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_379 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_381 = torch.nn.functional.dropout(x_380, 0.0, False, False) + x_380 = None + x_382 = mul_60 + x_381 + mul_60 = x_381 = None + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_61 = x_382 * view_21 + view_21 = None + layer_norm_54 = torch.nn.functional.layer_norm( + x_382, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_382 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) + x_383 = torch._C._nn.linear( + layer_norm_54, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_54 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_40 = torch.nn.functional.relu(x_383, inplace=False) + x_383 = None + pow_41 = relu_40**2 + relu_40 = None + mul_62 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + * pow_41 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( + pow_41 + ) = None + x_384 = ( + mul_62 + + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + ) + mul_62 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) + x_385 = torch.nn.functional.dropout(x_384, 0.0, False, False) + x_384 = None + x_386 = torch._C._nn.linear( + x_385, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_385 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_387 = torch.nn.functional.dropout(x_386, 0.0, False, False) + x_386 = None + x_388 = mul_61 + x_387 + mul_61 = x_387 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_63 = x_388 * view_22 + view_22 = None + layer_norm_55 = torch.nn.functional.layer_norm( + x_388, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_388 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_55, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_55 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_23 = linear_44.reshape(1, pow_52, 3, 16, 32) + linear_44 = None + qkv_11 = reshape_23.permute(2, 0, 3, 1, 4) + reshape_23 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_389 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_389.transpose(1, 2) + x_389 = None + x_390 = transpose_12.reshape(1, pow_52, 512) + transpose_12 = None + x_391 = torch._C._nn.linear( + x_390, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_390 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_392 = torch.nn.functional.dropout(x_391, 0.0, False, False) + x_391 = None + x_393 = mul_63 + x_392 + mul_63 = x_392 = None + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_64 = x_393 * view_23 + view_23 = None + layer_norm_56 = torch.nn.functional.layer_norm( + x_393, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_393 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) + x_394 = torch._C._nn.linear( + layer_norm_56, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_56 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_41 = torch.nn.functional.relu(x_394, inplace=False) + x_394 = None + pow_42 = relu_41**2 + relu_41 = None + mul_65 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + * pow_42 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( + pow_42 + ) = None + x_395 = ( + mul_65 + + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + ) + mul_65 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) + x_396 = torch.nn.functional.dropout(x_395, 0.0, False, False) + x_395 = None + x_397 = torch._C._nn.linear( + x_396, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_396 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_398 = torch.nn.functional.dropout(x_397, 0.0, False, False) + x_397 = None + x_399 = mul_64 + x_398 + mul_64 = x_398 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_66 = x_399 * view_24 + view_24 = None + layer_norm_57 = torch.nn.functional.layer_norm( + x_399, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_399 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_57, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_57 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_25 = linear_48.reshape(1, pow_52, 3, 16, 32) + linear_48 = None + qkv_12 = reshape_25.permute(2, 0, 3, 1, 4) + reshape_25 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_12 = unbind_12[0] + k_12 = unbind_12[1] + v_12 = unbind_12[2] + unbind_12 = None + x_400 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_13 = x_400.transpose(1, 2) + x_400 = None + x_401 = transpose_13.reshape(1, pow_52, 512) + transpose_13 = None + x_402 = torch._C._nn.linear( + x_401, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_401 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_403 = torch.nn.functional.dropout(x_402, 0.0, False, False) + x_402 = None + x_404 = mul_66 + x_403 + mul_66 = x_403 = None + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_67 = x_404 * view_25 + view_25 = None + layer_norm_58 = torch.nn.functional.layer_norm( + x_404, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_404 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = (None) + x_405 = torch._C._nn.linear( + layer_norm_58, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_58 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_42 = torch.nn.functional.relu(x_405, inplace=False) + x_405 = None + pow_43 = relu_42**2 + relu_42 = None + mul_68 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ + * pow_43 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = ( + pow_43 + ) = None + x_406 = ( + mul_68 + + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ + ) + mul_68 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = (None) + x_407 = torch.nn.functional.dropout(x_406, 0.0, False, False) + x_406 = None + x_408 = torch._C._nn.linear( + x_407, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_407 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_409 = torch.nn.functional.dropout(x_408, 0.0, False, False) + x_408 = None + x_410 = mul_67 + x_409 + mul_67 = x_409 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_69 = x_410 * view_26 + view_26 = None + layer_norm_59 = torch.nn.functional.layer_norm( + x_410, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_410 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = (None) + linear_52 = torch._C._nn.linear( + layer_norm_59, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_59 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_27 = linear_52.reshape(1, pow_52, 3, 16, 32) + linear_52 = None + qkv_13 = reshape_27.permute(2, 0, 3, 1, 4) + reshape_27 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_13 = unbind_13[0] + k_13 = unbind_13[1] + v_13 = unbind_13[2] + unbind_13 = None + x_411 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_14 = x_411.transpose(1, 2) + x_411 = None + x_412 = transpose_14.reshape(1, pow_52, 512) + transpose_14 = None + x_413 = torch._C._nn.linear( + x_412, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_412 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_414 = torch.nn.functional.dropout(x_413, 0.0, False, False) + x_413 = None + x_415 = mul_69 + x_414 + mul_69 = x_414 = None + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_70 = x_415 * view_27 + view_27 = None + layer_norm_60 = torch.nn.functional.layer_norm( + x_415, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_415 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = (None) + x_416 = torch._C._nn.linear( + layer_norm_60, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_60 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_43 = torch.nn.functional.relu(x_416, inplace=False) + x_416 = None + pow_44 = relu_43**2 + relu_43 = None + mul_71 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ + * pow_44 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = ( + pow_44 + ) = None + x_417 = ( + mul_71 + + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ + ) + mul_71 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = (None) + x_418 = torch.nn.functional.dropout(x_417, 0.0, False, False) + x_417 = None + x_419 = torch._C._nn.linear( + x_418, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_418 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_420 = torch.nn.functional.dropout(x_419, 0.0, False, False) + x_419 = None + x_421 = mul_70 + x_420 + mul_70 = x_420 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_72 = x_421 * view_28 + view_28 = None + layer_norm_61 = torch.nn.functional.layer_norm( + x_421, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_421 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_61, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_61 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_29 = linear_56.reshape(1, pow_52, 3, 16, 32) + linear_56 = None + qkv_14 = reshape_29.permute(2, 0, 3, 1, 4) + reshape_29 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_14 = unbind_14[0] + k_14 = unbind_14[1] + v_14 = unbind_14[2] + unbind_14 = None + x_422 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_15 = x_422.transpose(1, 2) + x_422 = None + x_423 = transpose_15.reshape(1, pow_52, 512) + transpose_15 = None + x_424 = torch._C._nn.linear( + x_423, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_423 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_425 = torch.nn.functional.dropout(x_424, 0.0, False, False) + x_424 = None + x_426 = mul_72 + x_425 + mul_72 = x_425 = None + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_73 = x_426 * view_29 + view_29 = None + layer_norm_62 = torch.nn.functional.layer_norm( + x_426, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_426 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = (None) + x_427 = torch._C._nn.linear( + layer_norm_62, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_62 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_44 = torch.nn.functional.relu(x_427, inplace=False) + x_427 = None + pow_45 = relu_44**2 + relu_44 = None + mul_74 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ + * pow_45 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = ( + pow_45 + ) = None + x_428 = ( + mul_74 + + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ + ) + mul_74 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = (None) + x_429 = torch.nn.functional.dropout(x_428, 0.0, False, False) + x_428 = None + x_430 = torch._C._nn.linear( + x_429, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_429 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_431 = torch.nn.functional.dropout(x_430, 0.0, False, False) + x_430 = None + x_432 = mul_73 + x_431 + mul_73 = x_431 = None + view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_75 = x_432 * view_30 + view_30 = None + layer_norm_63 = torch.nn.functional.layer_norm( + x_432, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_432 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = (None) + linear_60 = torch._C._nn.linear( + layer_norm_63, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_63 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_31 = linear_60.reshape(1, pow_52, 3, 16, 32) + linear_60 = None + qkv_15 = reshape_31.permute(2, 0, 3, 1, 4) + reshape_31 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_15 = unbind_15[0] + k_15 = unbind_15[1] + v_15 = unbind_15[2] + unbind_15 = None + x_433 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_16 = x_433.transpose(1, 2) + x_433 = None + x_434 = transpose_16.reshape(1, pow_52, 512) + transpose_16 = None + x_435 = torch._C._nn.linear( + x_434, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_434 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_436 = torch.nn.functional.dropout(x_435, 0.0, False, False) + x_435 = None + x_437 = mul_75 + x_436 + mul_75 = x_436 = None + view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_76 = x_437 * view_31 + view_31 = None + layer_norm_64 = torch.nn.functional.layer_norm( + x_437, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_437 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = (None) + x_438 = torch._C._nn.linear( + layer_norm_64, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_64 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_45 = torch.nn.functional.relu(x_438, inplace=False) + x_438 = None + pow_46 = relu_45**2 + relu_45 = None + mul_77 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ + * pow_46 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = ( + pow_46 + ) = None + x_439 = ( + mul_77 + + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ + ) + mul_77 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = (None) + x_440 = torch.nn.functional.dropout(x_439, 0.0, False, False) + x_439 = None + x_441 = torch._C._nn.linear( + x_440, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_440 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_442 = torch.nn.functional.dropout(x_441, 0.0, False, False) + x_441 = None + x_443 = mul_76 + x_442 + mul_76 = x_442 = None + view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_78 = x_443 * view_32 + view_32 = None + layer_norm_65 = torch.nn.functional.layer_norm( + x_443, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_443 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_65, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_65 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_33 = linear_64.reshape(1, pow_52, 3, 16, 32) + linear_64 = None + qkv_16 = reshape_33.permute(2, 0, 3, 1, 4) + reshape_33 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_16 = unbind_16[0] + k_16 = unbind_16[1] + v_16 = unbind_16[2] + unbind_16 = None + x_444 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_17 = x_444.transpose(1, 2) + x_444 = None + x_445 = transpose_17.reshape(1, pow_52, 512) + transpose_17 = None + x_446 = torch._C._nn.linear( + x_445, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_445 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_447 = torch.nn.functional.dropout(x_446, 0.0, False, False) + x_446 = None + x_448 = mul_78 + x_447 + mul_78 = x_447 = None + view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_79 = x_448 * view_33 + view_33 = None + layer_norm_66 = torch.nn.functional.layer_norm( + x_448, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_448 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = (None) + x_449 = torch._C._nn.linear( + layer_norm_66, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_66 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_46 = torch.nn.functional.relu(x_449, inplace=False) + x_449 = None + pow_47 = relu_46**2 + relu_46 = None + mul_80 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ + * pow_47 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = ( + pow_47 + ) = None + x_450 = ( + mul_80 + + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ + ) + mul_80 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = (None) + x_451 = torch.nn.functional.dropout(x_450, 0.0, False, False) + x_450 = None + x_452 = torch._C._nn.linear( + x_451, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_451 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_453 = torch.nn.functional.dropout(x_452, 0.0, False, False) + x_452 = None + x_454 = mul_79 + x_453 + mul_79 = x_453 = None + view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_81 = x_454 * view_34 + view_34 = None + layer_norm_67 = torch.nn.functional.layer_norm( + x_454, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_454 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = (None) + linear_68 = torch._C._nn.linear( + layer_norm_67, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_67 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_35 = linear_68.reshape(1, pow_52, 3, 16, 32) + linear_68 = None + qkv_17 = reshape_35.permute(2, 0, 3, 1, 4) + reshape_35 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_17 = unbind_17[0] + k_17 = unbind_17[1] + v_17 = unbind_17[2] + unbind_17 = None + x_455 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_18 = x_455.transpose(1, 2) + x_455 = None + x_456 = transpose_18.reshape(1, pow_52, 512) + transpose_18 = pow_52 = None + x_457 = torch._C._nn.linear( + x_456, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_456 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_458 = torch.nn.functional.dropout(x_457, 0.0, False, False) + x_457 = None + x_459 = mul_81 + x_458 + mul_81 = x_458 = None + view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_82 = x_459 * view_35 + view_35 = None + layer_norm_68 = torch.nn.functional.layer_norm( + x_459, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_459 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = (None) + x_460 = torch._C._nn.linear( + layer_norm_68, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_68 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_47 = torch.nn.functional.relu(x_460, inplace=False) + x_460 = None + pow_48 = relu_47**2 + relu_47 = None + mul_83 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ + * pow_48 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = ( + pow_48 + ) = None + x_461 = ( + mul_83 + + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ + ) + mul_83 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = (None) + x_462 = torch.nn.functional.dropout(x_461, 0.0, False, False) + x_461 = None + x_463 = torch._C._nn.linear( + x_462, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_462 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_464 = torch.nn.functional.dropout(x_463, 0.0, False, False) + x_463 = None + x_465 = mul_82 + x_464 + mul_82 = x_464 = None + transpose_19 = x_465.transpose(1, 2) + x_465 = None + x_466 = transpose_19.reshape(1, 512, sym_sum_3, sym_sum_3) + transpose_19 = sym_sum_3 = None + x_467 = x_466.permute(0, 2, 3, 1) + x_466 = None + x_468 = torch.nn.functional.layer_norm( + x_467, + (512,), + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_467 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) + x_469 = x_468.permute(0, 3, 1, 2) + x_468 = None + x_470 = torch.conv2d( + x_469, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_469 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_3 = sym_sum // 32 + sym_sum = None + sym_sum_4 = torch.sym_sum([1, floordiv_3]) + floordiv_3 = None + reshape_38 = x_470.reshape(1, 768, -1) + x_470 = None + x_471 = reshape_38.transpose(1, 2) + reshape_38 = None + view_36 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_84 = x_471 * view_36 + view_36 = None + layer_norm_70 = torch.nn.functional.layer_norm( + x_471, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_471 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_53 = sym_sum_4**2 + linear_72 = torch._C._nn.linear( + layer_norm_70, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_70 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_39 = linear_72.reshape(1, pow_53, 3, 24, 32) + linear_72 = None + qkv_18 = reshape_39.permute(2, 0, 3, 1, 4) + reshape_39 = None + unbind_18 = qkv_18.unbind(0) + qkv_18 = None + q_18 = unbind_18[0] + k_18 = unbind_18[1] + v_18 = unbind_18[2] + unbind_18 = None + x_472 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_21 = x_472.transpose(1, 2) + x_472 = None + x_473 = transpose_21.reshape(1, pow_53, 768) + transpose_21 = None + x_474 = torch._C._nn.linear( + x_473, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_473 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_475 = torch.nn.functional.dropout(x_474, 0.0, False, False) + x_474 = None + x_476 = mul_84 + x_475 + mul_84 = x_475 = None + view_37 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_85 = x_476 * view_37 + view_37 = None + layer_norm_71 = torch.nn.functional.layer_norm( + x_476, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_476 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_477 = torch._C._nn.linear( + layer_norm_71, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_71 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_48 = torch.nn.functional.relu(x_477, inplace=False) + x_477 = None + pow_49 = relu_48**2 + relu_48 = None + mul_86 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_49 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_49 + ) = None + x_478 = ( + mul_86 + + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_86 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_479 = torch.nn.functional.dropout(x_478, 0.0, False, False) + x_478 = None + x_480 = torch._C._nn.linear( + x_479, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_479 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_481 = torch.nn.functional.dropout(x_480, 0.0, False, False) + x_480 = None + x_482 = mul_85 + x_481 + mul_85 = x_481 = None + view_38 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_87 = x_482 * view_38 + view_38 = None + layer_norm_72 = torch.nn.functional.layer_norm( + x_482, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_482 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_76 = torch._C._nn.linear( + layer_norm_72, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_72 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_41 = linear_76.reshape(1, pow_53, 3, 24, 32) + linear_76 = None + qkv_19 = reshape_41.permute(2, 0, 3, 1, 4) + reshape_41 = None + unbind_19 = qkv_19.unbind(0) + qkv_19 = None + q_19 = unbind_19[0] + k_19 = unbind_19[1] + v_19 = unbind_19[2] + unbind_19 = None + x_483 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_22 = x_483.transpose(1, 2) + x_483 = None + x_484 = transpose_22.reshape(1, pow_53, 768) + transpose_22 = None + x_485 = torch._C._nn.linear( + x_484, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_484 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_486 = torch.nn.functional.dropout(x_485, 0.0, False, False) + x_485 = None + x_487 = mul_87 + x_486 + mul_87 = x_486 = None + view_39 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_88 = x_487 * view_39 + view_39 = None + layer_norm_73 = torch.nn.functional.layer_norm( + x_487, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_487 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_488 = torch._C._nn.linear( + layer_norm_73, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_73 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_49 = torch.nn.functional.relu(x_488, inplace=False) + x_488 = None + pow_50 = relu_49**2 + relu_49 = None + mul_89 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_50 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_50 + ) = None + x_489 = ( + mul_89 + + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_89 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_490 = torch.nn.functional.dropout(x_489, 0.0, False, False) + x_489 = None + x_491 = torch._C._nn.linear( + x_490, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_490 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_492 = torch.nn.functional.dropout(x_491, 0.0, False, False) + x_491 = None + x_493 = mul_88 + x_492 + mul_88 = x_492 = None + view_40 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_90 = x_493 * view_40 + view_40 = None + layer_norm_74 = torch.nn.functional.layer_norm( + x_493, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_493 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_74, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_74 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_43 = linear_80.reshape(1, pow_53, 3, 24, 32) + linear_80 = None + qkv_20 = reshape_43.permute(2, 0, 3, 1, 4) + reshape_43 = None + unbind_20 = qkv_20.unbind(0) + qkv_20 = None + q_20 = unbind_20[0] + k_20 = unbind_20[1] + v_20 = unbind_20[2] + unbind_20 = None + x_494 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_23 = x_494.transpose(1, 2) + x_494 = None + x_495 = transpose_23.reshape(1, pow_53, 768) + transpose_23 = pow_53 = None + x_496 = torch._C._nn.linear( + x_495, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_495 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_497 = torch.nn.functional.dropout(x_496, 0.0, False, False) + x_496 = None + x_498 = mul_90 + x_497 + mul_90 = x_497 = None + view_41 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (768,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_91 = x_498 * view_41 + view_41 = None + layer_norm_75 = torch.nn.functional.layer_norm( + x_498, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_498 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_499 = torch._C._nn.linear( + layer_norm_75, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_75 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_50 = torch.nn.functional.relu(x_499, inplace=False) + x_499 = None + pow_51 = relu_50**2 + relu_50 = None + mul_92 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_51 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_51 + ) = None + x_500 = ( + mul_92 + + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_92 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_501 = torch.nn.functional.dropout(x_500, 0.0, False, False) + x_500 = None + x_502 = torch._C._nn.linear( + x_501, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_501 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_503 = torch.nn.functional.dropout(x_502, 0.0, False, False) + x_502 = None + x_504 = mul_91 + x_503 + mul_91 = x_503 = None + transpose_24 = x_504.transpose(1, 2) + x_504 = None + x_505 = transpose_24.reshape(1, 768, sym_sum_4, sym_sum_4) + transpose_24 = sym_sum_4 = None + x_506 = torch.nn.functional.adaptive_avg_pool2d(x_505, 1) + x_505 = None + x_507 = x_506.permute(0, 2, 3, 1) + x_506 = None + x_508 = torch.nn.functional.layer_norm( + x_507, + (768,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_507 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_509 = x_508.permute(0, 3, 1, 2) + x_508 = None + x_510 = x_509.flatten(1, -1) + x_509 = None + x_511 = torch.nn.functional.dropout(x_510, 0.0, False, False) + x_510 = None + x_512 = torch._C._nn.linear( + x_511, + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, + ) + x_511 = ( + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None + relu_51 = torch.nn.functional.relu(x_512, inplace=False) + x_512 = None + x_513 = torch.square(relu_51) + relu_51 = None + x_514 = torch.nn.functional.layer_norm( + x_513, + (3072,), + l_self_modules_head_modules_fc_modules_norm_parameters_weight_, + l_self_modules_head_modules_fc_modules_norm_parameters_bias_, + 1e-06, + ) + x_513 = ( + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None + x_515 = torch.nn.functional.dropout(x_514, 0.0, False, False) + x_514 = None + x_516 = torch._C._nn.linear( + x_515, + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, + ) + x_515 = ( + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None + return (x_516,) diff --git a/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/weight_meta.py b/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/weight_meta.py new file mode 100644 index 0000000000..6cf290fca4 --- /dev/null +++ b/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/weight_meta.py @@ -0,0 +1,3968 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv_parameters_weight_" + shape = [128, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: + name = "L_self_modules_stem_modules_conv_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.226 + std = 1.284 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: + name = "L_self_modules_stem_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" + shape = [768, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" + shape = [1000, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/caformer_m36.sail_in1k/graph_net.json b/samples/timm/caformer_m36.sail_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/caformer_m36.sail_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/caformer_m36.sail_in1k/input_meta.py b/samples/timm/caformer_m36.sail_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_m36.sail_in1k/input_tensor_constraints.py b/samples/timm/caformer_m36.sail_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_m36.sail_in1k/model.py b/samples/timm/caformer_m36.sail_in1k/model.py new file mode 100644 index 0000000000..dcce286c5c --- /dev/null +++ b/samples/timm/caformer_m36.sail_in1k/model.py @@ -0,0 +1,4754 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv_parameters_bias_ = ( + L_self_modules_stem_modules_conv_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_norm_parameters_weight_ = ( + L_self_modules_stem_modules_norm_parameters_weight_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv_parameters_weight_, + l_self_modules_stem_modules_conv_parameters_bias_, + (4, 4), + (2, 2), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_conv_parameters_weight_ + ) = l_self_modules_stem_modules_conv_parameters_bias_ = None + x_1 = x.permute(0, 2, 3, 1) + x = None + x_2 = torch.nn.functional.layer_norm( + x_1, (96,), l_self_modules_stem_modules_norm_parameters_weight_, None, 1e-06 + ) + x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None + x_3 = x_2.permute(0, 3, 1, 2) + x_2 = None + sym_sum = torch.sym_sum([-3, s1]) + s1 = None + floordiv = sym_sum // 4 + sym_sum_1 = torch.sym_sum([1, floordiv]) + floordiv = sym_sum_1 = None + x_4 = x_3.permute(0, 2, 3, 1) + x_5 = torch.nn.functional.layer_norm( + x_4, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu = torch.nn.functional.relu(x_7, inplace=False) + x_7 = None + pow_1 = relu**2 + relu = None + mul = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_1 + ) = None + x_8 = ( + mul + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_11 = x_3 + x_10 + x_3 = x_10 = None + x_12 = x_11.permute(0, 2, 3, 1) + x_13 = torch.nn.functional.layer_norm( + x_12, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_14 = x_13.permute(0, 3, 1, 2) + x_13 = None + x_15 = torch.conv2d( + x_14, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_1 = torch.nn.functional.relu(x_15, inplace=False) + x_15 = None + pow_2 = relu_1**2 + relu_1 = None + mul_1 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_2 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_2 + ) = None + x_16 = ( + mul_1 + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = x_11 + x_19 + x_11 = x_19 = None + x_21 = x_20.permute(0, 2, 3, 1) + x_22 = torch.nn.functional.layer_norm( + x_21, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = torch.conv2d( + x_23, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_2 = torch.nn.functional.relu(x_24, inplace=False) + x_24 = None + pow_3 = relu_2**2 + relu_2 = None + mul_2 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_3 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_3 + ) = None + x_25 = ( + mul_2 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_27 = torch.conv2d( + x_26, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_28 = x_20 + x_27 + x_20 = x_27 = None + x_29 = x_28.permute(0, 2, 3, 1) + x_30 = torch.nn.functional.layer_norm( + x_29, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_3 = torch.nn.functional.relu(x_32, inplace=False) + x_32 = None + pow_4 = relu_3**2 + relu_3 = None + mul_3 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_4 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_4 + ) = None + x_33 = ( + mul_3 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + x_37 = x_28 + x_36 + x_28 = x_36 = None + x_38 = x_37.permute(0, 2, 3, 1) + x_39 = torch.nn.functional.layer_norm( + x_38, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_40 = x_39.permute(0, 3, 1, 2) + x_39 = None + x_41 = torch.conv2d( + x_40, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_4 = torch.nn.functional.relu(x_41, inplace=False) + x_41 = None + pow_5 = relu_4**2 + relu_4 = None + mul_4 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_5 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_5 + ) = None + x_42 = ( + mul_4 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_44 = torch.conv2d( + x_43, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_45 = x_37 + x_44 + x_37 = x_44 = None + x_46 = x_45.permute(0, 2, 3, 1) + x_47 = torch.nn.functional.layer_norm( + x_46, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_5 = torch.nn.functional.relu(x_49, inplace=False) + x_49 = None + pow_6 = relu_5**2 + relu_5 = None + mul_5 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_6 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_6 + ) = None + x_50 = ( + mul_5 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = x_45 + x_53 + x_45 = x_53 = None + x_55 = x_54.permute(0, 2, 3, 1) + x_54 = None + x_56 = torch.nn.functional.layer_norm( + x_55, + (96,), + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) + x_57 = x_56.permute(0, 3, 1, 2) + x_56 = None + x_58 = torch.conv2d( + x_57, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_1 = sym_sum // 8 + sym_sum_2 = torch.sym_sum([1, floordiv_1]) + floordiv_1 = sym_sum_2 = None + x_59 = x_58.permute(0, 2, 3, 1) + x_60 = torch.nn.functional.layer_norm( + x_59, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_61 = x_60.permute(0, 3, 1, 2) + x_60 = None + x_62 = torch.conv2d( + x_61, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_6 = torch.nn.functional.relu(x_62, inplace=False) + x_62 = None + pow_7 = relu_6**2 + relu_6 = None + mul_6 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_7 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_7 + ) = None + x_63 = ( + mul_6 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_66 = x_58 + x_65 + x_58 = x_65 = None + x_67 = x_66.permute(0, 2, 3, 1) + x_68 = torch.nn.functional.layer_norm( + x_67, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + x_70 = torch.conv2d( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_7 = torch.nn.functional.relu(x_70, inplace=False) + x_70 = None + pow_8 = relu_7**2 + relu_7 = None + mul_7 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_8 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_8 + ) = None + x_71 = ( + mul_7 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + x_73 = torch.conv2d( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + x_75 = x_66 + x_74 + x_66 = x_74 = None + x_76 = x_75.permute(0, 2, 3, 1) + x_77 = torch.nn.functional.layer_norm( + x_76, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_8 = torch.nn.functional.relu(x_79, inplace=False) + x_79 = None + pow_9 = relu_8**2 + relu_8 = None + mul_8 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_9 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_9 + ) = None + x_80 = ( + mul_8 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_81 = torch.conv2d( + x_80, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_83 = x_75 + x_82 + x_75 = x_82 = None + x_84 = x_83.permute(0, 2, 3, 1) + x_85 = torch.nn.functional.layer_norm( + x_84, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_86 = x_85.permute(0, 3, 1, 2) + x_85 = None + x_87 = torch.conv2d( + x_86, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_9 = torch.nn.functional.relu(x_87, inplace=False) + x_87 = None + pow_10 = relu_9**2 + relu_9 = None + mul_9 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_10 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_10 + ) = None + x_88 = ( + mul_9 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = x_83 + x_91 + x_83 = x_91 = None + x_93 = x_92.permute(0, 2, 3, 1) + x_94 = torch.nn.functional.layer_norm( + x_93, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = torch.conv2d( + x_95, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_10 = torch.nn.functional.relu(x_96, inplace=False) + x_96 = None + pow_11 = relu_10**2 + relu_10 = None + mul_10 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_11 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_11 + ) = None + x_97 = ( + mul_10 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_98 = torch.conv2d( + x_97, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_99 = torch.conv2d( + x_98, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_100 = x_92 + x_99 + x_92 = x_99 = None + x_101 = x_100.permute(0, 2, 3, 1) + x_102 = torch.nn.functional.layer_norm( + x_101, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_103 = x_102.permute(0, 3, 1, 2) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_11 = torch.nn.functional.relu(x_104, inplace=False) + x_104 = None + pow_12 = relu_11**2 + relu_11 = None + mul_11 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_12 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_12 + ) = None + x_105 = ( + mul_11 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_100 + x_108 + x_100 = x_108 = None + x_110 = x_109.permute(0, 2, 3, 1) + x_111 = torch.nn.functional.layer_norm( + x_110, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_110 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) + x_112 = x_111.permute(0, 3, 1, 2) + x_111 = None + x_113 = torch.conv2d( + x_112, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_112 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_12 = torch.nn.functional.relu(x_113, inplace=False) + x_113 = None + pow_13 = relu_12**2 + relu_12 = None + mul_12 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ + * pow_13 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_13 + ) = None + x_114 = ( + mul_12 + + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_12 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_115 = torch.conv2d( + x_114, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_114 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_117 = x_109 + x_116 + x_109 = x_116 = None + x_118 = x_117.permute(0, 2, 3, 1) + x_119 = torch.nn.functional.layer_norm( + x_118, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_118 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) + x_120 = x_119.permute(0, 3, 1, 2) + x_119 = None + x_121 = torch.conv2d( + x_120, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_120 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_13 = torch.nn.functional.relu(x_121, inplace=False) + x_121 = None + pow_14 = relu_13**2 + relu_13 = None + mul_13 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + * pow_14 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( + pow_14 + ) = None + x_122 = ( + mul_13 + + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + ) + mul_13 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) + x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) + x_122 = None + x_124 = torch.conv2d( + x_123, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_123 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_126 = x_117 + x_125 + x_117 = x_125 = None + x_127 = x_126.permute(0, 2, 3, 1) + x_128 = torch.nn.functional.layer_norm( + x_127, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_127 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) + x_129 = x_128.permute(0, 3, 1, 2) + x_128 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_14 = torch.nn.functional.relu(x_130, inplace=False) + x_130 = None + pow_15 = relu_14**2 + relu_14 = None + mul_14 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ + * pow_15 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_15 + ) = None + x_131 = ( + mul_14 + + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_14 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_132 = torch.conv2d( + x_131, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_131 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_132 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_134 = x_126 + x_133 + x_126 = x_133 = None + x_135 = x_134.permute(0, 2, 3, 1) + x_136 = torch.nn.functional.layer_norm( + x_135, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_135 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) + x_137 = x_136.permute(0, 3, 1, 2) + x_136 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_15 = torch.nn.functional.relu(x_138, inplace=False) + x_138 = None + pow_16 = relu_15**2 + relu_15 = None + mul_15 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + * pow_16 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( + pow_16 + ) = None + x_139 = ( + mul_15 + + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + ) + mul_15 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) + x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) + x_139 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_140 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_134 + x_142 + x_134 = x_142 = None + x_144 = x_143.permute(0, 2, 3, 1) + x_145 = torch.nn.functional.layer_norm( + x_144, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_144 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) + x_146 = x_145.permute(0, 3, 1, 2) + x_145 = None + x_147 = torch.conv2d( + x_146, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_146 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_16 = torch.nn.functional.relu(x_147, inplace=False) + x_147 = None + pow_17 = relu_16**2 + relu_16 = None + mul_16 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ + * pow_17 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_17 + ) = None + x_148 = ( + mul_16 + + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_16 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_149 = torch.conv2d( + x_148, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_148 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_150 = torch.conv2d( + x_149, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_149 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_151 = x_143 + x_150 + x_143 = x_150 = None + x_152 = x_151.permute(0, 2, 3, 1) + x_153 = torch.nn.functional.layer_norm( + x_152, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_152 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) + x_154 = x_153.permute(0, 3, 1, 2) + x_153 = None + x_155 = torch.conv2d( + x_154, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_154 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_17 = torch.nn.functional.relu(x_155, inplace=False) + x_155 = None + pow_18 = relu_17**2 + relu_17 = None + mul_17 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + * pow_18 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( + pow_18 + ) = None + x_156 = ( + mul_17 + + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + ) + mul_17 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch.conv2d( + x_157, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_157 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_151 + x_159 + x_151 = x_159 = None + x_161 = x_160.permute(0, 2, 3, 1) + x_162 = torch.nn.functional.layer_norm( + x_161, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_161 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) + x_163 = x_162.permute(0, 3, 1, 2) + x_162 = None + x_164 = torch.conv2d( + x_163, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_163 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_18 = torch.nn.functional.relu(x_164, inplace=False) + x_164 = None + pow_19 = relu_18**2 + relu_18 = None + mul_18 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ + * pow_19 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_19 + ) = None + x_165 = ( + mul_18 + + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_18 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_166 = torch.conv2d( + x_165, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_165 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_167 = torch.conv2d( + x_166, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_166 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_168 = x_160 + x_167 + x_160 = x_167 = None + x_169 = x_168.permute(0, 2, 3, 1) + x_170 = torch.nn.functional.layer_norm( + x_169, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_169 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) + x_171 = x_170.permute(0, 3, 1, 2) + x_170 = None + x_172 = torch.conv2d( + x_171, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_171 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_19 = torch.nn.functional.relu(x_172, inplace=False) + x_172 = None + pow_20 = relu_19**2 + relu_19 = None + mul_19 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + * pow_20 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( + pow_20 + ) = None + x_173 = ( + mul_19 + + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + ) + mul_19 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) + x_174 = torch.nn.functional.dropout(x_173, 0.0, False, False) + x_173 = None + x_175 = torch.conv2d( + x_174, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_174 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = x_168 + x_176 + x_168 = x_176 = None + x_178 = x_177.permute(0, 2, 3, 1) + x_179 = torch.nn.functional.layer_norm( + x_178, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_178 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) + x_180 = x_179.permute(0, 3, 1, 2) + x_179 = None + x_181 = torch.conv2d( + x_180, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_180 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_20 = torch.nn.functional.relu(x_181, inplace=False) + x_181 = None + pow_21 = relu_20**2 + relu_20 = None + mul_20 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ + * pow_21 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_21 + ) = None + x_182 = ( + mul_20 + + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_20 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_183 = torch.conv2d( + x_182, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_182 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_184 = torch.conv2d( + x_183, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_183 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_185 = x_177 + x_184 + x_177 = x_184 = None + x_186 = x_185.permute(0, 2, 3, 1) + x_187 = torch.nn.functional.layer_norm( + x_186, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) + x_188 = x_187.permute(0, 3, 1, 2) + x_187 = None + x_189 = torch.conv2d( + x_188, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_188 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_21 = torch.nn.functional.relu(x_189, inplace=False) + x_189 = None + pow_22 = relu_21**2 + relu_21 = None + mul_21 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + * pow_22 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( + pow_22 + ) = None + x_190 = ( + mul_21 + + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + ) + mul_21 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) + x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) + x_190 = None + x_192 = torch.conv2d( + x_191, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_191 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_185 + x_193 + x_185 = x_193 = None + x_195 = x_194.permute(0, 2, 3, 1) + x_196 = torch.nn.functional.layer_norm( + x_195, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_195 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) + x_197 = x_196.permute(0, 3, 1, 2) + x_196 = None + x_198 = torch.conv2d( + x_197, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_197 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_22 = torch.nn.functional.relu(x_198, inplace=False) + x_198 = None + pow_23 = relu_22**2 + relu_22 = None + mul_22 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ + * pow_23 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_23 + ) = None + x_199 = ( + mul_22 + + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_22 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_200 = torch.conv2d( + x_199, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_199 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_201 = torch.conv2d( + x_200, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_200 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_202 = x_194 + x_201 + x_194 = x_201 = None + x_203 = x_202.permute(0, 2, 3, 1) + x_204 = torch.nn.functional.layer_norm( + x_203, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_203 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) + x_205 = x_204.permute(0, 3, 1, 2) + x_204 = None + x_206 = torch.conv2d( + x_205, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_205 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_23 = torch.nn.functional.relu(x_206, inplace=False) + x_206 = None + pow_24 = relu_23**2 + relu_23 = None + mul_23 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + * pow_24 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( + pow_24 + ) = None + x_207 = ( + mul_23 + + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + ) + mul_23 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) + x_208 = torch.nn.functional.dropout(x_207, 0.0, False, False) + x_207 = None + x_209 = torch.conv2d( + x_208, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_208 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) + x_209 = None + x_211 = x_202 + x_210 + x_202 = x_210 = None + x_212 = x_211.permute(0, 2, 3, 1) + x_213 = torch.nn.functional.layer_norm( + x_212, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_212 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) + x_214 = x_213.permute(0, 3, 1, 2) + x_213 = None + x_215 = torch.conv2d( + x_214, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_214 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_24 = torch.nn.functional.relu(x_215, inplace=False) + x_215 = None + pow_25 = relu_24**2 + relu_24 = None + mul_24 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ + * pow_25 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_25 + ) = None + x_216 = ( + mul_24 + + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_24 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_217 = torch.conv2d( + x_216, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_216 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_217 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_219 = x_211 + x_218 + x_211 = x_218 = None + x_220 = x_219.permute(0, 2, 3, 1) + x_221 = torch.nn.functional.layer_norm( + x_220, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_220 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) + x_222 = x_221.permute(0, 3, 1, 2) + x_221 = None + x_223 = torch.conv2d( + x_222, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_222 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_25 = torch.nn.functional.relu(x_223, inplace=False) + x_223 = None + pow_26 = relu_25**2 + relu_25 = None + mul_25 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + * pow_26 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( + pow_26 + ) = None + x_224 = ( + mul_25 + + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + ) + mul_25 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = torch.conv2d( + x_225, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_225 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) + x_226 = None + x_228 = x_219 + x_227 + x_219 = x_227 = None + x_229 = x_228.permute(0, 2, 3, 1) + x_230 = torch.nn.functional.layer_norm( + x_229, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_229 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) + x_231 = x_230.permute(0, 3, 1, 2) + x_230 = None + x_232 = torch.conv2d( + x_231, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_231 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_26 = torch.nn.functional.relu(x_232, inplace=False) + x_232 = None + pow_27 = relu_26**2 + relu_26 = None + mul_26 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ + * pow_27 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_27 + ) = None + x_233 = ( + mul_26 + + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_26 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_234 = torch.conv2d( + x_233, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_233 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_235 = torch.conv2d( + x_234, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_234 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_236 = x_228 + x_235 + x_228 = x_235 = None + x_237 = x_236.permute(0, 2, 3, 1) + x_238 = torch.nn.functional.layer_norm( + x_237, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_237 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) + x_239 = x_238.permute(0, 3, 1, 2) + x_238 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_239 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_27 = torch.nn.functional.relu(x_240, inplace=False) + x_240 = None + pow_28 = relu_27**2 + relu_27 = None + mul_27 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + * pow_28 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( + pow_28 + ) = None + x_241 = ( + mul_27 + + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + ) + mul_27 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) + x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) + x_241 = None + x_243 = torch.conv2d( + x_242, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_242 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_244 = torch.nn.functional.dropout(x_243, 0.0, False, False) + x_243 = None + x_245 = x_236 + x_244 + x_236 = x_244 = None + x_246 = x_245.permute(0, 2, 3, 1) + x_247 = torch.nn.functional.layer_norm( + x_246, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_246 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) + x_248 = x_247.permute(0, 3, 1, 2) + x_247 = None + x_249 = torch.conv2d( + x_248, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_248 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_28 = torch.nn.functional.relu(x_249, inplace=False) + x_249 = None + pow_29 = relu_28**2 + relu_28 = None + mul_28 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ + * pow_29 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_29 + ) = None + x_250 = ( + mul_28 + + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_28 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_250 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_252 = torch.conv2d( + x_251, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_251 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_253 = x_245 + x_252 + x_245 = x_252 = None + x_254 = x_253.permute(0, 2, 3, 1) + x_255 = torch.nn.functional.layer_norm( + x_254, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_254 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) + x_256 = x_255.permute(0, 3, 1, 2) + x_255 = None + x_257 = torch.conv2d( + x_256, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_256 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_29 = torch.nn.functional.relu(x_257, inplace=False) + x_257 = None + pow_30 = relu_29**2 + relu_29 = None + mul_29 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + * pow_30 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( + pow_30 + ) = None + x_258 = ( + mul_29 + + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + ) + mul_29 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = torch.conv2d( + x_259, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_259 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_261 = torch.nn.functional.dropout(x_260, 0.0, False, False) + x_260 = None + x_262 = x_253 + x_261 + x_253 = x_261 = None + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (192,), + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) + x_265 = x_264.permute(0, 3, 1, 2) + x_264 = None + x_266 = torch.conv2d( + x_265, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_265 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_2 = sym_sum // 16 + sym_sum_3 = torch.sym_sum([1, floordiv_2]) + floordiv_2 = None + reshape = x_266.reshape(1, 384, -1) + x_266 = None + x_267 = reshape.transpose(1, 2) + reshape = None + view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_30 = x_267 * view + view = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_267, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_52 = sym_sum_3**2 + linear = torch._C._nn.linear( + layer_norm_33, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_33 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_1 = linear.reshape(1, pow_52, 3, 12, 32) + linear = None + qkv = reshape_1.permute(2, 0, 3, 1, 4) + reshape_1 = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_268 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) + q = k = v = None + transpose_1 = x_268.transpose(1, 2) + x_268 = None + x_269 = transpose_1.reshape(1, pow_52, 384) + transpose_1 = None + x_270 = torch._C._nn.linear( + x_269, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_269 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_271 = torch.nn.functional.dropout(x_270, 0.0, False, False) + x_270 = None + x_272 = mul_30 + x_271 + mul_30 = x_271 = None + view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_31 = x_272 * view_1 + view_1 = None + layer_norm_34 = torch.nn.functional.layer_norm( + x_272, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_272 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_273 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_34 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_30 = torch.nn.functional.relu(x_273, inplace=False) + x_273 = None + pow_31 = relu_30**2 + relu_30 = None + mul_32 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_31 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_31 + ) = None + x_274 = ( + mul_32 + + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_275 = torch.nn.functional.dropout(x_274, 0.0, False, False) + x_274 = None + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_277 = torch.nn.functional.dropout(x_276, 0.0, False, False) + x_276 = None + x_278 = mul_31 + x_277 + mul_31 = x_277 = None + view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_33 = x_278 * view_2 + view_2 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_278, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_3 = linear_4.reshape(1, pow_52, 3, 12, 32) + linear_4 = None + qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) + reshape_3 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_279 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) + q_1 = k_1 = v_1 = None + transpose_2 = x_279.transpose(1, 2) + x_279 = None + x_280 = transpose_2.reshape(1, pow_52, 384) + transpose_2 = None + x_281 = torch._C._nn.linear( + x_280, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_280 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) + x_281 = None + x_283 = mul_33 + x_282 + mul_33 = x_282 = None + view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_34 = x_283 * view_3 + view_3 = None + layer_norm_36 = torch.nn.functional.layer_norm( + x_283, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_283 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_284 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_36 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_31 = torch.nn.functional.relu(x_284, inplace=False) + x_284 = None + pow_32 = relu_31**2 + relu_31 = None + mul_35 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_32 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_32 + ) = None + x_285 = ( + mul_35 + + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_286 = torch.nn.functional.dropout(x_285, 0.0, False, False) + x_285 = None + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) + x_287 = None + x_289 = mul_34 + x_288 + mul_34 = x_288 = None + view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_36 = x_289 * view_4 + view_4 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_289, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_37 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_5 = linear_8.reshape(1, pow_52, 3, 12, 32) + linear_8 = None + qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) + reshape_5 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_290 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) + q_2 = k_2 = v_2 = None + transpose_3 = x_290.transpose(1, 2) + x_290 = None + x_291 = transpose_3.reshape(1, pow_52, 384) + transpose_3 = None + x_292 = torch._C._nn.linear( + x_291, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_291 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_293 = torch.nn.functional.dropout(x_292, 0.0, False, False) + x_292 = None + x_294 = mul_36 + x_293 + mul_36 = x_293 = None + view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_37 = x_294 * view_5 + view_5 = None + layer_norm_38 = torch.nn.functional.layer_norm( + x_294, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_294 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_295 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_32 = torch.nn.functional.relu(x_295, inplace=False) + x_295 = None + pow_33 = relu_32**2 + relu_32 = None + mul_38 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_33 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_33 + ) = None + x_296 = ( + mul_38 + + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_297 = torch.nn.functional.dropout(x_296, 0.0, False, False) + x_296 = None + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_299 = torch.nn.functional.dropout(x_298, 0.0, False, False) + x_298 = None + x_300 = mul_37 + x_299 + mul_37 = x_299 = None + view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_39 = x_300 * view_6 + view_6 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_300, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_39 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_7 = linear_12.reshape(1, pow_52, 3, 12, 32) + linear_12 = None + qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) + reshape_7 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_301 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) + q_3 = k_3 = v_3 = None + transpose_4 = x_301.transpose(1, 2) + x_301 = None + x_302 = transpose_4.reshape(1, pow_52, 384) + transpose_4 = None + x_303 = torch._C._nn.linear( + x_302, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_302 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_304 = torch.nn.functional.dropout(x_303, 0.0, False, False) + x_303 = None + x_305 = mul_39 + x_304 + mul_39 = x_304 = None + view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_40 = x_305 * view_7 + view_7 = None + layer_norm_40 = torch.nn.functional.layer_norm( + x_305, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_305 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) + x_306 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_40 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_33 = torch.nn.functional.relu(x_306, inplace=False) + x_306 = None + pow_34 = relu_33**2 + relu_33 = None + mul_41 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + * pow_34 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( + pow_34 + ) = None + x_307 = ( + mul_41 + + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + ) + mul_41 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) + x_308 = torch.nn.functional.dropout(x_307, 0.0, False, False) + x_307 = None + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) + x_309 = None + x_311 = mul_40 + x_310 + mul_40 = x_310 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_42 = x_311 * view_8 + view_8 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_311, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_41 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_9 = linear_16.reshape(1, pow_52, 3, 12, 32) + linear_16 = None + qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) + reshape_9 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_312 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) + q_4 = k_4 = v_4 = None + transpose_5 = x_312.transpose(1, 2) + x_312 = None + x_313 = transpose_5.reshape(1, pow_52, 384) + transpose_5 = None + x_314 = torch._C._nn.linear( + x_313, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_313 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) + x_314 = None + x_316 = mul_42 + x_315 + mul_42 = x_315 = None + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_43 = x_316 * view_9 + view_9 = None + layer_norm_42 = torch.nn.functional.layer_norm( + x_316, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_316 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) + x_317 = torch._C._nn.linear( + layer_norm_42, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_42 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_34 = torch.nn.functional.relu(x_317, inplace=False) + x_317 = None + pow_35 = relu_34**2 + relu_34 = None + mul_44 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + * pow_35 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( + pow_35 + ) = None + x_318 = ( + mul_44 + + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + ) + mul_44 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) + x_319 = torch.nn.functional.dropout(x_318, 0.0, False, False) + x_318 = None + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_321 = torch.nn.functional.dropout(x_320, 0.0, False, False) + x_320 = None + x_322 = mul_43 + x_321 + mul_43 = x_321 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_45 = x_322 * view_10 + view_10 = None + layer_norm_43 = torch.nn.functional.layer_norm( + x_322, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) + linear_20 = torch._C._nn.linear( + layer_norm_43, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_43 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_11 = linear_20.reshape(1, pow_52, 3, 12, 32) + linear_20 = None + qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) + reshape_11 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_323 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) + q_5 = k_5 = v_5 = None + transpose_6 = x_323.transpose(1, 2) + x_323 = None + x_324 = transpose_6.reshape(1, pow_52, 384) + transpose_6 = None + x_325 = torch._C._nn.linear( + x_324, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_324 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_326 = torch.nn.functional.dropout(x_325, 0.0, False, False) + x_325 = None + x_327 = mul_45 + x_326 + mul_45 = x_326 = None + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_46 = x_327 * view_11 + view_11 = None + layer_norm_44 = torch.nn.functional.layer_norm( + x_327, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_327 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) + x_328 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_44 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_35 = torch.nn.functional.relu(x_328, inplace=False) + x_328 = None + pow_36 = relu_35**2 + relu_35 = None + mul_47 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + * pow_36 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( + pow_36 + ) = None + x_329 = ( + mul_47 + + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + ) + mul_47 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) + x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) + x_329 = None + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_332 = torch.nn.functional.dropout(x_331, 0.0, False, False) + x_331 = None + x_333 = mul_46 + x_332 + mul_46 = x_332 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_48 = x_333 * view_12 + view_12 = None + layer_norm_45 = torch.nn.functional.layer_norm( + x_333, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) + linear_24 = torch._C._nn.linear( + layer_norm_45, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_45 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_13 = linear_24.reshape(1, pow_52, 3, 12, 32) + linear_24 = None + qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) + reshape_13 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_334 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) + q_6 = k_6 = v_6 = None + transpose_7 = x_334.transpose(1, 2) + x_334 = None + x_335 = transpose_7.reshape(1, pow_52, 384) + transpose_7 = None + x_336 = torch._C._nn.linear( + x_335, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_335 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_337 = torch.nn.functional.dropout(x_336, 0.0, False, False) + x_336 = None + x_338 = mul_48 + x_337 + mul_48 = x_337 = None + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_49 = x_338 * view_13 + view_13 = None + layer_norm_46 = torch.nn.functional.layer_norm( + x_338, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_338 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) + x_339 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_46 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_36 = torch.nn.functional.relu(x_339, inplace=False) + x_339 = None + pow_37 = relu_36**2 + relu_36 = None + mul_50 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + * pow_37 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( + pow_37 + ) = None + x_340 = ( + mul_50 + + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + ) + mul_50 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) + x_341 = torch.nn.functional.dropout(x_340, 0.0, False, False) + x_340 = None + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_343 = torch.nn.functional.dropout(x_342, 0.0, False, False) + x_342 = None + x_344 = mul_49 + x_343 + mul_49 = x_343 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_51 = x_344 * view_14 + view_14 = None + layer_norm_47 = torch.nn.functional.layer_norm( + x_344, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) + linear_28 = torch._C._nn.linear( + layer_norm_47, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_47 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_15 = linear_28.reshape(1, pow_52, 3, 12, 32) + linear_28 = None + qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) + reshape_15 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_345 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) + q_7 = k_7 = v_7 = None + transpose_8 = x_345.transpose(1, 2) + x_345 = None + x_346 = transpose_8.reshape(1, pow_52, 384) + transpose_8 = None + x_347 = torch._C._nn.linear( + x_346, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_346 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) + x_347 = None + x_349 = mul_51 + x_348 + mul_51 = x_348 = None + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_52 = x_349 * view_15 + view_15 = None + layer_norm_48 = torch.nn.functional.layer_norm( + x_349, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_349 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) + x_350 = torch._C._nn.linear( + layer_norm_48, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_48 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_37 = torch.nn.functional.relu(x_350, inplace=False) + x_350 = None + pow_38 = relu_37**2 + relu_37 = None + mul_53 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + * pow_38 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( + pow_38 + ) = None + x_351 = ( + mul_53 + + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + ) + mul_53 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) + x_352 = torch.nn.functional.dropout(x_351, 0.0, False, False) + x_351 = None + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_354 = torch.nn.functional.dropout(x_353, 0.0, False, False) + x_353 = None + x_355 = mul_52 + x_354 + mul_52 = x_354 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_54 = x_355 * view_16 + view_16 = None + layer_norm_49 = torch.nn.functional.layer_norm( + x_355, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) + linear_32 = torch._C._nn.linear( + layer_norm_49, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_49 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_17 = linear_32.reshape(1, pow_52, 3, 12, 32) + linear_32 = None + qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) + reshape_17 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_356 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) + q_8 = k_8 = v_8 = None + transpose_9 = x_356.transpose(1, 2) + x_356 = None + x_357 = transpose_9.reshape(1, pow_52, 384) + transpose_9 = None + x_358 = torch._C._nn.linear( + x_357, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_357 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_359 = torch.nn.functional.dropout(x_358, 0.0, False, False) + x_358 = None + x_360 = mul_54 + x_359 + mul_54 = x_359 = None + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_55 = x_360 * view_17 + view_17 = None + layer_norm_50 = torch.nn.functional.layer_norm( + x_360, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_360 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) + x_361 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_50 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_38 = torch.nn.functional.relu(x_361, inplace=False) + x_361 = None + pow_39 = relu_38**2 + relu_38 = None + mul_56 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + * pow_39 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( + pow_39 + ) = None + x_362 = ( + mul_56 + + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + ) + mul_56 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) + x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) + x_362 = None + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_365 = torch.nn.functional.dropout(x_364, 0.0, False, False) + x_364 = None + x_366 = mul_55 + x_365 + mul_55 = x_365 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_57 = x_366 * view_18 + view_18 = None + layer_norm_51 = torch.nn.functional.layer_norm( + x_366, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_51, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_51 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_19 = linear_36.reshape(1, pow_52, 3, 12, 32) + linear_36 = None + qkv_9 = reshape_19.permute(2, 0, 3, 1, 4) + reshape_19 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_367 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) + q_9 = k_9 = v_9 = None + transpose_10 = x_367.transpose(1, 2) + x_367 = None + x_368 = transpose_10.reshape(1, pow_52, 384) + transpose_10 = None + x_369 = torch._C._nn.linear( + x_368, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_368 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_370 = torch.nn.functional.dropout(x_369, 0.0, False, False) + x_369 = None + x_371 = mul_57 + x_370 + mul_57 = x_370 = None + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_58 = x_371 * view_19 + view_19 = None + layer_norm_52 = torch.nn.functional.layer_norm( + x_371, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_371 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) + x_372 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_52 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_39 = torch.nn.functional.relu(x_372, inplace=False) + x_372 = None + pow_40 = relu_39**2 + relu_39 = None + mul_59 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + * pow_40 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( + pow_40 + ) = None + x_373 = ( + mul_59 + + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + ) + mul_59 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) + x_374 = torch.nn.functional.dropout(x_373, 0.0, False, False) + x_373 = None + x_375 = torch._C._nn.linear( + x_374, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_374 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_376 = torch.nn.functional.dropout(x_375, 0.0, False, False) + x_375 = None + x_377 = mul_58 + x_376 + mul_58 = x_376 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_60 = x_377 * view_20 + view_20 = None + layer_norm_53 = torch.nn.functional.layer_norm( + x_377, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_377 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_53, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_53 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_21 = linear_40.reshape(1, pow_52, 3, 12, 32) + linear_40 = None + qkv_10 = reshape_21.permute(2, 0, 3, 1, 4) + reshape_21 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_378 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_378.transpose(1, 2) + x_378 = None + x_379 = transpose_11.reshape(1, pow_52, 384) + transpose_11 = None + x_380 = torch._C._nn.linear( + x_379, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_379 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_381 = torch.nn.functional.dropout(x_380, 0.0, False, False) + x_380 = None + x_382 = mul_60 + x_381 + mul_60 = x_381 = None + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_61 = x_382 * view_21 + view_21 = None + layer_norm_54 = torch.nn.functional.layer_norm( + x_382, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_382 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) + x_383 = torch._C._nn.linear( + layer_norm_54, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_54 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_40 = torch.nn.functional.relu(x_383, inplace=False) + x_383 = None + pow_41 = relu_40**2 + relu_40 = None + mul_62 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + * pow_41 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( + pow_41 + ) = None + x_384 = ( + mul_62 + + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + ) + mul_62 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) + x_385 = torch.nn.functional.dropout(x_384, 0.0, False, False) + x_384 = None + x_386 = torch._C._nn.linear( + x_385, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_385 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_387 = torch.nn.functional.dropout(x_386, 0.0, False, False) + x_386 = None + x_388 = mul_61 + x_387 + mul_61 = x_387 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_63 = x_388 * view_22 + view_22 = None + layer_norm_55 = torch.nn.functional.layer_norm( + x_388, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_388 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_55, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_55 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_23 = linear_44.reshape(1, pow_52, 3, 12, 32) + linear_44 = None + qkv_11 = reshape_23.permute(2, 0, 3, 1, 4) + reshape_23 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_389 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_389.transpose(1, 2) + x_389 = None + x_390 = transpose_12.reshape(1, pow_52, 384) + transpose_12 = None + x_391 = torch._C._nn.linear( + x_390, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_390 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_392 = torch.nn.functional.dropout(x_391, 0.0, False, False) + x_391 = None + x_393 = mul_63 + x_392 + mul_63 = x_392 = None + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_64 = x_393 * view_23 + view_23 = None + layer_norm_56 = torch.nn.functional.layer_norm( + x_393, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_393 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) + x_394 = torch._C._nn.linear( + layer_norm_56, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_56 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_41 = torch.nn.functional.relu(x_394, inplace=False) + x_394 = None + pow_42 = relu_41**2 + relu_41 = None + mul_65 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + * pow_42 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( + pow_42 + ) = None + x_395 = ( + mul_65 + + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + ) + mul_65 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) + x_396 = torch.nn.functional.dropout(x_395, 0.0, False, False) + x_395 = None + x_397 = torch._C._nn.linear( + x_396, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_396 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_398 = torch.nn.functional.dropout(x_397, 0.0, False, False) + x_397 = None + x_399 = mul_64 + x_398 + mul_64 = x_398 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_66 = x_399 * view_24 + view_24 = None + layer_norm_57 = torch.nn.functional.layer_norm( + x_399, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_399 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_57, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_57 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_25 = linear_48.reshape(1, pow_52, 3, 12, 32) + linear_48 = None + qkv_12 = reshape_25.permute(2, 0, 3, 1, 4) + reshape_25 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_12 = unbind_12[0] + k_12 = unbind_12[1] + v_12 = unbind_12[2] + unbind_12 = None + x_400 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_13 = x_400.transpose(1, 2) + x_400 = None + x_401 = transpose_13.reshape(1, pow_52, 384) + transpose_13 = None + x_402 = torch._C._nn.linear( + x_401, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_401 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_403 = torch.nn.functional.dropout(x_402, 0.0, False, False) + x_402 = None + x_404 = mul_66 + x_403 + mul_66 = x_403 = None + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_67 = x_404 * view_25 + view_25 = None + layer_norm_58 = torch.nn.functional.layer_norm( + x_404, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_404 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = (None) + x_405 = torch._C._nn.linear( + layer_norm_58, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_58 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_42 = torch.nn.functional.relu(x_405, inplace=False) + x_405 = None + pow_43 = relu_42**2 + relu_42 = None + mul_68 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ + * pow_43 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = ( + pow_43 + ) = None + x_406 = ( + mul_68 + + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ + ) + mul_68 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = (None) + x_407 = torch.nn.functional.dropout(x_406, 0.0, False, False) + x_406 = None + x_408 = torch._C._nn.linear( + x_407, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_407 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_409 = torch.nn.functional.dropout(x_408, 0.0, False, False) + x_408 = None + x_410 = mul_67 + x_409 + mul_67 = x_409 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_69 = x_410 * view_26 + view_26 = None + layer_norm_59 = torch.nn.functional.layer_norm( + x_410, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_410 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = (None) + linear_52 = torch._C._nn.linear( + layer_norm_59, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_59 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_27 = linear_52.reshape(1, pow_52, 3, 12, 32) + linear_52 = None + qkv_13 = reshape_27.permute(2, 0, 3, 1, 4) + reshape_27 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_13 = unbind_13[0] + k_13 = unbind_13[1] + v_13 = unbind_13[2] + unbind_13 = None + x_411 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_14 = x_411.transpose(1, 2) + x_411 = None + x_412 = transpose_14.reshape(1, pow_52, 384) + transpose_14 = None + x_413 = torch._C._nn.linear( + x_412, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_412 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_414 = torch.nn.functional.dropout(x_413, 0.0, False, False) + x_413 = None + x_415 = mul_69 + x_414 + mul_69 = x_414 = None + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_70 = x_415 * view_27 + view_27 = None + layer_norm_60 = torch.nn.functional.layer_norm( + x_415, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_415 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = (None) + x_416 = torch._C._nn.linear( + layer_norm_60, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_60 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_43 = torch.nn.functional.relu(x_416, inplace=False) + x_416 = None + pow_44 = relu_43**2 + relu_43 = None + mul_71 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ + * pow_44 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = ( + pow_44 + ) = None + x_417 = ( + mul_71 + + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ + ) + mul_71 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = (None) + x_418 = torch.nn.functional.dropout(x_417, 0.0, False, False) + x_417 = None + x_419 = torch._C._nn.linear( + x_418, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_418 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_420 = torch.nn.functional.dropout(x_419, 0.0, False, False) + x_419 = None + x_421 = mul_70 + x_420 + mul_70 = x_420 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_72 = x_421 * view_28 + view_28 = None + layer_norm_61 = torch.nn.functional.layer_norm( + x_421, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_421 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_61, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_61 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_29 = linear_56.reshape(1, pow_52, 3, 12, 32) + linear_56 = None + qkv_14 = reshape_29.permute(2, 0, 3, 1, 4) + reshape_29 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_14 = unbind_14[0] + k_14 = unbind_14[1] + v_14 = unbind_14[2] + unbind_14 = None + x_422 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_15 = x_422.transpose(1, 2) + x_422 = None + x_423 = transpose_15.reshape(1, pow_52, 384) + transpose_15 = None + x_424 = torch._C._nn.linear( + x_423, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_423 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_425 = torch.nn.functional.dropout(x_424, 0.0, False, False) + x_424 = None + x_426 = mul_72 + x_425 + mul_72 = x_425 = None + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_73 = x_426 * view_29 + view_29 = None + layer_norm_62 = torch.nn.functional.layer_norm( + x_426, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_426 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = (None) + x_427 = torch._C._nn.linear( + layer_norm_62, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_62 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_44 = torch.nn.functional.relu(x_427, inplace=False) + x_427 = None + pow_45 = relu_44**2 + relu_44 = None + mul_74 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ + * pow_45 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = ( + pow_45 + ) = None + x_428 = ( + mul_74 + + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ + ) + mul_74 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = (None) + x_429 = torch.nn.functional.dropout(x_428, 0.0, False, False) + x_428 = None + x_430 = torch._C._nn.linear( + x_429, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_429 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_431 = torch.nn.functional.dropout(x_430, 0.0, False, False) + x_430 = None + x_432 = mul_73 + x_431 + mul_73 = x_431 = None + view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_75 = x_432 * view_30 + view_30 = None + layer_norm_63 = torch.nn.functional.layer_norm( + x_432, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_432 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = (None) + linear_60 = torch._C._nn.linear( + layer_norm_63, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_63 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_31 = linear_60.reshape(1, pow_52, 3, 12, 32) + linear_60 = None + qkv_15 = reshape_31.permute(2, 0, 3, 1, 4) + reshape_31 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_15 = unbind_15[0] + k_15 = unbind_15[1] + v_15 = unbind_15[2] + unbind_15 = None + x_433 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_16 = x_433.transpose(1, 2) + x_433 = None + x_434 = transpose_16.reshape(1, pow_52, 384) + transpose_16 = None + x_435 = torch._C._nn.linear( + x_434, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_434 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_436 = torch.nn.functional.dropout(x_435, 0.0, False, False) + x_435 = None + x_437 = mul_75 + x_436 + mul_75 = x_436 = None + view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_76 = x_437 * view_31 + view_31 = None + layer_norm_64 = torch.nn.functional.layer_norm( + x_437, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_437 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = (None) + x_438 = torch._C._nn.linear( + layer_norm_64, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_64 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_45 = torch.nn.functional.relu(x_438, inplace=False) + x_438 = None + pow_46 = relu_45**2 + relu_45 = None + mul_77 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ + * pow_46 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = ( + pow_46 + ) = None + x_439 = ( + mul_77 + + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ + ) + mul_77 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = (None) + x_440 = torch.nn.functional.dropout(x_439, 0.0, False, False) + x_439 = None + x_441 = torch._C._nn.linear( + x_440, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_440 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_442 = torch.nn.functional.dropout(x_441, 0.0, False, False) + x_441 = None + x_443 = mul_76 + x_442 + mul_76 = x_442 = None + view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_78 = x_443 * view_32 + view_32 = None + layer_norm_65 = torch.nn.functional.layer_norm( + x_443, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_443 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_65, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_65 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_33 = linear_64.reshape(1, pow_52, 3, 12, 32) + linear_64 = None + qkv_16 = reshape_33.permute(2, 0, 3, 1, 4) + reshape_33 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_16 = unbind_16[0] + k_16 = unbind_16[1] + v_16 = unbind_16[2] + unbind_16 = None + x_444 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_17 = x_444.transpose(1, 2) + x_444 = None + x_445 = transpose_17.reshape(1, pow_52, 384) + transpose_17 = None + x_446 = torch._C._nn.linear( + x_445, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_445 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_447 = torch.nn.functional.dropout(x_446, 0.0, False, False) + x_446 = None + x_448 = mul_78 + x_447 + mul_78 = x_447 = None + view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_79 = x_448 * view_33 + view_33 = None + layer_norm_66 = torch.nn.functional.layer_norm( + x_448, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_448 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = (None) + x_449 = torch._C._nn.linear( + layer_norm_66, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_66 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_46 = torch.nn.functional.relu(x_449, inplace=False) + x_449 = None + pow_47 = relu_46**2 + relu_46 = None + mul_80 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ + * pow_47 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = ( + pow_47 + ) = None + x_450 = ( + mul_80 + + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ + ) + mul_80 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = (None) + x_451 = torch.nn.functional.dropout(x_450, 0.0, False, False) + x_450 = None + x_452 = torch._C._nn.linear( + x_451, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_451 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_453 = torch.nn.functional.dropout(x_452, 0.0, False, False) + x_452 = None + x_454 = mul_79 + x_453 + mul_79 = x_453 = None + view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_81 = x_454 * view_34 + view_34 = None + layer_norm_67 = torch.nn.functional.layer_norm( + x_454, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_454 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = (None) + linear_68 = torch._C._nn.linear( + layer_norm_67, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_67 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_35 = linear_68.reshape(1, pow_52, 3, 12, 32) + linear_68 = None + qkv_17 = reshape_35.permute(2, 0, 3, 1, 4) + reshape_35 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_17 = unbind_17[0] + k_17 = unbind_17[1] + v_17 = unbind_17[2] + unbind_17 = None + x_455 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_18 = x_455.transpose(1, 2) + x_455 = None + x_456 = transpose_18.reshape(1, pow_52, 384) + transpose_18 = pow_52 = None + x_457 = torch._C._nn.linear( + x_456, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_456 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_458 = torch.nn.functional.dropout(x_457, 0.0, False, False) + x_457 = None + x_459 = mul_81 + x_458 + mul_81 = x_458 = None + view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_82 = x_459 * view_35 + view_35 = None + layer_norm_68 = torch.nn.functional.layer_norm( + x_459, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_459 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = (None) + x_460 = torch._C._nn.linear( + layer_norm_68, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_68 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_47 = torch.nn.functional.relu(x_460, inplace=False) + x_460 = None + pow_48 = relu_47**2 + relu_47 = None + mul_83 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ + * pow_48 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = ( + pow_48 + ) = None + x_461 = ( + mul_83 + + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ + ) + mul_83 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = (None) + x_462 = torch.nn.functional.dropout(x_461, 0.0, False, False) + x_461 = None + x_463 = torch._C._nn.linear( + x_462, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_462 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_464 = torch.nn.functional.dropout(x_463, 0.0, False, False) + x_463 = None + x_465 = mul_82 + x_464 + mul_82 = x_464 = None + transpose_19 = x_465.transpose(1, 2) + x_465 = None + x_466 = transpose_19.reshape(1, 384, sym_sum_3, sym_sum_3) + transpose_19 = sym_sum_3 = None + x_467 = x_466.permute(0, 2, 3, 1) + x_466 = None + x_468 = torch.nn.functional.layer_norm( + x_467, + (384,), + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_467 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) + x_469 = x_468.permute(0, 3, 1, 2) + x_468 = None + x_470 = torch.conv2d( + x_469, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_469 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_3 = sym_sum // 32 + sym_sum = None + sym_sum_4 = torch.sym_sum([1, floordiv_3]) + floordiv_3 = None + reshape_38 = x_470.reshape(1, 576, -1) + x_470 = None + x_471 = reshape_38.transpose(1, 2) + reshape_38 = None + view_36 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_84 = x_471 * view_36 + view_36 = None + layer_norm_70 = torch.nn.functional.layer_norm( + x_471, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_471 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_53 = sym_sum_4**2 + linear_72 = torch._C._nn.linear( + layer_norm_70, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_70 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_39 = linear_72.reshape(1, pow_53, 3, 18, 32) + linear_72 = None + qkv_18 = reshape_39.permute(2, 0, 3, 1, 4) + reshape_39 = None + unbind_18 = qkv_18.unbind(0) + qkv_18 = None + q_18 = unbind_18[0] + k_18 = unbind_18[1] + v_18 = unbind_18[2] + unbind_18 = None + x_472 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_21 = x_472.transpose(1, 2) + x_472 = None + x_473 = transpose_21.reshape(1, pow_53, 576) + transpose_21 = None + x_474 = torch._C._nn.linear( + x_473, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_473 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_475 = torch.nn.functional.dropout(x_474, 0.0, False, False) + x_474 = None + x_476 = mul_84 + x_475 + mul_84 = x_475 = None + view_37 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_85 = x_476 * view_37 + view_37 = None + layer_norm_71 = torch.nn.functional.layer_norm( + x_476, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_476 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_477 = torch._C._nn.linear( + layer_norm_71, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_71 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_48 = torch.nn.functional.relu(x_477, inplace=False) + x_477 = None + pow_49 = relu_48**2 + relu_48 = None + mul_86 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_49 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_49 + ) = None + x_478 = ( + mul_86 + + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_86 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_479 = torch.nn.functional.dropout(x_478, 0.0, False, False) + x_478 = None + x_480 = torch._C._nn.linear( + x_479, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_479 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_481 = torch.nn.functional.dropout(x_480, 0.0, False, False) + x_480 = None + x_482 = mul_85 + x_481 + mul_85 = x_481 = None + view_38 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_87 = x_482 * view_38 + view_38 = None + layer_norm_72 = torch.nn.functional.layer_norm( + x_482, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_482 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_76 = torch._C._nn.linear( + layer_norm_72, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_72 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_41 = linear_76.reshape(1, pow_53, 3, 18, 32) + linear_76 = None + qkv_19 = reshape_41.permute(2, 0, 3, 1, 4) + reshape_41 = None + unbind_19 = qkv_19.unbind(0) + qkv_19 = None + q_19 = unbind_19[0] + k_19 = unbind_19[1] + v_19 = unbind_19[2] + unbind_19 = None + x_483 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_22 = x_483.transpose(1, 2) + x_483 = None + x_484 = transpose_22.reshape(1, pow_53, 576) + transpose_22 = None + x_485 = torch._C._nn.linear( + x_484, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_484 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_486 = torch.nn.functional.dropout(x_485, 0.0, False, False) + x_485 = None + x_487 = mul_87 + x_486 + mul_87 = x_486 = None + view_39 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_88 = x_487 * view_39 + view_39 = None + layer_norm_73 = torch.nn.functional.layer_norm( + x_487, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_487 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_488 = torch._C._nn.linear( + layer_norm_73, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_73 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_49 = torch.nn.functional.relu(x_488, inplace=False) + x_488 = None + pow_50 = relu_49**2 + relu_49 = None + mul_89 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_50 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_50 + ) = None + x_489 = ( + mul_89 + + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_89 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_490 = torch.nn.functional.dropout(x_489, 0.0, False, False) + x_489 = None + x_491 = torch._C._nn.linear( + x_490, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_490 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_492 = torch.nn.functional.dropout(x_491, 0.0, False, False) + x_491 = None + x_493 = mul_88 + x_492 + mul_88 = x_492 = None + view_40 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_90 = x_493 * view_40 + view_40 = None + layer_norm_74 = torch.nn.functional.layer_norm( + x_493, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_493 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_74, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_74 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_43 = linear_80.reshape(1, pow_53, 3, 18, 32) + linear_80 = None + qkv_20 = reshape_43.permute(2, 0, 3, 1, 4) + reshape_43 = None + unbind_20 = qkv_20.unbind(0) + qkv_20 = None + q_20 = unbind_20[0] + k_20 = unbind_20[1] + v_20 = unbind_20[2] + unbind_20 = None + x_494 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_23 = x_494.transpose(1, 2) + x_494 = None + x_495 = transpose_23.reshape(1, pow_53, 576) + transpose_23 = pow_53 = None + x_496 = torch._C._nn.linear( + x_495, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_495 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_497 = torch.nn.functional.dropout(x_496, 0.0, False, False) + x_496 = None + x_498 = mul_90 + x_497 + mul_90 = x_497 = None + view_41 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_91 = x_498 * view_41 + view_41 = None + layer_norm_75 = torch.nn.functional.layer_norm( + x_498, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_498 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_499 = torch._C._nn.linear( + layer_norm_75, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_75 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_50 = torch.nn.functional.relu(x_499, inplace=False) + x_499 = None + pow_51 = relu_50**2 + relu_50 = None + mul_92 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_51 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_51 + ) = None + x_500 = ( + mul_92 + + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_92 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_501 = torch.nn.functional.dropout(x_500, 0.0, False, False) + x_500 = None + x_502 = torch._C._nn.linear( + x_501, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_501 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_503 = torch.nn.functional.dropout(x_502, 0.0, False, False) + x_502 = None + x_504 = mul_91 + x_503 + mul_91 = x_503 = None + transpose_24 = x_504.transpose(1, 2) + x_504 = None + x_505 = transpose_24.reshape(1, 576, sym_sum_4, sym_sum_4) + transpose_24 = sym_sum_4 = None + x_506 = torch.nn.functional.adaptive_avg_pool2d(x_505, 1) + x_505 = None + x_507 = x_506.permute(0, 2, 3, 1) + x_506 = None + x_508 = torch.nn.functional.layer_norm( + x_507, + (576,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_507 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_509 = x_508.permute(0, 3, 1, 2) + x_508 = None + x_510 = x_509.flatten(1, -1) + x_509 = None + x_511 = torch.nn.functional.dropout(x_510, 0.0, False, False) + x_510 = None + x_512 = torch._C._nn.linear( + x_511, + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, + ) + x_511 = ( + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None + relu_51 = torch.nn.functional.relu(x_512, inplace=False) + x_512 = None + x_513 = torch.square(relu_51) + relu_51 = None + x_514 = torch.nn.functional.layer_norm( + x_513, + (2304,), + l_self_modules_head_modules_fc_modules_norm_parameters_weight_, + l_self_modules_head_modules_fc_modules_norm_parameters_bias_, + 1e-06, + ) + x_513 = ( + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None + x_515 = torch.nn.functional.dropout(x_514, 0.0, False, False) + x_514 = None + x_516 = torch._C._nn.linear( + x_515, + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, + ) + x_515 = ( + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None + return (x_516,) diff --git a/samples/timm/caformer_m36.sail_in1k/weight_meta.py b/samples/timm/caformer_m36.sail_in1k/weight_meta.py new file mode 100644 index 0000000000..a2c0914908 --- /dev/null +++ b/samples/timm/caformer_m36.sail_in1k/weight_meta.py @@ -0,0 +1,3968 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv_parameters_weight_" + shape = [96, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: + name = "L_self_modules_stem_modules_conv_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.223 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: + name = "L_self_modules_stem_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [192, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [96, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [192, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [96, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [192, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [96, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" + shape = [192, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" + shape = [384, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" + shape = [576, 384, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1728, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [576, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2304, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [576, 2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1728, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [576, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2304, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [576, 2304] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1728, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [576, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2304, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [576, 2304] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" + shape = [2304, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" + shape = [1000, 2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/caformer_m36.sail_in1k_384/graph_net.json b/samples/timm/caformer_m36.sail_in1k_384/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/caformer_m36.sail_in1k_384/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/caformer_m36.sail_in1k_384/input_meta.py b/samples/timm/caformer_m36.sail_in1k_384/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_m36.sail_in1k_384/input_tensor_constraints.py b/samples/timm/caformer_m36.sail_in1k_384/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_m36.sail_in1k_384/model.py b/samples/timm/caformer_m36.sail_in1k_384/model.py new file mode 100644 index 0000000000..dcce286c5c --- /dev/null +++ b/samples/timm/caformer_m36.sail_in1k_384/model.py @@ -0,0 +1,4754 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv_parameters_bias_ = ( + L_self_modules_stem_modules_conv_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_norm_parameters_weight_ = ( + L_self_modules_stem_modules_norm_parameters_weight_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv_parameters_weight_, + l_self_modules_stem_modules_conv_parameters_bias_, + (4, 4), + (2, 2), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_conv_parameters_weight_ + ) = l_self_modules_stem_modules_conv_parameters_bias_ = None + x_1 = x.permute(0, 2, 3, 1) + x = None + x_2 = torch.nn.functional.layer_norm( + x_1, (96,), l_self_modules_stem_modules_norm_parameters_weight_, None, 1e-06 + ) + x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None + x_3 = x_2.permute(0, 3, 1, 2) + x_2 = None + sym_sum = torch.sym_sum([-3, s1]) + s1 = None + floordiv = sym_sum // 4 + sym_sum_1 = torch.sym_sum([1, floordiv]) + floordiv = sym_sum_1 = None + x_4 = x_3.permute(0, 2, 3, 1) + x_5 = torch.nn.functional.layer_norm( + x_4, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu = torch.nn.functional.relu(x_7, inplace=False) + x_7 = None + pow_1 = relu**2 + relu = None + mul = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_1 + ) = None + x_8 = ( + mul + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_11 = x_3 + x_10 + x_3 = x_10 = None + x_12 = x_11.permute(0, 2, 3, 1) + x_13 = torch.nn.functional.layer_norm( + x_12, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_14 = x_13.permute(0, 3, 1, 2) + x_13 = None + x_15 = torch.conv2d( + x_14, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_1 = torch.nn.functional.relu(x_15, inplace=False) + x_15 = None + pow_2 = relu_1**2 + relu_1 = None + mul_1 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_2 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_2 + ) = None + x_16 = ( + mul_1 + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = x_11 + x_19 + x_11 = x_19 = None + x_21 = x_20.permute(0, 2, 3, 1) + x_22 = torch.nn.functional.layer_norm( + x_21, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = torch.conv2d( + x_23, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_2 = torch.nn.functional.relu(x_24, inplace=False) + x_24 = None + pow_3 = relu_2**2 + relu_2 = None + mul_2 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_3 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_3 + ) = None + x_25 = ( + mul_2 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_27 = torch.conv2d( + x_26, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_28 = x_20 + x_27 + x_20 = x_27 = None + x_29 = x_28.permute(0, 2, 3, 1) + x_30 = torch.nn.functional.layer_norm( + x_29, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_3 = torch.nn.functional.relu(x_32, inplace=False) + x_32 = None + pow_4 = relu_3**2 + relu_3 = None + mul_3 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_4 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_4 + ) = None + x_33 = ( + mul_3 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + x_37 = x_28 + x_36 + x_28 = x_36 = None + x_38 = x_37.permute(0, 2, 3, 1) + x_39 = torch.nn.functional.layer_norm( + x_38, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_40 = x_39.permute(0, 3, 1, 2) + x_39 = None + x_41 = torch.conv2d( + x_40, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_4 = torch.nn.functional.relu(x_41, inplace=False) + x_41 = None + pow_5 = relu_4**2 + relu_4 = None + mul_4 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_5 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_5 + ) = None + x_42 = ( + mul_4 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_44 = torch.conv2d( + x_43, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_45 = x_37 + x_44 + x_37 = x_44 = None + x_46 = x_45.permute(0, 2, 3, 1) + x_47 = torch.nn.functional.layer_norm( + x_46, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_5 = torch.nn.functional.relu(x_49, inplace=False) + x_49 = None + pow_6 = relu_5**2 + relu_5 = None + mul_5 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_6 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_6 + ) = None + x_50 = ( + mul_5 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = x_45 + x_53 + x_45 = x_53 = None + x_55 = x_54.permute(0, 2, 3, 1) + x_54 = None + x_56 = torch.nn.functional.layer_norm( + x_55, + (96,), + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) + x_57 = x_56.permute(0, 3, 1, 2) + x_56 = None + x_58 = torch.conv2d( + x_57, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_1 = sym_sum // 8 + sym_sum_2 = torch.sym_sum([1, floordiv_1]) + floordiv_1 = sym_sum_2 = None + x_59 = x_58.permute(0, 2, 3, 1) + x_60 = torch.nn.functional.layer_norm( + x_59, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_61 = x_60.permute(0, 3, 1, 2) + x_60 = None + x_62 = torch.conv2d( + x_61, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_6 = torch.nn.functional.relu(x_62, inplace=False) + x_62 = None + pow_7 = relu_6**2 + relu_6 = None + mul_6 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_7 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_7 + ) = None + x_63 = ( + mul_6 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_66 = x_58 + x_65 + x_58 = x_65 = None + x_67 = x_66.permute(0, 2, 3, 1) + x_68 = torch.nn.functional.layer_norm( + x_67, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + x_70 = torch.conv2d( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_7 = torch.nn.functional.relu(x_70, inplace=False) + x_70 = None + pow_8 = relu_7**2 + relu_7 = None + mul_7 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_8 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_8 + ) = None + x_71 = ( + mul_7 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + x_73 = torch.conv2d( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + x_75 = x_66 + x_74 + x_66 = x_74 = None + x_76 = x_75.permute(0, 2, 3, 1) + x_77 = torch.nn.functional.layer_norm( + x_76, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_8 = torch.nn.functional.relu(x_79, inplace=False) + x_79 = None + pow_9 = relu_8**2 + relu_8 = None + mul_8 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_9 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_9 + ) = None + x_80 = ( + mul_8 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_81 = torch.conv2d( + x_80, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_83 = x_75 + x_82 + x_75 = x_82 = None + x_84 = x_83.permute(0, 2, 3, 1) + x_85 = torch.nn.functional.layer_norm( + x_84, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_86 = x_85.permute(0, 3, 1, 2) + x_85 = None + x_87 = torch.conv2d( + x_86, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_9 = torch.nn.functional.relu(x_87, inplace=False) + x_87 = None + pow_10 = relu_9**2 + relu_9 = None + mul_9 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_10 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_10 + ) = None + x_88 = ( + mul_9 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = x_83 + x_91 + x_83 = x_91 = None + x_93 = x_92.permute(0, 2, 3, 1) + x_94 = torch.nn.functional.layer_norm( + x_93, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = torch.conv2d( + x_95, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_10 = torch.nn.functional.relu(x_96, inplace=False) + x_96 = None + pow_11 = relu_10**2 + relu_10 = None + mul_10 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_11 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_11 + ) = None + x_97 = ( + mul_10 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_98 = torch.conv2d( + x_97, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_99 = torch.conv2d( + x_98, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_100 = x_92 + x_99 + x_92 = x_99 = None + x_101 = x_100.permute(0, 2, 3, 1) + x_102 = torch.nn.functional.layer_norm( + x_101, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_103 = x_102.permute(0, 3, 1, 2) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_11 = torch.nn.functional.relu(x_104, inplace=False) + x_104 = None + pow_12 = relu_11**2 + relu_11 = None + mul_11 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_12 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_12 + ) = None + x_105 = ( + mul_11 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_100 + x_108 + x_100 = x_108 = None + x_110 = x_109.permute(0, 2, 3, 1) + x_111 = torch.nn.functional.layer_norm( + x_110, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_110 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) + x_112 = x_111.permute(0, 3, 1, 2) + x_111 = None + x_113 = torch.conv2d( + x_112, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_112 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_12 = torch.nn.functional.relu(x_113, inplace=False) + x_113 = None + pow_13 = relu_12**2 + relu_12 = None + mul_12 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ + * pow_13 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_13 + ) = None + x_114 = ( + mul_12 + + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_12 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_115 = torch.conv2d( + x_114, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_114 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_117 = x_109 + x_116 + x_109 = x_116 = None + x_118 = x_117.permute(0, 2, 3, 1) + x_119 = torch.nn.functional.layer_norm( + x_118, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_118 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) + x_120 = x_119.permute(0, 3, 1, 2) + x_119 = None + x_121 = torch.conv2d( + x_120, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_120 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_13 = torch.nn.functional.relu(x_121, inplace=False) + x_121 = None + pow_14 = relu_13**2 + relu_13 = None + mul_13 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + * pow_14 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( + pow_14 + ) = None + x_122 = ( + mul_13 + + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + ) + mul_13 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) + x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) + x_122 = None + x_124 = torch.conv2d( + x_123, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_123 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_126 = x_117 + x_125 + x_117 = x_125 = None + x_127 = x_126.permute(0, 2, 3, 1) + x_128 = torch.nn.functional.layer_norm( + x_127, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_127 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) + x_129 = x_128.permute(0, 3, 1, 2) + x_128 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_14 = torch.nn.functional.relu(x_130, inplace=False) + x_130 = None + pow_15 = relu_14**2 + relu_14 = None + mul_14 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ + * pow_15 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_15 + ) = None + x_131 = ( + mul_14 + + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_14 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_132 = torch.conv2d( + x_131, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_131 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_132 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_134 = x_126 + x_133 + x_126 = x_133 = None + x_135 = x_134.permute(0, 2, 3, 1) + x_136 = torch.nn.functional.layer_norm( + x_135, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_135 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) + x_137 = x_136.permute(0, 3, 1, 2) + x_136 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_15 = torch.nn.functional.relu(x_138, inplace=False) + x_138 = None + pow_16 = relu_15**2 + relu_15 = None + mul_15 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + * pow_16 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( + pow_16 + ) = None + x_139 = ( + mul_15 + + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + ) + mul_15 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) + x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) + x_139 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_140 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_134 + x_142 + x_134 = x_142 = None + x_144 = x_143.permute(0, 2, 3, 1) + x_145 = torch.nn.functional.layer_norm( + x_144, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_144 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) + x_146 = x_145.permute(0, 3, 1, 2) + x_145 = None + x_147 = torch.conv2d( + x_146, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_146 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_16 = torch.nn.functional.relu(x_147, inplace=False) + x_147 = None + pow_17 = relu_16**2 + relu_16 = None + mul_16 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ + * pow_17 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_17 + ) = None + x_148 = ( + mul_16 + + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_16 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_149 = torch.conv2d( + x_148, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_148 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_150 = torch.conv2d( + x_149, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_149 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_151 = x_143 + x_150 + x_143 = x_150 = None + x_152 = x_151.permute(0, 2, 3, 1) + x_153 = torch.nn.functional.layer_norm( + x_152, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_152 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) + x_154 = x_153.permute(0, 3, 1, 2) + x_153 = None + x_155 = torch.conv2d( + x_154, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_154 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_17 = torch.nn.functional.relu(x_155, inplace=False) + x_155 = None + pow_18 = relu_17**2 + relu_17 = None + mul_17 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + * pow_18 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( + pow_18 + ) = None + x_156 = ( + mul_17 + + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + ) + mul_17 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch.conv2d( + x_157, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_157 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_151 + x_159 + x_151 = x_159 = None + x_161 = x_160.permute(0, 2, 3, 1) + x_162 = torch.nn.functional.layer_norm( + x_161, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_161 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) + x_163 = x_162.permute(0, 3, 1, 2) + x_162 = None + x_164 = torch.conv2d( + x_163, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_163 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_18 = torch.nn.functional.relu(x_164, inplace=False) + x_164 = None + pow_19 = relu_18**2 + relu_18 = None + mul_18 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ + * pow_19 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_19 + ) = None + x_165 = ( + mul_18 + + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_18 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_166 = torch.conv2d( + x_165, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_165 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_167 = torch.conv2d( + x_166, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_166 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_168 = x_160 + x_167 + x_160 = x_167 = None + x_169 = x_168.permute(0, 2, 3, 1) + x_170 = torch.nn.functional.layer_norm( + x_169, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_169 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) + x_171 = x_170.permute(0, 3, 1, 2) + x_170 = None + x_172 = torch.conv2d( + x_171, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_171 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_19 = torch.nn.functional.relu(x_172, inplace=False) + x_172 = None + pow_20 = relu_19**2 + relu_19 = None + mul_19 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + * pow_20 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( + pow_20 + ) = None + x_173 = ( + mul_19 + + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + ) + mul_19 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) + x_174 = torch.nn.functional.dropout(x_173, 0.0, False, False) + x_173 = None + x_175 = torch.conv2d( + x_174, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_174 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = x_168 + x_176 + x_168 = x_176 = None + x_178 = x_177.permute(0, 2, 3, 1) + x_179 = torch.nn.functional.layer_norm( + x_178, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_178 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) + x_180 = x_179.permute(0, 3, 1, 2) + x_179 = None + x_181 = torch.conv2d( + x_180, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_180 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_20 = torch.nn.functional.relu(x_181, inplace=False) + x_181 = None + pow_21 = relu_20**2 + relu_20 = None + mul_20 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ + * pow_21 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_21 + ) = None + x_182 = ( + mul_20 + + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_20 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_183 = torch.conv2d( + x_182, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_182 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_184 = torch.conv2d( + x_183, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_183 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_185 = x_177 + x_184 + x_177 = x_184 = None + x_186 = x_185.permute(0, 2, 3, 1) + x_187 = torch.nn.functional.layer_norm( + x_186, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) + x_188 = x_187.permute(0, 3, 1, 2) + x_187 = None + x_189 = torch.conv2d( + x_188, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_188 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_21 = torch.nn.functional.relu(x_189, inplace=False) + x_189 = None + pow_22 = relu_21**2 + relu_21 = None + mul_21 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + * pow_22 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( + pow_22 + ) = None + x_190 = ( + mul_21 + + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + ) + mul_21 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) + x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) + x_190 = None + x_192 = torch.conv2d( + x_191, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_191 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_185 + x_193 + x_185 = x_193 = None + x_195 = x_194.permute(0, 2, 3, 1) + x_196 = torch.nn.functional.layer_norm( + x_195, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_195 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) + x_197 = x_196.permute(0, 3, 1, 2) + x_196 = None + x_198 = torch.conv2d( + x_197, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_197 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_22 = torch.nn.functional.relu(x_198, inplace=False) + x_198 = None + pow_23 = relu_22**2 + relu_22 = None + mul_22 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ + * pow_23 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_23 + ) = None + x_199 = ( + mul_22 + + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_22 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_200 = torch.conv2d( + x_199, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_199 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_201 = torch.conv2d( + x_200, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_200 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_202 = x_194 + x_201 + x_194 = x_201 = None + x_203 = x_202.permute(0, 2, 3, 1) + x_204 = torch.nn.functional.layer_norm( + x_203, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_203 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) + x_205 = x_204.permute(0, 3, 1, 2) + x_204 = None + x_206 = torch.conv2d( + x_205, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_205 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_23 = torch.nn.functional.relu(x_206, inplace=False) + x_206 = None + pow_24 = relu_23**2 + relu_23 = None + mul_23 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + * pow_24 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( + pow_24 + ) = None + x_207 = ( + mul_23 + + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + ) + mul_23 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) + x_208 = torch.nn.functional.dropout(x_207, 0.0, False, False) + x_207 = None + x_209 = torch.conv2d( + x_208, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_208 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) + x_209 = None + x_211 = x_202 + x_210 + x_202 = x_210 = None + x_212 = x_211.permute(0, 2, 3, 1) + x_213 = torch.nn.functional.layer_norm( + x_212, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_212 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) + x_214 = x_213.permute(0, 3, 1, 2) + x_213 = None + x_215 = torch.conv2d( + x_214, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_214 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_24 = torch.nn.functional.relu(x_215, inplace=False) + x_215 = None + pow_25 = relu_24**2 + relu_24 = None + mul_24 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ + * pow_25 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_25 + ) = None + x_216 = ( + mul_24 + + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_24 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_217 = torch.conv2d( + x_216, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_216 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_217 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_219 = x_211 + x_218 + x_211 = x_218 = None + x_220 = x_219.permute(0, 2, 3, 1) + x_221 = torch.nn.functional.layer_norm( + x_220, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_220 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) + x_222 = x_221.permute(0, 3, 1, 2) + x_221 = None + x_223 = torch.conv2d( + x_222, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_222 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_25 = torch.nn.functional.relu(x_223, inplace=False) + x_223 = None + pow_26 = relu_25**2 + relu_25 = None + mul_25 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + * pow_26 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( + pow_26 + ) = None + x_224 = ( + mul_25 + + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + ) + mul_25 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = torch.conv2d( + x_225, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_225 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) + x_226 = None + x_228 = x_219 + x_227 + x_219 = x_227 = None + x_229 = x_228.permute(0, 2, 3, 1) + x_230 = torch.nn.functional.layer_norm( + x_229, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_229 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) + x_231 = x_230.permute(0, 3, 1, 2) + x_230 = None + x_232 = torch.conv2d( + x_231, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_231 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_26 = torch.nn.functional.relu(x_232, inplace=False) + x_232 = None + pow_27 = relu_26**2 + relu_26 = None + mul_26 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ + * pow_27 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_27 + ) = None + x_233 = ( + mul_26 + + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_26 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_234 = torch.conv2d( + x_233, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_233 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_235 = torch.conv2d( + x_234, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_234 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_236 = x_228 + x_235 + x_228 = x_235 = None + x_237 = x_236.permute(0, 2, 3, 1) + x_238 = torch.nn.functional.layer_norm( + x_237, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_237 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) + x_239 = x_238.permute(0, 3, 1, 2) + x_238 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_239 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_27 = torch.nn.functional.relu(x_240, inplace=False) + x_240 = None + pow_28 = relu_27**2 + relu_27 = None + mul_27 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + * pow_28 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( + pow_28 + ) = None + x_241 = ( + mul_27 + + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + ) + mul_27 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) + x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) + x_241 = None + x_243 = torch.conv2d( + x_242, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_242 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_244 = torch.nn.functional.dropout(x_243, 0.0, False, False) + x_243 = None + x_245 = x_236 + x_244 + x_236 = x_244 = None + x_246 = x_245.permute(0, 2, 3, 1) + x_247 = torch.nn.functional.layer_norm( + x_246, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_246 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) + x_248 = x_247.permute(0, 3, 1, 2) + x_247 = None + x_249 = torch.conv2d( + x_248, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_248 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_28 = torch.nn.functional.relu(x_249, inplace=False) + x_249 = None + pow_29 = relu_28**2 + relu_28 = None + mul_28 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ + * pow_29 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_29 + ) = None + x_250 = ( + mul_28 + + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_28 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_250 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_252 = torch.conv2d( + x_251, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_251 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_253 = x_245 + x_252 + x_245 = x_252 = None + x_254 = x_253.permute(0, 2, 3, 1) + x_255 = torch.nn.functional.layer_norm( + x_254, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_254 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) + x_256 = x_255.permute(0, 3, 1, 2) + x_255 = None + x_257 = torch.conv2d( + x_256, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_256 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_29 = torch.nn.functional.relu(x_257, inplace=False) + x_257 = None + pow_30 = relu_29**2 + relu_29 = None + mul_29 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + * pow_30 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( + pow_30 + ) = None + x_258 = ( + mul_29 + + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + ) + mul_29 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = torch.conv2d( + x_259, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_259 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_261 = torch.nn.functional.dropout(x_260, 0.0, False, False) + x_260 = None + x_262 = x_253 + x_261 + x_253 = x_261 = None + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (192,), + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) + x_265 = x_264.permute(0, 3, 1, 2) + x_264 = None + x_266 = torch.conv2d( + x_265, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_265 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_2 = sym_sum // 16 + sym_sum_3 = torch.sym_sum([1, floordiv_2]) + floordiv_2 = None + reshape = x_266.reshape(1, 384, -1) + x_266 = None + x_267 = reshape.transpose(1, 2) + reshape = None + view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_30 = x_267 * view + view = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_267, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_52 = sym_sum_3**2 + linear = torch._C._nn.linear( + layer_norm_33, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_33 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_1 = linear.reshape(1, pow_52, 3, 12, 32) + linear = None + qkv = reshape_1.permute(2, 0, 3, 1, 4) + reshape_1 = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_268 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) + q = k = v = None + transpose_1 = x_268.transpose(1, 2) + x_268 = None + x_269 = transpose_1.reshape(1, pow_52, 384) + transpose_1 = None + x_270 = torch._C._nn.linear( + x_269, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_269 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_271 = torch.nn.functional.dropout(x_270, 0.0, False, False) + x_270 = None + x_272 = mul_30 + x_271 + mul_30 = x_271 = None + view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_31 = x_272 * view_1 + view_1 = None + layer_norm_34 = torch.nn.functional.layer_norm( + x_272, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_272 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_273 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_34 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_30 = torch.nn.functional.relu(x_273, inplace=False) + x_273 = None + pow_31 = relu_30**2 + relu_30 = None + mul_32 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_31 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_31 + ) = None + x_274 = ( + mul_32 + + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_275 = torch.nn.functional.dropout(x_274, 0.0, False, False) + x_274 = None + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_277 = torch.nn.functional.dropout(x_276, 0.0, False, False) + x_276 = None + x_278 = mul_31 + x_277 + mul_31 = x_277 = None + view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_33 = x_278 * view_2 + view_2 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_278, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_3 = linear_4.reshape(1, pow_52, 3, 12, 32) + linear_4 = None + qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) + reshape_3 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_279 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) + q_1 = k_1 = v_1 = None + transpose_2 = x_279.transpose(1, 2) + x_279 = None + x_280 = transpose_2.reshape(1, pow_52, 384) + transpose_2 = None + x_281 = torch._C._nn.linear( + x_280, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_280 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) + x_281 = None + x_283 = mul_33 + x_282 + mul_33 = x_282 = None + view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_34 = x_283 * view_3 + view_3 = None + layer_norm_36 = torch.nn.functional.layer_norm( + x_283, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_283 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_284 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_36 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_31 = torch.nn.functional.relu(x_284, inplace=False) + x_284 = None + pow_32 = relu_31**2 + relu_31 = None + mul_35 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_32 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_32 + ) = None + x_285 = ( + mul_35 + + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_286 = torch.nn.functional.dropout(x_285, 0.0, False, False) + x_285 = None + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) + x_287 = None + x_289 = mul_34 + x_288 + mul_34 = x_288 = None + view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_36 = x_289 * view_4 + view_4 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_289, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_37 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_5 = linear_8.reshape(1, pow_52, 3, 12, 32) + linear_8 = None + qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) + reshape_5 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_290 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) + q_2 = k_2 = v_2 = None + transpose_3 = x_290.transpose(1, 2) + x_290 = None + x_291 = transpose_3.reshape(1, pow_52, 384) + transpose_3 = None + x_292 = torch._C._nn.linear( + x_291, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_291 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_293 = torch.nn.functional.dropout(x_292, 0.0, False, False) + x_292 = None + x_294 = mul_36 + x_293 + mul_36 = x_293 = None + view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_37 = x_294 * view_5 + view_5 = None + layer_norm_38 = torch.nn.functional.layer_norm( + x_294, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_294 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_295 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_32 = torch.nn.functional.relu(x_295, inplace=False) + x_295 = None + pow_33 = relu_32**2 + relu_32 = None + mul_38 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_33 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_33 + ) = None + x_296 = ( + mul_38 + + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_297 = torch.nn.functional.dropout(x_296, 0.0, False, False) + x_296 = None + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_299 = torch.nn.functional.dropout(x_298, 0.0, False, False) + x_298 = None + x_300 = mul_37 + x_299 + mul_37 = x_299 = None + view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_39 = x_300 * view_6 + view_6 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_300, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_39 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_7 = linear_12.reshape(1, pow_52, 3, 12, 32) + linear_12 = None + qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) + reshape_7 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_301 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) + q_3 = k_3 = v_3 = None + transpose_4 = x_301.transpose(1, 2) + x_301 = None + x_302 = transpose_4.reshape(1, pow_52, 384) + transpose_4 = None + x_303 = torch._C._nn.linear( + x_302, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_302 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_304 = torch.nn.functional.dropout(x_303, 0.0, False, False) + x_303 = None + x_305 = mul_39 + x_304 + mul_39 = x_304 = None + view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_40 = x_305 * view_7 + view_7 = None + layer_norm_40 = torch.nn.functional.layer_norm( + x_305, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_305 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) + x_306 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_40 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_33 = torch.nn.functional.relu(x_306, inplace=False) + x_306 = None + pow_34 = relu_33**2 + relu_33 = None + mul_41 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + * pow_34 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( + pow_34 + ) = None + x_307 = ( + mul_41 + + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + ) + mul_41 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) + x_308 = torch.nn.functional.dropout(x_307, 0.0, False, False) + x_307 = None + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) + x_309 = None + x_311 = mul_40 + x_310 + mul_40 = x_310 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_42 = x_311 * view_8 + view_8 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_311, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_41 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_9 = linear_16.reshape(1, pow_52, 3, 12, 32) + linear_16 = None + qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) + reshape_9 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_312 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) + q_4 = k_4 = v_4 = None + transpose_5 = x_312.transpose(1, 2) + x_312 = None + x_313 = transpose_5.reshape(1, pow_52, 384) + transpose_5 = None + x_314 = torch._C._nn.linear( + x_313, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_313 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) + x_314 = None + x_316 = mul_42 + x_315 + mul_42 = x_315 = None + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_43 = x_316 * view_9 + view_9 = None + layer_norm_42 = torch.nn.functional.layer_norm( + x_316, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_316 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) + x_317 = torch._C._nn.linear( + layer_norm_42, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_42 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_34 = torch.nn.functional.relu(x_317, inplace=False) + x_317 = None + pow_35 = relu_34**2 + relu_34 = None + mul_44 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + * pow_35 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( + pow_35 + ) = None + x_318 = ( + mul_44 + + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + ) + mul_44 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) + x_319 = torch.nn.functional.dropout(x_318, 0.0, False, False) + x_318 = None + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_321 = torch.nn.functional.dropout(x_320, 0.0, False, False) + x_320 = None + x_322 = mul_43 + x_321 + mul_43 = x_321 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_45 = x_322 * view_10 + view_10 = None + layer_norm_43 = torch.nn.functional.layer_norm( + x_322, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) + linear_20 = torch._C._nn.linear( + layer_norm_43, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_43 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_11 = linear_20.reshape(1, pow_52, 3, 12, 32) + linear_20 = None + qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) + reshape_11 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_323 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) + q_5 = k_5 = v_5 = None + transpose_6 = x_323.transpose(1, 2) + x_323 = None + x_324 = transpose_6.reshape(1, pow_52, 384) + transpose_6 = None + x_325 = torch._C._nn.linear( + x_324, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_324 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_326 = torch.nn.functional.dropout(x_325, 0.0, False, False) + x_325 = None + x_327 = mul_45 + x_326 + mul_45 = x_326 = None + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_46 = x_327 * view_11 + view_11 = None + layer_norm_44 = torch.nn.functional.layer_norm( + x_327, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_327 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) + x_328 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_44 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_35 = torch.nn.functional.relu(x_328, inplace=False) + x_328 = None + pow_36 = relu_35**2 + relu_35 = None + mul_47 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + * pow_36 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( + pow_36 + ) = None + x_329 = ( + mul_47 + + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + ) + mul_47 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) + x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) + x_329 = None + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_332 = torch.nn.functional.dropout(x_331, 0.0, False, False) + x_331 = None + x_333 = mul_46 + x_332 + mul_46 = x_332 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_48 = x_333 * view_12 + view_12 = None + layer_norm_45 = torch.nn.functional.layer_norm( + x_333, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) + linear_24 = torch._C._nn.linear( + layer_norm_45, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_45 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_13 = linear_24.reshape(1, pow_52, 3, 12, 32) + linear_24 = None + qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) + reshape_13 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_334 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) + q_6 = k_6 = v_6 = None + transpose_7 = x_334.transpose(1, 2) + x_334 = None + x_335 = transpose_7.reshape(1, pow_52, 384) + transpose_7 = None + x_336 = torch._C._nn.linear( + x_335, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_335 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_337 = torch.nn.functional.dropout(x_336, 0.0, False, False) + x_336 = None + x_338 = mul_48 + x_337 + mul_48 = x_337 = None + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_49 = x_338 * view_13 + view_13 = None + layer_norm_46 = torch.nn.functional.layer_norm( + x_338, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_338 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) + x_339 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_46 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_36 = torch.nn.functional.relu(x_339, inplace=False) + x_339 = None + pow_37 = relu_36**2 + relu_36 = None + mul_50 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + * pow_37 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( + pow_37 + ) = None + x_340 = ( + mul_50 + + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + ) + mul_50 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) + x_341 = torch.nn.functional.dropout(x_340, 0.0, False, False) + x_340 = None + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_343 = torch.nn.functional.dropout(x_342, 0.0, False, False) + x_342 = None + x_344 = mul_49 + x_343 + mul_49 = x_343 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_51 = x_344 * view_14 + view_14 = None + layer_norm_47 = torch.nn.functional.layer_norm( + x_344, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) + linear_28 = torch._C._nn.linear( + layer_norm_47, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_47 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_15 = linear_28.reshape(1, pow_52, 3, 12, 32) + linear_28 = None + qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) + reshape_15 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_345 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) + q_7 = k_7 = v_7 = None + transpose_8 = x_345.transpose(1, 2) + x_345 = None + x_346 = transpose_8.reshape(1, pow_52, 384) + transpose_8 = None + x_347 = torch._C._nn.linear( + x_346, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_346 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) + x_347 = None + x_349 = mul_51 + x_348 + mul_51 = x_348 = None + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_52 = x_349 * view_15 + view_15 = None + layer_norm_48 = torch.nn.functional.layer_norm( + x_349, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_349 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) + x_350 = torch._C._nn.linear( + layer_norm_48, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_48 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_37 = torch.nn.functional.relu(x_350, inplace=False) + x_350 = None + pow_38 = relu_37**2 + relu_37 = None + mul_53 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + * pow_38 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( + pow_38 + ) = None + x_351 = ( + mul_53 + + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + ) + mul_53 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) + x_352 = torch.nn.functional.dropout(x_351, 0.0, False, False) + x_351 = None + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_354 = torch.nn.functional.dropout(x_353, 0.0, False, False) + x_353 = None + x_355 = mul_52 + x_354 + mul_52 = x_354 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_54 = x_355 * view_16 + view_16 = None + layer_norm_49 = torch.nn.functional.layer_norm( + x_355, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) + linear_32 = torch._C._nn.linear( + layer_norm_49, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_49 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_17 = linear_32.reshape(1, pow_52, 3, 12, 32) + linear_32 = None + qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) + reshape_17 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_356 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) + q_8 = k_8 = v_8 = None + transpose_9 = x_356.transpose(1, 2) + x_356 = None + x_357 = transpose_9.reshape(1, pow_52, 384) + transpose_9 = None + x_358 = torch._C._nn.linear( + x_357, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_357 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_359 = torch.nn.functional.dropout(x_358, 0.0, False, False) + x_358 = None + x_360 = mul_54 + x_359 + mul_54 = x_359 = None + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_55 = x_360 * view_17 + view_17 = None + layer_norm_50 = torch.nn.functional.layer_norm( + x_360, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_360 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) + x_361 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_50 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_38 = torch.nn.functional.relu(x_361, inplace=False) + x_361 = None + pow_39 = relu_38**2 + relu_38 = None + mul_56 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + * pow_39 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( + pow_39 + ) = None + x_362 = ( + mul_56 + + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + ) + mul_56 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) + x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) + x_362 = None + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_365 = torch.nn.functional.dropout(x_364, 0.0, False, False) + x_364 = None + x_366 = mul_55 + x_365 + mul_55 = x_365 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_57 = x_366 * view_18 + view_18 = None + layer_norm_51 = torch.nn.functional.layer_norm( + x_366, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_51, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_51 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_19 = linear_36.reshape(1, pow_52, 3, 12, 32) + linear_36 = None + qkv_9 = reshape_19.permute(2, 0, 3, 1, 4) + reshape_19 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_367 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) + q_9 = k_9 = v_9 = None + transpose_10 = x_367.transpose(1, 2) + x_367 = None + x_368 = transpose_10.reshape(1, pow_52, 384) + transpose_10 = None + x_369 = torch._C._nn.linear( + x_368, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_368 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_370 = torch.nn.functional.dropout(x_369, 0.0, False, False) + x_369 = None + x_371 = mul_57 + x_370 + mul_57 = x_370 = None + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_58 = x_371 * view_19 + view_19 = None + layer_norm_52 = torch.nn.functional.layer_norm( + x_371, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_371 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) + x_372 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_52 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_39 = torch.nn.functional.relu(x_372, inplace=False) + x_372 = None + pow_40 = relu_39**2 + relu_39 = None + mul_59 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + * pow_40 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( + pow_40 + ) = None + x_373 = ( + mul_59 + + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + ) + mul_59 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) + x_374 = torch.nn.functional.dropout(x_373, 0.0, False, False) + x_373 = None + x_375 = torch._C._nn.linear( + x_374, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_374 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_376 = torch.nn.functional.dropout(x_375, 0.0, False, False) + x_375 = None + x_377 = mul_58 + x_376 + mul_58 = x_376 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_60 = x_377 * view_20 + view_20 = None + layer_norm_53 = torch.nn.functional.layer_norm( + x_377, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_377 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_53, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_53 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_21 = linear_40.reshape(1, pow_52, 3, 12, 32) + linear_40 = None + qkv_10 = reshape_21.permute(2, 0, 3, 1, 4) + reshape_21 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_378 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_378.transpose(1, 2) + x_378 = None + x_379 = transpose_11.reshape(1, pow_52, 384) + transpose_11 = None + x_380 = torch._C._nn.linear( + x_379, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_379 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_381 = torch.nn.functional.dropout(x_380, 0.0, False, False) + x_380 = None + x_382 = mul_60 + x_381 + mul_60 = x_381 = None + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_61 = x_382 * view_21 + view_21 = None + layer_norm_54 = torch.nn.functional.layer_norm( + x_382, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_382 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) + x_383 = torch._C._nn.linear( + layer_norm_54, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_54 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_40 = torch.nn.functional.relu(x_383, inplace=False) + x_383 = None + pow_41 = relu_40**2 + relu_40 = None + mul_62 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + * pow_41 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( + pow_41 + ) = None + x_384 = ( + mul_62 + + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + ) + mul_62 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) + x_385 = torch.nn.functional.dropout(x_384, 0.0, False, False) + x_384 = None + x_386 = torch._C._nn.linear( + x_385, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_385 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_387 = torch.nn.functional.dropout(x_386, 0.0, False, False) + x_386 = None + x_388 = mul_61 + x_387 + mul_61 = x_387 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_63 = x_388 * view_22 + view_22 = None + layer_norm_55 = torch.nn.functional.layer_norm( + x_388, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_388 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_55, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_55 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_23 = linear_44.reshape(1, pow_52, 3, 12, 32) + linear_44 = None + qkv_11 = reshape_23.permute(2, 0, 3, 1, 4) + reshape_23 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_389 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_389.transpose(1, 2) + x_389 = None + x_390 = transpose_12.reshape(1, pow_52, 384) + transpose_12 = None + x_391 = torch._C._nn.linear( + x_390, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_390 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_392 = torch.nn.functional.dropout(x_391, 0.0, False, False) + x_391 = None + x_393 = mul_63 + x_392 + mul_63 = x_392 = None + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_64 = x_393 * view_23 + view_23 = None + layer_norm_56 = torch.nn.functional.layer_norm( + x_393, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_393 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) + x_394 = torch._C._nn.linear( + layer_norm_56, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_56 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_41 = torch.nn.functional.relu(x_394, inplace=False) + x_394 = None + pow_42 = relu_41**2 + relu_41 = None + mul_65 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + * pow_42 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( + pow_42 + ) = None + x_395 = ( + mul_65 + + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + ) + mul_65 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) + x_396 = torch.nn.functional.dropout(x_395, 0.0, False, False) + x_395 = None + x_397 = torch._C._nn.linear( + x_396, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_396 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_398 = torch.nn.functional.dropout(x_397, 0.0, False, False) + x_397 = None + x_399 = mul_64 + x_398 + mul_64 = x_398 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_66 = x_399 * view_24 + view_24 = None + layer_norm_57 = torch.nn.functional.layer_norm( + x_399, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_399 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_57, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_57 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_25 = linear_48.reshape(1, pow_52, 3, 12, 32) + linear_48 = None + qkv_12 = reshape_25.permute(2, 0, 3, 1, 4) + reshape_25 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_12 = unbind_12[0] + k_12 = unbind_12[1] + v_12 = unbind_12[2] + unbind_12 = None + x_400 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_13 = x_400.transpose(1, 2) + x_400 = None + x_401 = transpose_13.reshape(1, pow_52, 384) + transpose_13 = None + x_402 = torch._C._nn.linear( + x_401, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_401 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_403 = torch.nn.functional.dropout(x_402, 0.0, False, False) + x_402 = None + x_404 = mul_66 + x_403 + mul_66 = x_403 = None + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_67 = x_404 * view_25 + view_25 = None + layer_norm_58 = torch.nn.functional.layer_norm( + x_404, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_404 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = (None) + x_405 = torch._C._nn.linear( + layer_norm_58, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_58 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_42 = torch.nn.functional.relu(x_405, inplace=False) + x_405 = None + pow_43 = relu_42**2 + relu_42 = None + mul_68 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ + * pow_43 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = ( + pow_43 + ) = None + x_406 = ( + mul_68 + + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ + ) + mul_68 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = (None) + x_407 = torch.nn.functional.dropout(x_406, 0.0, False, False) + x_406 = None + x_408 = torch._C._nn.linear( + x_407, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_407 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_409 = torch.nn.functional.dropout(x_408, 0.0, False, False) + x_408 = None + x_410 = mul_67 + x_409 + mul_67 = x_409 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_69 = x_410 * view_26 + view_26 = None + layer_norm_59 = torch.nn.functional.layer_norm( + x_410, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_410 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = (None) + linear_52 = torch._C._nn.linear( + layer_norm_59, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_59 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_27 = linear_52.reshape(1, pow_52, 3, 12, 32) + linear_52 = None + qkv_13 = reshape_27.permute(2, 0, 3, 1, 4) + reshape_27 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_13 = unbind_13[0] + k_13 = unbind_13[1] + v_13 = unbind_13[2] + unbind_13 = None + x_411 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_14 = x_411.transpose(1, 2) + x_411 = None + x_412 = transpose_14.reshape(1, pow_52, 384) + transpose_14 = None + x_413 = torch._C._nn.linear( + x_412, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_412 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_414 = torch.nn.functional.dropout(x_413, 0.0, False, False) + x_413 = None + x_415 = mul_69 + x_414 + mul_69 = x_414 = None + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_70 = x_415 * view_27 + view_27 = None + layer_norm_60 = torch.nn.functional.layer_norm( + x_415, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_415 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = (None) + x_416 = torch._C._nn.linear( + layer_norm_60, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_60 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_43 = torch.nn.functional.relu(x_416, inplace=False) + x_416 = None + pow_44 = relu_43**2 + relu_43 = None + mul_71 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ + * pow_44 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = ( + pow_44 + ) = None + x_417 = ( + mul_71 + + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ + ) + mul_71 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = (None) + x_418 = torch.nn.functional.dropout(x_417, 0.0, False, False) + x_417 = None + x_419 = torch._C._nn.linear( + x_418, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_418 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_420 = torch.nn.functional.dropout(x_419, 0.0, False, False) + x_419 = None + x_421 = mul_70 + x_420 + mul_70 = x_420 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_72 = x_421 * view_28 + view_28 = None + layer_norm_61 = torch.nn.functional.layer_norm( + x_421, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_421 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_61, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_61 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_29 = linear_56.reshape(1, pow_52, 3, 12, 32) + linear_56 = None + qkv_14 = reshape_29.permute(2, 0, 3, 1, 4) + reshape_29 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_14 = unbind_14[0] + k_14 = unbind_14[1] + v_14 = unbind_14[2] + unbind_14 = None + x_422 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_15 = x_422.transpose(1, 2) + x_422 = None + x_423 = transpose_15.reshape(1, pow_52, 384) + transpose_15 = None + x_424 = torch._C._nn.linear( + x_423, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_423 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_425 = torch.nn.functional.dropout(x_424, 0.0, False, False) + x_424 = None + x_426 = mul_72 + x_425 + mul_72 = x_425 = None + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_73 = x_426 * view_29 + view_29 = None + layer_norm_62 = torch.nn.functional.layer_norm( + x_426, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_426 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = (None) + x_427 = torch._C._nn.linear( + layer_norm_62, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_62 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_44 = torch.nn.functional.relu(x_427, inplace=False) + x_427 = None + pow_45 = relu_44**2 + relu_44 = None + mul_74 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ + * pow_45 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = ( + pow_45 + ) = None + x_428 = ( + mul_74 + + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ + ) + mul_74 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = (None) + x_429 = torch.nn.functional.dropout(x_428, 0.0, False, False) + x_428 = None + x_430 = torch._C._nn.linear( + x_429, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_429 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_431 = torch.nn.functional.dropout(x_430, 0.0, False, False) + x_430 = None + x_432 = mul_73 + x_431 + mul_73 = x_431 = None + view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_75 = x_432 * view_30 + view_30 = None + layer_norm_63 = torch.nn.functional.layer_norm( + x_432, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_432 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = (None) + linear_60 = torch._C._nn.linear( + layer_norm_63, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_63 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_31 = linear_60.reshape(1, pow_52, 3, 12, 32) + linear_60 = None + qkv_15 = reshape_31.permute(2, 0, 3, 1, 4) + reshape_31 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_15 = unbind_15[0] + k_15 = unbind_15[1] + v_15 = unbind_15[2] + unbind_15 = None + x_433 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_16 = x_433.transpose(1, 2) + x_433 = None + x_434 = transpose_16.reshape(1, pow_52, 384) + transpose_16 = None + x_435 = torch._C._nn.linear( + x_434, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_434 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_436 = torch.nn.functional.dropout(x_435, 0.0, False, False) + x_435 = None + x_437 = mul_75 + x_436 + mul_75 = x_436 = None + view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_76 = x_437 * view_31 + view_31 = None + layer_norm_64 = torch.nn.functional.layer_norm( + x_437, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_437 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = (None) + x_438 = torch._C._nn.linear( + layer_norm_64, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_64 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_45 = torch.nn.functional.relu(x_438, inplace=False) + x_438 = None + pow_46 = relu_45**2 + relu_45 = None + mul_77 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ + * pow_46 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = ( + pow_46 + ) = None + x_439 = ( + mul_77 + + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ + ) + mul_77 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = (None) + x_440 = torch.nn.functional.dropout(x_439, 0.0, False, False) + x_439 = None + x_441 = torch._C._nn.linear( + x_440, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_440 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_442 = torch.nn.functional.dropout(x_441, 0.0, False, False) + x_441 = None + x_443 = mul_76 + x_442 + mul_76 = x_442 = None + view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_78 = x_443 * view_32 + view_32 = None + layer_norm_65 = torch.nn.functional.layer_norm( + x_443, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_443 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_65, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_65 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_33 = linear_64.reshape(1, pow_52, 3, 12, 32) + linear_64 = None + qkv_16 = reshape_33.permute(2, 0, 3, 1, 4) + reshape_33 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_16 = unbind_16[0] + k_16 = unbind_16[1] + v_16 = unbind_16[2] + unbind_16 = None + x_444 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_17 = x_444.transpose(1, 2) + x_444 = None + x_445 = transpose_17.reshape(1, pow_52, 384) + transpose_17 = None + x_446 = torch._C._nn.linear( + x_445, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_445 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_447 = torch.nn.functional.dropout(x_446, 0.0, False, False) + x_446 = None + x_448 = mul_78 + x_447 + mul_78 = x_447 = None + view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_79 = x_448 * view_33 + view_33 = None + layer_norm_66 = torch.nn.functional.layer_norm( + x_448, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_448 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = (None) + x_449 = torch._C._nn.linear( + layer_norm_66, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_66 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_46 = torch.nn.functional.relu(x_449, inplace=False) + x_449 = None + pow_47 = relu_46**2 + relu_46 = None + mul_80 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ + * pow_47 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = ( + pow_47 + ) = None + x_450 = ( + mul_80 + + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ + ) + mul_80 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = (None) + x_451 = torch.nn.functional.dropout(x_450, 0.0, False, False) + x_450 = None + x_452 = torch._C._nn.linear( + x_451, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_451 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_453 = torch.nn.functional.dropout(x_452, 0.0, False, False) + x_452 = None + x_454 = mul_79 + x_453 + mul_79 = x_453 = None + view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_81 = x_454 * view_34 + view_34 = None + layer_norm_67 = torch.nn.functional.layer_norm( + x_454, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_454 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = (None) + linear_68 = torch._C._nn.linear( + layer_norm_67, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_67 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_35 = linear_68.reshape(1, pow_52, 3, 12, 32) + linear_68 = None + qkv_17 = reshape_35.permute(2, 0, 3, 1, 4) + reshape_35 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_17 = unbind_17[0] + k_17 = unbind_17[1] + v_17 = unbind_17[2] + unbind_17 = None + x_455 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_18 = x_455.transpose(1, 2) + x_455 = None + x_456 = transpose_18.reshape(1, pow_52, 384) + transpose_18 = pow_52 = None + x_457 = torch._C._nn.linear( + x_456, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_456 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_458 = torch.nn.functional.dropout(x_457, 0.0, False, False) + x_457 = None + x_459 = mul_81 + x_458 + mul_81 = x_458 = None + view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_82 = x_459 * view_35 + view_35 = None + layer_norm_68 = torch.nn.functional.layer_norm( + x_459, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_459 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = (None) + x_460 = torch._C._nn.linear( + layer_norm_68, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_68 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_47 = torch.nn.functional.relu(x_460, inplace=False) + x_460 = None + pow_48 = relu_47**2 + relu_47 = None + mul_83 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ + * pow_48 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = ( + pow_48 + ) = None + x_461 = ( + mul_83 + + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ + ) + mul_83 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = (None) + x_462 = torch.nn.functional.dropout(x_461, 0.0, False, False) + x_461 = None + x_463 = torch._C._nn.linear( + x_462, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_462 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_464 = torch.nn.functional.dropout(x_463, 0.0, False, False) + x_463 = None + x_465 = mul_82 + x_464 + mul_82 = x_464 = None + transpose_19 = x_465.transpose(1, 2) + x_465 = None + x_466 = transpose_19.reshape(1, 384, sym_sum_3, sym_sum_3) + transpose_19 = sym_sum_3 = None + x_467 = x_466.permute(0, 2, 3, 1) + x_466 = None + x_468 = torch.nn.functional.layer_norm( + x_467, + (384,), + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_467 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) + x_469 = x_468.permute(0, 3, 1, 2) + x_468 = None + x_470 = torch.conv2d( + x_469, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_469 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_3 = sym_sum // 32 + sym_sum = None + sym_sum_4 = torch.sym_sum([1, floordiv_3]) + floordiv_3 = None + reshape_38 = x_470.reshape(1, 576, -1) + x_470 = None + x_471 = reshape_38.transpose(1, 2) + reshape_38 = None + view_36 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_84 = x_471 * view_36 + view_36 = None + layer_norm_70 = torch.nn.functional.layer_norm( + x_471, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_471 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_53 = sym_sum_4**2 + linear_72 = torch._C._nn.linear( + layer_norm_70, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_70 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_39 = linear_72.reshape(1, pow_53, 3, 18, 32) + linear_72 = None + qkv_18 = reshape_39.permute(2, 0, 3, 1, 4) + reshape_39 = None + unbind_18 = qkv_18.unbind(0) + qkv_18 = None + q_18 = unbind_18[0] + k_18 = unbind_18[1] + v_18 = unbind_18[2] + unbind_18 = None + x_472 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_21 = x_472.transpose(1, 2) + x_472 = None + x_473 = transpose_21.reshape(1, pow_53, 576) + transpose_21 = None + x_474 = torch._C._nn.linear( + x_473, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_473 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_475 = torch.nn.functional.dropout(x_474, 0.0, False, False) + x_474 = None + x_476 = mul_84 + x_475 + mul_84 = x_475 = None + view_37 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_85 = x_476 * view_37 + view_37 = None + layer_norm_71 = torch.nn.functional.layer_norm( + x_476, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_476 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_477 = torch._C._nn.linear( + layer_norm_71, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_71 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_48 = torch.nn.functional.relu(x_477, inplace=False) + x_477 = None + pow_49 = relu_48**2 + relu_48 = None + mul_86 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_49 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_49 + ) = None + x_478 = ( + mul_86 + + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_86 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_479 = torch.nn.functional.dropout(x_478, 0.0, False, False) + x_478 = None + x_480 = torch._C._nn.linear( + x_479, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_479 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_481 = torch.nn.functional.dropout(x_480, 0.0, False, False) + x_480 = None + x_482 = mul_85 + x_481 + mul_85 = x_481 = None + view_38 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_87 = x_482 * view_38 + view_38 = None + layer_norm_72 = torch.nn.functional.layer_norm( + x_482, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_482 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_76 = torch._C._nn.linear( + layer_norm_72, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_72 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_41 = linear_76.reshape(1, pow_53, 3, 18, 32) + linear_76 = None + qkv_19 = reshape_41.permute(2, 0, 3, 1, 4) + reshape_41 = None + unbind_19 = qkv_19.unbind(0) + qkv_19 = None + q_19 = unbind_19[0] + k_19 = unbind_19[1] + v_19 = unbind_19[2] + unbind_19 = None + x_483 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_22 = x_483.transpose(1, 2) + x_483 = None + x_484 = transpose_22.reshape(1, pow_53, 576) + transpose_22 = None + x_485 = torch._C._nn.linear( + x_484, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_484 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_486 = torch.nn.functional.dropout(x_485, 0.0, False, False) + x_485 = None + x_487 = mul_87 + x_486 + mul_87 = x_486 = None + view_39 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_88 = x_487 * view_39 + view_39 = None + layer_norm_73 = torch.nn.functional.layer_norm( + x_487, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_487 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_488 = torch._C._nn.linear( + layer_norm_73, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_73 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_49 = torch.nn.functional.relu(x_488, inplace=False) + x_488 = None + pow_50 = relu_49**2 + relu_49 = None + mul_89 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_50 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_50 + ) = None + x_489 = ( + mul_89 + + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_89 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_490 = torch.nn.functional.dropout(x_489, 0.0, False, False) + x_489 = None + x_491 = torch._C._nn.linear( + x_490, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_490 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_492 = torch.nn.functional.dropout(x_491, 0.0, False, False) + x_491 = None + x_493 = mul_88 + x_492 + mul_88 = x_492 = None + view_40 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_90 = x_493 * view_40 + view_40 = None + layer_norm_74 = torch.nn.functional.layer_norm( + x_493, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_493 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_74, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_74 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_43 = linear_80.reshape(1, pow_53, 3, 18, 32) + linear_80 = None + qkv_20 = reshape_43.permute(2, 0, 3, 1, 4) + reshape_43 = None + unbind_20 = qkv_20.unbind(0) + qkv_20 = None + q_20 = unbind_20[0] + k_20 = unbind_20[1] + v_20 = unbind_20[2] + unbind_20 = None + x_494 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_23 = x_494.transpose(1, 2) + x_494 = None + x_495 = transpose_23.reshape(1, pow_53, 576) + transpose_23 = pow_53 = None + x_496 = torch._C._nn.linear( + x_495, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_495 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_497 = torch.nn.functional.dropout(x_496, 0.0, False, False) + x_496 = None + x_498 = mul_90 + x_497 + mul_90 = x_497 = None + view_41 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_91 = x_498 * view_41 + view_41 = None + layer_norm_75 = torch.nn.functional.layer_norm( + x_498, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_498 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_499 = torch._C._nn.linear( + layer_norm_75, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_75 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_50 = torch.nn.functional.relu(x_499, inplace=False) + x_499 = None + pow_51 = relu_50**2 + relu_50 = None + mul_92 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_51 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_51 + ) = None + x_500 = ( + mul_92 + + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_92 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_501 = torch.nn.functional.dropout(x_500, 0.0, False, False) + x_500 = None + x_502 = torch._C._nn.linear( + x_501, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_501 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_503 = torch.nn.functional.dropout(x_502, 0.0, False, False) + x_502 = None + x_504 = mul_91 + x_503 + mul_91 = x_503 = None + transpose_24 = x_504.transpose(1, 2) + x_504 = None + x_505 = transpose_24.reshape(1, 576, sym_sum_4, sym_sum_4) + transpose_24 = sym_sum_4 = None + x_506 = torch.nn.functional.adaptive_avg_pool2d(x_505, 1) + x_505 = None + x_507 = x_506.permute(0, 2, 3, 1) + x_506 = None + x_508 = torch.nn.functional.layer_norm( + x_507, + (576,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_507 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_509 = x_508.permute(0, 3, 1, 2) + x_508 = None + x_510 = x_509.flatten(1, -1) + x_509 = None + x_511 = torch.nn.functional.dropout(x_510, 0.0, False, False) + x_510 = None + x_512 = torch._C._nn.linear( + x_511, + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, + ) + x_511 = ( + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None + relu_51 = torch.nn.functional.relu(x_512, inplace=False) + x_512 = None + x_513 = torch.square(relu_51) + relu_51 = None + x_514 = torch.nn.functional.layer_norm( + x_513, + (2304,), + l_self_modules_head_modules_fc_modules_norm_parameters_weight_, + l_self_modules_head_modules_fc_modules_norm_parameters_bias_, + 1e-06, + ) + x_513 = ( + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None + x_515 = torch.nn.functional.dropout(x_514, 0.0, False, False) + x_514 = None + x_516 = torch._C._nn.linear( + x_515, + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, + ) + x_515 = ( + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None + return (x_516,) diff --git a/samples/timm/caformer_m36.sail_in1k_384/weight_meta.py b/samples/timm/caformer_m36.sail_in1k_384/weight_meta.py new file mode 100644 index 0000000000..1a948dec5d --- /dev/null +++ b/samples/timm/caformer_m36.sail_in1k_384/weight_meta.py @@ -0,0 +1,3968 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv_parameters_weight_" + shape = [96, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: + name = "L_self_modules_stem_modules_conv_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.229 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: + name = "L_self_modules_stem_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [192, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [96, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [192, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [96, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [192, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [96, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" + shape = [192, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" + shape = [384, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" + shape = [576, 384, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1728, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [576, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2304, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [576, 2304] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1728, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [576, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2304, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [576, 2304] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1728, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [576, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2304, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [576, 2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" + shape = [2304, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" + shape = [1000, 2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/caformer_m36.sail_in22k/graph_net.json b/samples/timm/caformer_m36.sail_in22k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/caformer_m36.sail_in22k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/caformer_m36.sail_in22k/input_meta.py b/samples/timm/caformer_m36.sail_in22k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_m36.sail_in22k/input_tensor_constraints.py b/samples/timm/caformer_m36.sail_in22k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_m36.sail_in22k/model.py b/samples/timm/caformer_m36.sail_in22k/model.py new file mode 100644 index 0000000000..dcce286c5c --- /dev/null +++ b/samples/timm/caformer_m36.sail_in22k/model.py @@ -0,0 +1,4754 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv_parameters_bias_ = ( + L_self_modules_stem_modules_conv_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_norm_parameters_weight_ = ( + L_self_modules_stem_modules_norm_parameters_weight_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv_parameters_weight_, + l_self_modules_stem_modules_conv_parameters_bias_, + (4, 4), + (2, 2), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_conv_parameters_weight_ + ) = l_self_modules_stem_modules_conv_parameters_bias_ = None + x_1 = x.permute(0, 2, 3, 1) + x = None + x_2 = torch.nn.functional.layer_norm( + x_1, (96,), l_self_modules_stem_modules_norm_parameters_weight_, None, 1e-06 + ) + x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None + x_3 = x_2.permute(0, 3, 1, 2) + x_2 = None + sym_sum = torch.sym_sum([-3, s1]) + s1 = None + floordiv = sym_sum // 4 + sym_sum_1 = torch.sym_sum([1, floordiv]) + floordiv = sym_sum_1 = None + x_4 = x_3.permute(0, 2, 3, 1) + x_5 = torch.nn.functional.layer_norm( + x_4, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu = torch.nn.functional.relu(x_7, inplace=False) + x_7 = None + pow_1 = relu**2 + relu = None + mul = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_1 + ) = None + x_8 = ( + mul + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_11 = x_3 + x_10 + x_3 = x_10 = None + x_12 = x_11.permute(0, 2, 3, 1) + x_13 = torch.nn.functional.layer_norm( + x_12, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_14 = x_13.permute(0, 3, 1, 2) + x_13 = None + x_15 = torch.conv2d( + x_14, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_1 = torch.nn.functional.relu(x_15, inplace=False) + x_15 = None + pow_2 = relu_1**2 + relu_1 = None + mul_1 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_2 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_2 + ) = None + x_16 = ( + mul_1 + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = x_11 + x_19 + x_11 = x_19 = None + x_21 = x_20.permute(0, 2, 3, 1) + x_22 = torch.nn.functional.layer_norm( + x_21, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = torch.conv2d( + x_23, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_2 = torch.nn.functional.relu(x_24, inplace=False) + x_24 = None + pow_3 = relu_2**2 + relu_2 = None + mul_2 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_3 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_3 + ) = None + x_25 = ( + mul_2 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_27 = torch.conv2d( + x_26, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_28 = x_20 + x_27 + x_20 = x_27 = None + x_29 = x_28.permute(0, 2, 3, 1) + x_30 = torch.nn.functional.layer_norm( + x_29, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_3 = torch.nn.functional.relu(x_32, inplace=False) + x_32 = None + pow_4 = relu_3**2 + relu_3 = None + mul_3 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_4 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_4 + ) = None + x_33 = ( + mul_3 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + x_37 = x_28 + x_36 + x_28 = x_36 = None + x_38 = x_37.permute(0, 2, 3, 1) + x_39 = torch.nn.functional.layer_norm( + x_38, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_40 = x_39.permute(0, 3, 1, 2) + x_39 = None + x_41 = torch.conv2d( + x_40, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_4 = torch.nn.functional.relu(x_41, inplace=False) + x_41 = None + pow_5 = relu_4**2 + relu_4 = None + mul_4 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_5 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_5 + ) = None + x_42 = ( + mul_4 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_44 = torch.conv2d( + x_43, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_45 = x_37 + x_44 + x_37 = x_44 = None + x_46 = x_45.permute(0, 2, 3, 1) + x_47 = torch.nn.functional.layer_norm( + x_46, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_5 = torch.nn.functional.relu(x_49, inplace=False) + x_49 = None + pow_6 = relu_5**2 + relu_5 = None + mul_5 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_6 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_6 + ) = None + x_50 = ( + mul_5 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = x_45 + x_53 + x_45 = x_53 = None + x_55 = x_54.permute(0, 2, 3, 1) + x_54 = None + x_56 = torch.nn.functional.layer_norm( + x_55, + (96,), + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) + x_57 = x_56.permute(0, 3, 1, 2) + x_56 = None + x_58 = torch.conv2d( + x_57, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_1 = sym_sum // 8 + sym_sum_2 = torch.sym_sum([1, floordiv_1]) + floordiv_1 = sym_sum_2 = None + x_59 = x_58.permute(0, 2, 3, 1) + x_60 = torch.nn.functional.layer_norm( + x_59, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_61 = x_60.permute(0, 3, 1, 2) + x_60 = None + x_62 = torch.conv2d( + x_61, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_6 = torch.nn.functional.relu(x_62, inplace=False) + x_62 = None + pow_7 = relu_6**2 + relu_6 = None + mul_6 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_7 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_7 + ) = None + x_63 = ( + mul_6 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_66 = x_58 + x_65 + x_58 = x_65 = None + x_67 = x_66.permute(0, 2, 3, 1) + x_68 = torch.nn.functional.layer_norm( + x_67, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + x_70 = torch.conv2d( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_7 = torch.nn.functional.relu(x_70, inplace=False) + x_70 = None + pow_8 = relu_7**2 + relu_7 = None + mul_7 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_8 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_8 + ) = None + x_71 = ( + mul_7 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + x_73 = torch.conv2d( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + x_75 = x_66 + x_74 + x_66 = x_74 = None + x_76 = x_75.permute(0, 2, 3, 1) + x_77 = torch.nn.functional.layer_norm( + x_76, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_8 = torch.nn.functional.relu(x_79, inplace=False) + x_79 = None + pow_9 = relu_8**2 + relu_8 = None + mul_8 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_9 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_9 + ) = None + x_80 = ( + mul_8 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_81 = torch.conv2d( + x_80, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_83 = x_75 + x_82 + x_75 = x_82 = None + x_84 = x_83.permute(0, 2, 3, 1) + x_85 = torch.nn.functional.layer_norm( + x_84, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_86 = x_85.permute(0, 3, 1, 2) + x_85 = None + x_87 = torch.conv2d( + x_86, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_9 = torch.nn.functional.relu(x_87, inplace=False) + x_87 = None + pow_10 = relu_9**2 + relu_9 = None + mul_9 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_10 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_10 + ) = None + x_88 = ( + mul_9 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = x_83 + x_91 + x_83 = x_91 = None + x_93 = x_92.permute(0, 2, 3, 1) + x_94 = torch.nn.functional.layer_norm( + x_93, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = torch.conv2d( + x_95, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_10 = torch.nn.functional.relu(x_96, inplace=False) + x_96 = None + pow_11 = relu_10**2 + relu_10 = None + mul_10 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_11 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_11 + ) = None + x_97 = ( + mul_10 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_98 = torch.conv2d( + x_97, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_99 = torch.conv2d( + x_98, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_100 = x_92 + x_99 + x_92 = x_99 = None + x_101 = x_100.permute(0, 2, 3, 1) + x_102 = torch.nn.functional.layer_norm( + x_101, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_103 = x_102.permute(0, 3, 1, 2) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_11 = torch.nn.functional.relu(x_104, inplace=False) + x_104 = None + pow_12 = relu_11**2 + relu_11 = None + mul_11 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_12 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_12 + ) = None + x_105 = ( + mul_11 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_100 + x_108 + x_100 = x_108 = None + x_110 = x_109.permute(0, 2, 3, 1) + x_111 = torch.nn.functional.layer_norm( + x_110, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_110 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) + x_112 = x_111.permute(0, 3, 1, 2) + x_111 = None + x_113 = torch.conv2d( + x_112, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_112 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_12 = torch.nn.functional.relu(x_113, inplace=False) + x_113 = None + pow_13 = relu_12**2 + relu_12 = None + mul_12 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ + * pow_13 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_13 + ) = None + x_114 = ( + mul_12 + + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_12 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_115 = torch.conv2d( + x_114, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_114 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_117 = x_109 + x_116 + x_109 = x_116 = None + x_118 = x_117.permute(0, 2, 3, 1) + x_119 = torch.nn.functional.layer_norm( + x_118, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_118 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) + x_120 = x_119.permute(0, 3, 1, 2) + x_119 = None + x_121 = torch.conv2d( + x_120, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_120 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_13 = torch.nn.functional.relu(x_121, inplace=False) + x_121 = None + pow_14 = relu_13**2 + relu_13 = None + mul_13 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + * pow_14 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( + pow_14 + ) = None + x_122 = ( + mul_13 + + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + ) + mul_13 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) + x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) + x_122 = None + x_124 = torch.conv2d( + x_123, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_123 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_126 = x_117 + x_125 + x_117 = x_125 = None + x_127 = x_126.permute(0, 2, 3, 1) + x_128 = torch.nn.functional.layer_norm( + x_127, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_127 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) + x_129 = x_128.permute(0, 3, 1, 2) + x_128 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_14 = torch.nn.functional.relu(x_130, inplace=False) + x_130 = None + pow_15 = relu_14**2 + relu_14 = None + mul_14 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ + * pow_15 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_15 + ) = None + x_131 = ( + mul_14 + + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_14 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_132 = torch.conv2d( + x_131, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_131 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_132 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_134 = x_126 + x_133 + x_126 = x_133 = None + x_135 = x_134.permute(0, 2, 3, 1) + x_136 = torch.nn.functional.layer_norm( + x_135, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_135 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) + x_137 = x_136.permute(0, 3, 1, 2) + x_136 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_15 = torch.nn.functional.relu(x_138, inplace=False) + x_138 = None + pow_16 = relu_15**2 + relu_15 = None + mul_15 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + * pow_16 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( + pow_16 + ) = None + x_139 = ( + mul_15 + + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + ) + mul_15 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) + x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) + x_139 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_140 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_134 + x_142 + x_134 = x_142 = None + x_144 = x_143.permute(0, 2, 3, 1) + x_145 = torch.nn.functional.layer_norm( + x_144, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_144 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) + x_146 = x_145.permute(0, 3, 1, 2) + x_145 = None + x_147 = torch.conv2d( + x_146, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_146 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_16 = torch.nn.functional.relu(x_147, inplace=False) + x_147 = None + pow_17 = relu_16**2 + relu_16 = None + mul_16 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ + * pow_17 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_17 + ) = None + x_148 = ( + mul_16 + + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_16 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_149 = torch.conv2d( + x_148, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_148 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_150 = torch.conv2d( + x_149, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_149 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_151 = x_143 + x_150 + x_143 = x_150 = None + x_152 = x_151.permute(0, 2, 3, 1) + x_153 = torch.nn.functional.layer_norm( + x_152, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_152 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) + x_154 = x_153.permute(0, 3, 1, 2) + x_153 = None + x_155 = torch.conv2d( + x_154, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_154 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_17 = torch.nn.functional.relu(x_155, inplace=False) + x_155 = None + pow_18 = relu_17**2 + relu_17 = None + mul_17 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + * pow_18 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( + pow_18 + ) = None + x_156 = ( + mul_17 + + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + ) + mul_17 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch.conv2d( + x_157, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_157 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_151 + x_159 + x_151 = x_159 = None + x_161 = x_160.permute(0, 2, 3, 1) + x_162 = torch.nn.functional.layer_norm( + x_161, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_161 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) + x_163 = x_162.permute(0, 3, 1, 2) + x_162 = None + x_164 = torch.conv2d( + x_163, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_163 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_18 = torch.nn.functional.relu(x_164, inplace=False) + x_164 = None + pow_19 = relu_18**2 + relu_18 = None + mul_18 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ + * pow_19 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_19 + ) = None + x_165 = ( + mul_18 + + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_18 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_166 = torch.conv2d( + x_165, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_165 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_167 = torch.conv2d( + x_166, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_166 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_168 = x_160 + x_167 + x_160 = x_167 = None + x_169 = x_168.permute(0, 2, 3, 1) + x_170 = torch.nn.functional.layer_norm( + x_169, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_169 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) + x_171 = x_170.permute(0, 3, 1, 2) + x_170 = None + x_172 = torch.conv2d( + x_171, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_171 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_19 = torch.nn.functional.relu(x_172, inplace=False) + x_172 = None + pow_20 = relu_19**2 + relu_19 = None + mul_19 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + * pow_20 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( + pow_20 + ) = None + x_173 = ( + mul_19 + + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + ) + mul_19 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) + x_174 = torch.nn.functional.dropout(x_173, 0.0, False, False) + x_173 = None + x_175 = torch.conv2d( + x_174, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_174 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = x_168 + x_176 + x_168 = x_176 = None + x_178 = x_177.permute(0, 2, 3, 1) + x_179 = torch.nn.functional.layer_norm( + x_178, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_178 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) + x_180 = x_179.permute(0, 3, 1, 2) + x_179 = None + x_181 = torch.conv2d( + x_180, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_180 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_20 = torch.nn.functional.relu(x_181, inplace=False) + x_181 = None + pow_21 = relu_20**2 + relu_20 = None + mul_20 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ + * pow_21 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_21 + ) = None + x_182 = ( + mul_20 + + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_20 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_183 = torch.conv2d( + x_182, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_182 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_184 = torch.conv2d( + x_183, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_183 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_185 = x_177 + x_184 + x_177 = x_184 = None + x_186 = x_185.permute(0, 2, 3, 1) + x_187 = torch.nn.functional.layer_norm( + x_186, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) + x_188 = x_187.permute(0, 3, 1, 2) + x_187 = None + x_189 = torch.conv2d( + x_188, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_188 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_21 = torch.nn.functional.relu(x_189, inplace=False) + x_189 = None + pow_22 = relu_21**2 + relu_21 = None + mul_21 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + * pow_22 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( + pow_22 + ) = None + x_190 = ( + mul_21 + + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + ) + mul_21 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) + x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) + x_190 = None + x_192 = torch.conv2d( + x_191, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_191 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_185 + x_193 + x_185 = x_193 = None + x_195 = x_194.permute(0, 2, 3, 1) + x_196 = torch.nn.functional.layer_norm( + x_195, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_195 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) + x_197 = x_196.permute(0, 3, 1, 2) + x_196 = None + x_198 = torch.conv2d( + x_197, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_197 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_22 = torch.nn.functional.relu(x_198, inplace=False) + x_198 = None + pow_23 = relu_22**2 + relu_22 = None + mul_22 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ + * pow_23 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_23 + ) = None + x_199 = ( + mul_22 + + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_22 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_200 = torch.conv2d( + x_199, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_199 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_201 = torch.conv2d( + x_200, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_200 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_202 = x_194 + x_201 + x_194 = x_201 = None + x_203 = x_202.permute(0, 2, 3, 1) + x_204 = torch.nn.functional.layer_norm( + x_203, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_203 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) + x_205 = x_204.permute(0, 3, 1, 2) + x_204 = None + x_206 = torch.conv2d( + x_205, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_205 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_23 = torch.nn.functional.relu(x_206, inplace=False) + x_206 = None + pow_24 = relu_23**2 + relu_23 = None + mul_23 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + * pow_24 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( + pow_24 + ) = None + x_207 = ( + mul_23 + + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + ) + mul_23 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) + x_208 = torch.nn.functional.dropout(x_207, 0.0, False, False) + x_207 = None + x_209 = torch.conv2d( + x_208, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_208 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) + x_209 = None + x_211 = x_202 + x_210 + x_202 = x_210 = None + x_212 = x_211.permute(0, 2, 3, 1) + x_213 = torch.nn.functional.layer_norm( + x_212, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_212 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) + x_214 = x_213.permute(0, 3, 1, 2) + x_213 = None + x_215 = torch.conv2d( + x_214, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_214 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_24 = torch.nn.functional.relu(x_215, inplace=False) + x_215 = None + pow_25 = relu_24**2 + relu_24 = None + mul_24 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ + * pow_25 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_25 + ) = None + x_216 = ( + mul_24 + + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_24 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_217 = torch.conv2d( + x_216, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_216 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_217 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_219 = x_211 + x_218 + x_211 = x_218 = None + x_220 = x_219.permute(0, 2, 3, 1) + x_221 = torch.nn.functional.layer_norm( + x_220, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_220 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) + x_222 = x_221.permute(0, 3, 1, 2) + x_221 = None + x_223 = torch.conv2d( + x_222, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_222 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_25 = torch.nn.functional.relu(x_223, inplace=False) + x_223 = None + pow_26 = relu_25**2 + relu_25 = None + mul_25 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + * pow_26 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( + pow_26 + ) = None + x_224 = ( + mul_25 + + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + ) + mul_25 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = torch.conv2d( + x_225, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_225 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) + x_226 = None + x_228 = x_219 + x_227 + x_219 = x_227 = None + x_229 = x_228.permute(0, 2, 3, 1) + x_230 = torch.nn.functional.layer_norm( + x_229, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_229 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) + x_231 = x_230.permute(0, 3, 1, 2) + x_230 = None + x_232 = torch.conv2d( + x_231, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_231 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_26 = torch.nn.functional.relu(x_232, inplace=False) + x_232 = None + pow_27 = relu_26**2 + relu_26 = None + mul_26 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ + * pow_27 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_27 + ) = None + x_233 = ( + mul_26 + + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_26 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_234 = torch.conv2d( + x_233, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_233 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_235 = torch.conv2d( + x_234, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_234 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_236 = x_228 + x_235 + x_228 = x_235 = None + x_237 = x_236.permute(0, 2, 3, 1) + x_238 = torch.nn.functional.layer_norm( + x_237, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_237 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) + x_239 = x_238.permute(0, 3, 1, 2) + x_238 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_239 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_27 = torch.nn.functional.relu(x_240, inplace=False) + x_240 = None + pow_28 = relu_27**2 + relu_27 = None + mul_27 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + * pow_28 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( + pow_28 + ) = None + x_241 = ( + mul_27 + + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + ) + mul_27 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) + x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) + x_241 = None + x_243 = torch.conv2d( + x_242, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_242 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_244 = torch.nn.functional.dropout(x_243, 0.0, False, False) + x_243 = None + x_245 = x_236 + x_244 + x_236 = x_244 = None + x_246 = x_245.permute(0, 2, 3, 1) + x_247 = torch.nn.functional.layer_norm( + x_246, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_246 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) + x_248 = x_247.permute(0, 3, 1, 2) + x_247 = None + x_249 = torch.conv2d( + x_248, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_248 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_28 = torch.nn.functional.relu(x_249, inplace=False) + x_249 = None + pow_29 = relu_28**2 + relu_28 = None + mul_28 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ + * pow_29 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_29 + ) = None + x_250 = ( + mul_28 + + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_28 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_250 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_252 = torch.conv2d( + x_251, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_251 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_253 = x_245 + x_252 + x_245 = x_252 = None + x_254 = x_253.permute(0, 2, 3, 1) + x_255 = torch.nn.functional.layer_norm( + x_254, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_254 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) + x_256 = x_255.permute(0, 3, 1, 2) + x_255 = None + x_257 = torch.conv2d( + x_256, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_256 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_29 = torch.nn.functional.relu(x_257, inplace=False) + x_257 = None + pow_30 = relu_29**2 + relu_29 = None + mul_29 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + * pow_30 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( + pow_30 + ) = None + x_258 = ( + mul_29 + + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + ) + mul_29 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = torch.conv2d( + x_259, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_259 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_261 = torch.nn.functional.dropout(x_260, 0.0, False, False) + x_260 = None + x_262 = x_253 + x_261 + x_253 = x_261 = None + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (192,), + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) + x_265 = x_264.permute(0, 3, 1, 2) + x_264 = None + x_266 = torch.conv2d( + x_265, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_265 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_2 = sym_sum // 16 + sym_sum_3 = torch.sym_sum([1, floordiv_2]) + floordiv_2 = None + reshape = x_266.reshape(1, 384, -1) + x_266 = None + x_267 = reshape.transpose(1, 2) + reshape = None + view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_30 = x_267 * view + view = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_267, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_52 = sym_sum_3**2 + linear = torch._C._nn.linear( + layer_norm_33, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_33 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_1 = linear.reshape(1, pow_52, 3, 12, 32) + linear = None + qkv = reshape_1.permute(2, 0, 3, 1, 4) + reshape_1 = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_268 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) + q = k = v = None + transpose_1 = x_268.transpose(1, 2) + x_268 = None + x_269 = transpose_1.reshape(1, pow_52, 384) + transpose_1 = None + x_270 = torch._C._nn.linear( + x_269, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_269 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_271 = torch.nn.functional.dropout(x_270, 0.0, False, False) + x_270 = None + x_272 = mul_30 + x_271 + mul_30 = x_271 = None + view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_31 = x_272 * view_1 + view_1 = None + layer_norm_34 = torch.nn.functional.layer_norm( + x_272, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_272 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_273 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_34 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_30 = torch.nn.functional.relu(x_273, inplace=False) + x_273 = None + pow_31 = relu_30**2 + relu_30 = None + mul_32 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_31 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_31 + ) = None + x_274 = ( + mul_32 + + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_275 = torch.nn.functional.dropout(x_274, 0.0, False, False) + x_274 = None + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_277 = torch.nn.functional.dropout(x_276, 0.0, False, False) + x_276 = None + x_278 = mul_31 + x_277 + mul_31 = x_277 = None + view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_33 = x_278 * view_2 + view_2 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_278, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_3 = linear_4.reshape(1, pow_52, 3, 12, 32) + linear_4 = None + qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) + reshape_3 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_279 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) + q_1 = k_1 = v_1 = None + transpose_2 = x_279.transpose(1, 2) + x_279 = None + x_280 = transpose_2.reshape(1, pow_52, 384) + transpose_2 = None + x_281 = torch._C._nn.linear( + x_280, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_280 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) + x_281 = None + x_283 = mul_33 + x_282 + mul_33 = x_282 = None + view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_34 = x_283 * view_3 + view_3 = None + layer_norm_36 = torch.nn.functional.layer_norm( + x_283, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_283 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_284 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_36 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_31 = torch.nn.functional.relu(x_284, inplace=False) + x_284 = None + pow_32 = relu_31**2 + relu_31 = None + mul_35 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_32 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_32 + ) = None + x_285 = ( + mul_35 + + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_286 = torch.nn.functional.dropout(x_285, 0.0, False, False) + x_285 = None + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) + x_287 = None + x_289 = mul_34 + x_288 + mul_34 = x_288 = None + view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_36 = x_289 * view_4 + view_4 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_289, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_37 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_5 = linear_8.reshape(1, pow_52, 3, 12, 32) + linear_8 = None + qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) + reshape_5 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_290 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) + q_2 = k_2 = v_2 = None + transpose_3 = x_290.transpose(1, 2) + x_290 = None + x_291 = transpose_3.reshape(1, pow_52, 384) + transpose_3 = None + x_292 = torch._C._nn.linear( + x_291, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_291 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_293 = torch.nn.functional.dropout(x_292, 0.0, False, False) + x_292 = None + x_294 = mul_36 + x_293 + mul_36 = x_293 = None + view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_37 = x_294 * view_5 + view_5 = None + layer_norm_38 = torch.nn.functional.layer_norm( + x_294, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_294 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_295 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_32 = torch.nn.functional.relu(x_295, inplace=False) + x_295 = None + pow_33 = relu_32**2 + relu_32 = None + mul_38 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_33 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_33 + ) = None + x_296 = ( + mul_38 + + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_297 = torch.nn.functional.dropout(x_296, 0.0, False, False) + x_296 = None + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_299 = torch.nn.functional.dropout(x_298, 0.0, False, False) + x_298 = None + x_300 = mul_37 + x_299 + mul_37 = x_299 = None + view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_39 = x_300 * view_6 + view_6 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_300, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_39 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_7 = linear_12.reshape(1, pow_52, 3, 12, 32) + linear_12 = None + qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) + reshape_7 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_301 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) + q_3 = k_3 = v_3 = None + transpose_4 = x_301.transpose(1, 2) + x_301 = None + x_302 = transpose_4.reshape(1, pow_52, 384) + transpose_4 = None + x_303 = torch._C._nn.linear( + x_302, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_302 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_304 = torch.nn.functional.dropout(x_303, 0.0, False, False) + x_303 = None + x_305 = mul_39 + x_304 + mul_39 = x_304 = None + view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_40 = x_305 * view_7 + view_7 = None + layer_norm_40 = torch.nn.functional.layer_norm( + x_305, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_305 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) + x_306 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_40 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_33 = torch.nn.functional.relu(x_306, inplace=False) + x_306 = None + pow_34 = relu_33**2 + relu_33 = None + mul_41 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + * pow_34 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( + pow_34 + ) = None + x_307 = ( + mul_41 + + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + ) + mul_41 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) + x_308 = torch.nn.functional.dropout(x_307, 0.0, False, False) + x_307 = None + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) + x_309 = None + x_311 = mul_40 + x_310 + mul_40 = x_310 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_42 = x_311 * view_8 + view_8 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_311, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_41 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_9 = linear_16.reshape(1, pow_52, 3, 12, 32) + linear_16 = None + qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) + reshape_9 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_312 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) + q_4 = k_4 = v_4 = None + transpose_5 = x_312.transpose(1, 2) + x_312 = None + x_313 = transpose_5.reshape(1, pow_52, 384) + transpose_5 = None + x_314 = torch._C._nn.linear( + x_313, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_313 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) + x_314 = None + x_316 = mul_42 + x_315 + mul_42 = x_315 = None + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_43 = x_316 * view_9 + view_9 = None + layer_norm_42 = torch.nn.functional.layer_norm( + x_316, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_316 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) + x_317 = torch._C._nn.linear( + layer_norm_42, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_42 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_34 = torch.nn.functional.relu(x_317, inplace=False) + x_317 = None + pow_35 = relu_34**2 + relu_34 = None + mul_44 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + * pow_35 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( + pow_35 + ) = None + x_318 = ( + mul_44 + + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + ) + mul_44 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) + x_319 = torch.nn.functional.dropout(x_318, 0.0, False, False) + x_318 = None + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_321 = torch.nn.functional.dropout(x_320, 0.0, False, False) + x_320 = None + x_322 = mul_43 + x_321 + mul_43 = x_321 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_45 = x_322 * view_10 + view_10 = None + layer_norm_43 = torch.nn.functional.layer_norm( + x_322, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) + linear_20 = torch._C._nn.linear( + layer_norm_43, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_43 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_11 = linear_20.reshape(1, pow_52, 3, 12, 32) + linear_20 = None + qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) + reshape_11 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_323 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) + q_5 = k_5 = v_5 = None + transpose_6 = x_323.transpose(1, 2) + x_323 = None + x_324 = transpose_6.reshape(1, pow_52, 384) + transpose_6 = None + x_325 = torch._C._nn.linear( + x_324, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_324 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_326 = torch.nn.functional.dropout(x_325, 0.0, False, False) + x_325 = None + x_327 = mul_45 + x_326 + mul_45 = x_326 = None + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_46 = x_327 * view_11 + view_11 = None + layer_norm_44 = torch.nn.functional.layer_norm( + x_327, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_327 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) + x_328 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_44 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_35 = torch.nn.functional.relu(x_328, inplace=False) + x_328 = None + pow_36 = relu_35**2 + relu_35 = None + mul_47 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + * pow_36 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( + pow_36 + ) = None + x_329 = ( + mul_47 + + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + ) + mul_47 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) + x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) + x_329 = None + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_332 = torch.nn.functional.dropout(x_331, 0.0, False, False) + x_331 = None + x_333 = mul_46 + x_332 + mul_46 = x_332 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_48 = x_333 * view_12 + view_12 = None + layer_norm_45 = torch.nn.functional.layer_norm( + x_333, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) + linear_24 = torch._C._nn.linear( + layer_norm_45, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_45 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_13 = linear_24.reshape(1, pow_52, 3, 12, 32) + linear_24 = None + qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) + reshape_13 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_334 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) + q_6 = k_6 = v_6 = None + transpose_7 = x_334.transpose(1, 2) + x_334 = None + x_335 = transpose_7.reshape(1, pow_52, 384) + transpose_7 = None + x_336 = torch._C._nn.linear( + x_335, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_335 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_337 = torch.nn.functional.dropout(x_336, 0.0, False, False) + x_336 = None + x_338 = mul_48 + x_337 + mul_48 = x_337 = None + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_49 = x_338 * view_13 + view_13 = None + layer_norm_46 = torch.nn.functional.layer_norm( + x_338, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_338 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) + x_339 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_46 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_36 = torch.nn.functional.relu(x_339, inplace=False) + x_339 = None + pow_37 = relu_36**2 + relu_36 = None + mul_50 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + * pow_37 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( + pow_37 + ) = None + x_340 = ( + mul_50 + + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + ) + mul_50 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) + x_341 = torch.nn.functional.dropout(x_340, 0.0, False, False) + x_340 = None + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_343 = torch.nn.functional.dropout(x_342, 0.0, False, False) + x_342 = None + x_344 = mul_49 + x_343 + mul_49 = x_343 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_51 = x_344 * view_14 + view_14 = None + layer_norm_47 = torch.nn.functional.layer_norm( + x_344, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) + linear_28 = torch._C._nn.linear( + layer_norm_47, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_47 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_15 = linear_28.reshape(1, pow_52, 3, 12, 32) + linear_28 = None + qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) + reshape_15 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_345 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) + q_7 = k_7 = v_7 = None + transpose_8 = x_345.transpose(1, 2) + x_345 = None + x_346 = transpose_8.reshape(1, pow_52, 384) + transpose_8 = None + x_347 = torch._C._nn.linear( + x_346, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_346 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) + x_347 = None + x_349 = mul_51 + x_348 + mul_51 = x_348 = None + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_52 = x_349 * view_15 + view_15 = None + layer_norm_48 = torch.nn.functional.layer_norm( + x_349, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_349 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) + x_350 = torch._C._nn.linear( + layer_norm_48, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_48 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_37 = torch.nn.functional.relu(x_350, inplace=False) + x_350 = None + pow_38 = relu_37**2 + relu_37 = None + mul_53 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + * pow_38 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( + pow_38 + ) = None + x_351 = ( + mul_53 + + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + ) + mul_53 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) + x_352 = torch.nn.functional.dropout(x_351, 0.0, False, False) + x_351 = None + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_354 = torch.nn.functional.dropout(x_353, 0.0, False, False) + x_353 = None + x_355 = mul_52 + x_354 + mul_52 = x_354 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_54 = x_355 * view_16 + view_16 = None + layer_norm_49 = torch.nn.functional.layer_norm( + x_355, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) + linear_32 = torch._C._nn.linear( + layer_norm_49, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_49 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_17 = linear_32.reshape(1, pow_52, 3, 12, 32) + linear_32 = None + qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) + reshape_17 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_356 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) + q_8 = k_8 = v_8 = None + transpose_9 = x_356.transpose(1, 2) + x_356 = None + x_357 = transpose_9.reshape(1, pow_52, 384) + transpose_9 = None + x_358 = torch._C._nn.linear( + x_357, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_357 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_359 = torch.nn.functional.dropout(x_358, 0.0, False, False) + x_358 = None + x_360 = mul_54 + x_359 + mul_54 = x_359 = None + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_55 = x_360 * view_17 + view_17 = None + layer_norm_50 = torch.nn.functional.layer_norm( + x_360, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_360 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) + x_361 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_50 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_38 = torch.nn.functional.relu(x_361, inplace=False) + x_361 = None + pow_39 = relu_38**2 + relu_38 = None + mul_56 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + * pow_39 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( + pow_39 + ) = None + x_362 = ( + mul_56 + + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + ) + mul_56 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) + x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) + x_362 = None + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_365 = torch.nn.functional.dropout(x_364, 0.0, False, False) + x_364 = None + x_366 = mul_55 + x_365 + mul_55 = x_365 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_57 = x_366 * view_18 + view_18 = None + layer_norm_51 = torch.nn.functional.layer_norm( + x_366, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_51, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_51 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_19 = linear_36.reshape(1, pow_52, 3, 12, 32) + linear_36 = None + qkv_9 = reshape_19.permute(2, 0, 3, 1, 4) + reshape_19 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_367 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) + q_9 = k_9 = v_9 = None + transpose_10 = x_367.transpose(1, 2) + x_367 = None + x_368 = transpose_10.reshape(1, pow_52, 384) + transpose_10 = None + x_369 = torch._C._nn.linear( + x_368, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_368 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_370 = torch.nn.functional.dropout(x_369, 0.0, False, False) + x_369 = None + x_371 = mul_57 + x_370 + mul_57 = x_370 = None + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_58 = x_371 * view_19 + view_19 = None + layer_norm_52 = torch.nn.functional.layer_norm( + x_371, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_371 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) + x_372 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_52 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_39 = torch.nn.functional.relu(x_372, inplace=False) + x_372 = None + pow_40 = relu_39**2 + relu_39 = None + mul_59 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + * pow_40 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( + pow_40 + ) = None + x_373 = ( + mul_59 + + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + ) + mul_59 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) + x_374 = torch.nn.functional.dropout(x_373, 0.0, False, False) + x_373 = None + x_375 = torch._C._nn.linear( + x_374, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_374 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_376 = torch.nn.functional.dropout(x_375, 0.0, False, False) + x_375 = None + x_377 = mul_58 + x_376 + mul_58 = x_376 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_60 = x_377 * view_20 + view_20 = None + layer_norm_53 = torch.nn.functional.layer_norm( + x_377, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_377 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_53, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_53 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_21 = linear_40.reshape(1, pow_52, 3, 12, 32) + linear_40 = None + qkv_10 = reshape_21.permute(2, 0, 3, 1, 4) + reshape_21 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_378 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_378.transpose(1, 2) + x_378 = None + x_379 = transpose_11.reshape(1, pow_52, 384) + transpose_11 = None + x_380 = torch._C._nn.linear( + x_379, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_379 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_381 = torch.nn.functional.dropout(x_380, 0.0, False, False) + x_380 = None + x_382 = mul_60 + x_381 + mul_60 = x_381 = None + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_61 = x_382 * view_21 + view_21 = None + layer_norm_54 = torch.nn.functional.layer_norm( + x_382, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_382 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) + x_383 = torch._C._nn.linear( + layer_norm_54, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_54 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_40 = torch.nn.functional.relu(x_383, inplace=False) + x_383 = None + pow_41 = relu_40**2 + relu_40 = None + mul_62 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + * pow_41 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( + pow_41 + ) = None + x_384 = ( + mul_62 + + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + ) + mul_62 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) + x_385 = torch.nn.functional.dropout(x_384, 0.0, False, False) + x_384 = None + x_386 = torch._C._nn.linear( + x_385, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_385 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_387 = torch.nn.functional.dropout(x_386, 0.0, False, False) + x_386 = None + x_388 = mul_61 + x_387 + mul_61 = x_387 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_63 = x_388 * view_22 + view_22 = None + layer_norm_55 = torch.nn.functional.layer_norm( + x_388, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_388 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_55, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_55 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_23 = linear_44.reshape(1, pow_52, 3, 12, 32) + linear_44 = None + qkv_11 = reshape_23.permute(2, 0, 3, 1, 4) + reshape_23 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_389 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_389.transpose(1, 2) + x_389 = None + x_390 = transpose_12.reshape(1, pow_52, 384) + transpose_12 = None + x_391 = torch._C._nn.linear( + x_390, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_390 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_392 = torch.nn.functional.dropout(x_391, 0.0, False, False) + x_391 = None + x_393 = mul_63 + x_392 + mul_63 = x_392 = None + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_64 = x_393 * view_23 + view_23 = None + layer_norm_56 = torch.nn.functional.layer_norm( + x_393, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_393 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) + x_394 = torch._C._nn.linear( + layer_norm_56, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_56 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_41 = torch.nn.functional.relu(x_394, inplace=False) + x_394 = None + pow_42 = relu_41**2 + relu_41 = None + mul_65 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + * pow_42 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( + pow_42 + ) = None + x_395 = ( + mul_65 + + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + ) + mul_65 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) + x_396 = torch.nn.functional.dropout(x_395, 0.0, False, False) + x_395 = None + x_397 = torch._C._nn.linear( + x_396, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_396 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_398 = torch.nn.functional.dropout(x_397, 0.0, False, False) + x_397 = None + x_399 = mul_64 + x_398 + mul_64 = x_398 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_66 = x_399 * view_24 + view_24 = None + layer_norm_57 = torch.nn.functional.layer_norm( + x_399, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_399 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_57, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_57 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_25 = linear_48.reshape(1, pow_52, 3, 12, 32) + linear_48 = None + qkv_12 = reshape_25.permute(2, 0, 3, 1, 4) + reshape_25 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_12 = unbind_12[0] + k_12 = unbind_12[1] + v_12 = unbind_12[2] + unbind_12 = None + x_400 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_13 = x_400.transpose(1, 2) + x_400 = None + x_401 = transpose_13.reshape(1, pow_52, 384) + transpose_13 = None + x_402 = torch._C._nn.linear( + x_401, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_401 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_403 = torch.nn.functional.dropout(x_402, 0.0, False, False) + x_402 = None + x_404 = mul_66 + x_403 + mul_66 = x_403 = None + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_67 = x_404 * view_25 + view_25 = None + layer_norm_58 = torch.nn.functional.layer_norm( + x_404, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_404 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = (None) + x_405 = torch._C._nn.linear( + layer_norm_58, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_58 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_42 = torch.nn.functional.relu(x_405, inplace=False) + x_405 = None + pow_43 = relu_42**2 + relu_42 = None + mul_68 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ + * pow_43 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = ( + pow_43 + ) = None + x_406 = ( + mul_68 + + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ + ) + mul_68 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = (None) + x_407 = torch.nn.functional.dropout(x_406, 0.0, False, False) + x_406 = None + x_408 = torch._C._nn.linear( + x_407, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_407 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_409 = torch.nn.functional.dropout(x_408, 0.0, False, False) + x_408 = None + x_410 = mul_67 + x_409 + mul_67 = x_409 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_69 = x_410 * view_26 + view_26 = None + layer_norm_59 = torch.nn.functional.layer_norm( + x_410, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_410 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = (None) + linear_52 = torch._C._nn.linear( + layer_norm_59, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_59 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_27 = linear_52.reshape(1, pow_52, 3, 12, 32) + linear_52 = None + qkv_13 = reshape_27.permute(2, 0, 3, 1, 4) + reshape_27 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_13 = unbind_13[0] + k_13 = unbind_13[1] + v_13 = unbind_13[2] + unbind_13 = None + x_411 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_14 = x_411.transpose(1, 2) + x_411 = None + x_412 = transpose_14.reshape(1, pow_52, 384) + transpose_14 = None + x_413 = torch._C._nn.linear( + x_412, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_412 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_414 = torch.nn.functional.dropout(x_413, 0.0, False, False) + x_413 = None + x_415 = mul_69 + x_414 + mul_69 = x_414 = None + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_70 = x_415 * view_27 + view_27 = None + layer_norm_60 = torch.nn.functional.layer_norm( + x_415, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_415 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = (None) + x_416 = torch._C._nn.linear( + layer_norm_60, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_60 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_43 = torch.nn.functional.relu(x_416, inplace=False) + x_416 = None + pow_44 = relu_43**2 + relu_43 = None + mul_71 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ + * pow_44 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = ( + pow_44 + ) = None + x_417 = ( + mul_71 + + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ + ) + mul_71 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = (None) + x_418 = torch.nn.functional.dropout(x_417, 0.0, False, False) + x_417 = None + x_419 = torch._C._nn.linear( + x_418, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_418 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_420 = torch.nn.functional.dropout(x_419, 0.0, False, False) + x_419 = None + x_421 = mul_70 + x_420 + mul_70 = x_420 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_72 = x_421 * view_28 + view_28 = None + layer_norm_61 = torch.nn.functional.layer_norm( + x_421, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_421 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_61, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_61 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_29 = linear_56.reshape(1, pow_52, 3, 12, 32) + linear_56 = None + qkv_14 = reshape_29.permute(2, 0, 3, 1, 4) + reshape_29 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_14 = unbind_14[0] + k_14 = unbind_14[1] + v_14 = unbind_14[2] + unbind_14 = None + x_422 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_15 = x_422.transpose(1, 2) + x_422 = None + x_423 = transpose_15.reshape(1, pow_52, 384) + transpose_15 = None + x_424 = torch._C._nn.linear( + x_423, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_423 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_425 = torch.nn.functional.dropout(x_424, 0.0, False, False) + x_424 = None + x_426 = mul_72 + x_425 + mul_72 = x_425 = None + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_73 = x_426 * view_29 + view_29 = None + layer_norm_62 = torch.nn.functional.layer_norm( + x_426, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_426 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = (None) + x_427 = torch._C._nn.linear( + layer_norm_62, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_62 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_44 = torch.nn.functional.relu(x_427, inplace=False) + x_427 = None + pow_45 = relu_44**2 + relu_44 = None + mul_74 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ + * pow_45 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = ( + pow_45 + ) = None + x_428 = ( + mul_74 + + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ + ) + mul_74 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = (None) + x_429 = torch.nn.functional.dropout(x_428, 0.0, False, False) + x_428 = None + x_430 = torch._C._nn.linear( + x_429, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_429 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_431 = torch.nn.functional.dropout(x_430, 0.0, False, False) + x_430 = None + x_432 = mul_73 + x_431 + mul_73 = x_431 = None + view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_75 = x_432 * view_30 + view_30 = None + layer_norm_63 = torch.nn.functional.layer_norm( + x_432, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_432 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = (None) + linear_60 = torch._C._nn.linear( + layer_norm_63, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_63 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_31 = linear_60.reshape(1, pow_52, 3, 12, 32) + linear_60 = None + qkv_15 = reshape_31.permute(2, 0, 3, 1, 4) + reshape_31 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_15 = unbind_15[0] + k_15 = unbind_15[1] + v_15 = unbind_15[2] + unbind_15 = None + x_433 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_16 = x_433.transpose(1, 2) + x_433 = None + x_434 = transpose_16.reshape(1, pow_52, 384) + transpose_16 = None + x_435 = torch._C._nn.linear( + x_434, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_434 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_436 = torch.nn.functional.dropout(x_435, 0.0, False, False) + x_435 = None + x_437 = mul_75 + x_436 + mul_75 = x_436 = None + view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_76 = x_437 * view_31 + view_31 = None + layer_norm_64 = torch.nn.functional.layer_norm( + x_437, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_437 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = (None) + x_438 = torch._C._nn.linear( + layer_norm_64, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_64 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_45 = torch.nn.functional.relu(x_438, inplace=False) + x_438 = None + pow_46 = relu_45**2 + relu_45 = None + mul_77 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ + * pow_46 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = ( + pow_46 + ) = None + x_439 = ( + mul_77 + + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ + ) + mul_77 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = (None) + x_440 = torch.nn.functional.dropout(x_439, 0.0, False, False) + x_439 = None + x_441 = torch._C._nn.linear( + x_440, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_440 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_442 = torch.nn.functional.dropout(x_441, 0.0, False, False) + x_441 = None + x_443 = mul_76 + x_442 + mul_76 = x_442 = None + view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_78 = x_443 * view_32 + view_32 = None + layer_norm_65 = torch.nn.functional.layer_norm( + x_443, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_443 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_65, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_65 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_33 = linear_64.reshape(1, pow_52, 3, 12, 32) + linear_64 = None + qkv_16 = reshape_33.permute(2, 0, 3, 1, 4) + reshape_33 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_16 = unbind_16[0] + k_16 = unbind_16[1] + v_16 = unbind_16[2] + unbind_16 = None + x_444 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_17 = x_444.transpose(1, 2) + x_444 = None + x_445 = transpose_17.reshape(1, pow_52, 384) + transpose_17 = None + x_446 = torch._C._nn.linear( + x_445, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_445 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_447 = torch.nn.functional.dropout(x_446, 0.0, False, False) + x_446 = None + x_448 = mul_78 + x_447 + mul_78 = x_447 = None + view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_79 = x_448 * view_33 + view_33 = None + layer_norm_66 = torch.nn.functional.layer_norm( + x_448, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_448 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = (None) + x_449 = torch._C._nn.linear( + layer_norm_66, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_66 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_46 = torch.nn.functional.relu(x_449, inplace=False) + x_449 = None + pow_47 = relu_46**2 + relu_46 = None + mul_80 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ + * pow_47 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = ( + pow_47 + ) = None + x_450 = ( + mul_80 + + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ + ) + mul_80 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = (None) + x_451 = torch.nn.functional.dropout(x_450, 0.0, False, False) + x_450 = None + x_452 = torch._C._nn.linear( + x_451, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_451 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_453 = torch.nn.functional.dropout(x_452, 0.0, False, False) + x_452 = None + x_454 = mul_79 + x_453 + mul_79 = x_453 = None + view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_81 = x_454 * view_34 + view_34 = None + layer_norm_67 = torch.nn.functional.layer_norm( + x_454, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_454 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = (None) + linear_68 = torch._C._nn.linear( + layer_norm_67, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_67 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_35 = linear_68.reshape(1, pow_52, 3, 12, 32) + linear_68 = None + qkv_17 = reshape_35.permute(2, 0, 3, 1, 4) + reshape_35 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_17 = unbind_17[0] + k_17 = unbind_17[1] + v_17 = unbind_17[2] + unbind_17 = None + x_455 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_18 = x_455.transpose(1, 2) + x_455 = None + x_456 = transpose_18.reshape(1, pow_52, 384) + transpose_18 = pow_52 = None + x_457 = torch._C._nn.linear( + x_456, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_456 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_458 = torch.nn.functional.dropout(x_457, 0.0, False, False) + x_457 = None + x_459 = mul_81 + x_458 + mul_81 = x_458 = None + view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_82 = x_459 * view_35 + view_35 = None + layer_norm_68 = torch.nn.functional.layer_norm( + x_459, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_459 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = (None) + x_460 = torch._C._nn.linear( + layer_norm_68, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_68 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_47 = torch.nn.functional.relu(x_460, inplace=False) + x_460 = None + pow_48 = relu_47**2 + relu_47 = None + mul_83 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ + * pow_48 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = ( + pow_48 + ) = None + x_461 = ( + mul_83 + + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ + ) + mul_83 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = (None) + x_462 = torch.nn.functional.dropout(x_461, 0.0, False, False) + x_461 = None + x_463 = torch._C._nn.linear( + x_462, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_462 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_464 = torch.nn.functional.dropout(x_463, 0.0, False, False) + x_463 = None + x_465 = mul_82 + x_464 + mul_82 = x_464 = None + transpose_19 = x_465.transpose(1, 2) + x_465 = None + x_466 = transpose_19.reshape(1, 384, sym_sum_3, sym_sum_3) + transpose_19 = sym_sum_3 = None + x_467 = x_466.permute(0, 2, 3, 1) + x_466 = None + x_468 = torch.nn.functional.layer_norm( + x_467, + (384,), + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_467 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) + x_469 = x_468.permute(0, 3, 1, 2) + x_468 = None + x_470 = torch.conv2d( + x_469, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_469 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_3 = sym_sum // 32 + sym_sum = None + sym_sum_4 = torch.sym_sum([1, floordiv_3]) + floordiv_3 = None + reshape_38 = x_470.reshape(1, 576, -1) + x_470 = None + x_471 = reshape_38.transpose(1, 2) + reshape_38 = None + view_36 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_84 = x_471 * view_36 + view_36 = None + layer_norm_70 = torch.nn.functional.layer_norm( + x_471, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_471 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_53 = sym_sum_4**2 + linear_72 = torch._C._nn.linear( + layer_norm_70, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_70 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_39 = linear_72.reshape(1, pow_53, 3, 18, 32) + linear_72 = None + qkv_18 = reshape_39.permute(2, 0, 3, 1, 4) + reshape_39 = None + unbind_18 = qkv_18.unbind(0) + qkv_18 = None + q_18 = unbind_18[0] + k_18 = unbind_18[1] + v_18 = unbind_18[2] + unbind_18 = None + x_472 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_21 = x_472.transpose(1, 2) + x_472 = None + x_473 = transpose_21.reshape(1, pow_53, 576) + transpose_21 = None + x_474 = torch._C._nn.linear( + x_473, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_473 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_475 = torch.nn.functional.dropout(x_474, 0.0, False, False) + x_474 = None + x_476 = mul_84 + x_475 + mul_84 = x_475 = None + view_37 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_85 = x_476 * view_37 + view_37 = None + layer_norm_71 = torch.nn.functional.layer_norm( + x_476, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_476 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_477 = torch._C._nn.linear( + layer_norm_71, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_71 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_48 = torch.nn.functional.relu(x_477, inplace=False) + x_477 = None + pow_49 = relu_48**2 + relu_48 = None + mul_86 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_49 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_49 + ) = None + x_478 = ( + mul_86 + + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_86 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_479 = torch.nn.functional.dropout(x_478, 0.0, False, False) + x_478 = None + x_480 = torch._C._nn.linear( + x_479, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_479 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_481 = torch.nn.functional.dropout(x_480, 0.0, False, False) + x_480 = None + x_482 = mul_85 + x_481 + mul_85 = x_481 = None + view_38 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_87 = x_482 * view_38 + view_38 = None + layer_norm_72 = torch.nn.functional.layer_norm( + x_482, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_482 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_76 = torch._C._nn.linear( + layer_norm_72, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_72 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_41 = linear_76.reshape(1, pow_53, 3, 18, 32) + linear_76 = None + qkv_19 = reshape_41.permute(2, 0, 3, 1, 4) + reshape_41 = None + unbind_19 = qkv_19.unbind(0) + qkv_19 = None + q_19 = unbind_19[0] + k_19 = unbind_19[1] + v_19 = unbind_19[2] + unbind_19 = None + x_483 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_22 = x_483.transpose(1, 2) + x_483 = None + x_484 = transpose_22.reshape(1, pow_53, 576) + transpose_22 = None + x_485 = torch._C._nn.linear( + x_484, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_484 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_486 = torch.nn.functional.dropout(x_485, 0.0, False, False) + x_485 = None + x_487 = mul_87 + x_486 + mul_87 = x_486 = None + view_39 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_88 = x_487 * view_39 + view_39 = None + layer_norm_73 = torch.nn.functional.layer_norm( + x_487, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_487 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_488 = torch._C._nn.linear( + layer_norm_73, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_73 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_49 = torch.nn.functional.relu(x_488, inplace=False) + x_488 = None + pow_50 = relu_49**2 + relu_49 = None + mul_89 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_50 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_50 + ) = None + x_489 = ( + mul_89 + + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_89 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_490 = torch.nn.functional.dropout(x_489, 0.0, False, False) + x_489 = None + x_491 = torch._C._nn.linear( + x_490, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_490 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_492 = torch.nn.functional.dropout(x_491, 0.0, False, False) + x_491 = None + x_493 = mul_88 + x_492 + mul_88 = x_492 = None + view_40 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_90 = x_493 * view_40 + view_40 = None + layer_norm_74 = torch.nn.functional.layer_norm( + x_493, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_493 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_74, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_74 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_43 = linear_80.reshape(1, pow_53, 3, 18, 32) + linear_80 = None + qkv_20 = reshape_43.permute(2, 0, 3, 1, 4) + reshape_43 = None + unbind_20 = qkv_20.unbind(0) + qkv_20 = None + q_20 = unbind_20[0] + k_20 = unbind_20[1] + v_20 = unbind_20[2] + unbind_20 = None + x_494 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_23 = x_494.transpose(1, 2) + x_494 = None + x_495 = transpose_23.reshape(1, pow_53, 576) + transpose_23 = pow_53 = None + x_496 = torch._C._nn.linear( + x_495, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_495 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_497 = torch.nn.functional.dropout(x_496, 0.0, False, False) + x_496 = None + x_498 = mul_90 + x_497 + mul_90 = x_497 = None + view_41 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_91 = x_498 * view_41 + view_41 = None + layer_norm_75 = torch.nn.functional.layer_norm( + x_498, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_498 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_499 = torch._C._nn.linear( + layer_norm_75, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_75 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_50 = torch.nn.functional.relu(x_499, inplace=False) + x_499 = None + pow_51 = relu_50**2 + relu_50 = None + mul_92 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_51 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_51 + ) = None + x_500 = ( + mul_92 + + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_92 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_501 = torch.nn.functional.dropout(x_500, 0.0, False, False) + x_500 = None + x_502 = torch._C._nn.linear( + x_501, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_501 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_503 = torch.nn.functional.dropout(x_502, 0.0, False, False) + x_502 = None + x_504 = mul_91 + x_503 + mul_91 = x_503 = None + transpose_24 = x_504.transpose(1, 2) + x_504 = None + x_505 = transpose_24.reshape(1, 576, sym_sum_4, sym_sum_4) + transpose_24 = sym_sum_4 = None + x_506 = torch.nn.functional.adaptive_avg_pool2d(x_505, 1) + x_505 = None + x_507 = x_506.permute(0, 2, 3, 1) + x_506 = None + x_508 = torch.nn.functional.layer_norm( + x_507, + (576,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_507 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_509 = x_508.permute(0, 3, 1, 2) + x_508 = None + x_510 = x_509.flatten(1, -1) + x_509 = None + x_511 = torch.nn.functional.dropout(x_510, 0.0, False, False) + x_510 = None + x_512 = torch._C._nn.linear( + x_511, + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, + ) + x_511 = ( + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None + relu_51 = torch.nn.functional.relu(x_512, inplace=False) + x_512 = None + x_513 = torch.square(relu_51) + relu_51 = None + x_514 = torch.nn.functional.layer_norm( + x_513, + (2304,), + l_self_modules_head_modules_fc_modules_norm_parameters_weight_, + l_self_modules_head_modules_fc_modules_norm_parameters_bias_, + 1e-06, + ) + x_513 = ( + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None + x_515 = torch.nn.functional.dropout(x_514, 0.0, False, False) + x_514 = None + x_516 = torch._C._nn.linear( + x_515, + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, + ) + x_515 = ( + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None + return (x_516,) diff --git a/samples/timm/caformer_m36.sail_in22k/weight_meta.py b/samples/timm/caformer_m36.sail_in22k/weight_meta.py new file mode 100644 index 0000000000..7f331146f7 --- /dev/null +++ b/samples/timm/caformer_m36.sail_in22k/weight_meta.py @@ -0,0 +1,3968 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv_parameters_weight_" + shape = [96, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: + name = "L_self_modules_stem_modules_conv_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.227 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: + name = "L_self_modules_stem_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [192, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [96, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [192, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [96, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [192, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [96, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" + shape = [192, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" + shape = [384, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" + shape = [576, 384, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1728, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [576, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2304, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [576, 2304] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1728, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [576, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2304, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [576, 2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1728, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [576, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2304, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [576, 2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" + shape = [2304, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" + shape = [21841, 2304] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" + shape = [21841] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/caformer_m36.sail_in22k_ft_in1k/graph_net.json b/samples/timm/caformer_m36.sail_in22k_ft_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/caformer_m36.sail_in22k_ft_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/caformer_m36.sail_in22k_ft_in1k/input_meta.py b/samples/timm/caformer_m36.sail_in22k_ft_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_m36.sail_in22k_ft_in1k/input_tensor_constraints.py b/samples/timm/caformer_m36.sail_in22k_ft_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_m36.sail_in22k_ft_in1k/model.py b/samples/timm/caformer_m36.sail_in22k_ft_in1k/model.py new file mode 100644 index 0000000000..dcce286c5c --- /dev/null +++ b/samples/timm/caformer_m36.sail_in22k_ft_in1k/model.py @@ -0,0 +1,4754 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv_parameters_bias_ = ( + L_self_modules_stem_modules_conv_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_norm_parameters_weight_ = ( + L_self_modules_stem_modules_norm_parameters_weight_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv_parameters_weight_, + l_self_modules_stem_modules_conv_parameters_bias_, + (4, 4), + (2, 2), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_conv_parameters_weight_ + ) = l_self_modules_stem_modules_conv_parameters_bias_ = None + x_1 = x.permute(0, 2, 3, 1) + x = None + x_2 = torch.nn.functional.layer_norm( + x_1, (96,), l_self_modules_stem_modules_norm_parameters_weight_, None, 1e-06 + ) + x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None + x_3 = x_2.permute(0, 3, 1, 2) + x_2 = None + sym_sum = torch.sym_sum([-3, s1]) + s1 = None + floordiv = sym_sum // 4 + sym_sum_1 = torch.sym_sum([1, floordiv]) + floordiv = sym_sum_1 = None + x_4 = x_3.permute(0, 2, 3, 1) + x_5 = torch.nn.functional.layer_norm( + x_4, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu = torch.nn.functional.relu(x_7, inplace=False) + x_7 = None + pow_1 = relu**2 + relu = None + mul = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_1 + ) = None + x_8 = ( + mul + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_11 = x_3 + x_10 + x_3 = x_10 = None + x_12 = x_11.permute(0, 2, 3, 1) + x_13 = torch.nn.functional.layer_norm( + x_12, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_14 = x_13.permute(0, 3, 1, 2) + x_13 = None + x_15 = torch.conv2d( + x_14, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_1 = torch.nn.functional.relu(x_15, inplace=False) + x_15 = None + pow_2 = relu_1**2 + relu_1 = None + mul_1 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_2 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_2 + ) = None + x_16 = ( + mul_1 + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = x_11 + x_19 + x_11 = x_19 = None + x_21 = x_20.permute(0, 2, 3, 1) + x_22 = torch.nn.functional.layer_norm( + x_21, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = torch.conv2d( + x_23, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_2 = torch.nn.functional.relu(x_24, inplace=False) + x_24 = None + pow_3 = relu_2**2 + relu_2 = None + mul_2 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_3 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_3 + ) = None + x_25 = ( + mul_2 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_27 = torch.conv2d( + x_26, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_28 = x_20 + x_27 + x_20 = x_27 = None + x_29 = x_28.permute(0, 2, 3, 1) + x_30 = torch.nn.functional.layer_norm( + x_29, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_3 = torch.nn.functional.relu(x_32, inplace=False) + x_32 = None + pow_4 = relu_3**2 + relu_3 = None + mul_3 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_4 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_4 + ) = None + x_33 = ( + mul_3 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + x_37 = x_28 + x_36 + x_28 = x_36 = None + x_38 = x_37.permute(0, 2, 3, 1) + x_39 = torch.nn.functional.layer_norm( + x_38, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_40 = x_39.permute(0, 3, 1, 2) + x_39 = None + x_41 = torch.conv2d( + x_40, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_4 = torch.nn.functional.relu(x_41, inplace=False) + x_41 = None + pow_5 = relu_4**2 + relu_4 = None + mul_4 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_5 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_5 + ) = None + x_42 = ( + mul_4 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_44 = torch.conv2d( + x_43, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_45 = x_37 + x_44 + x_37 = x_44 = None + x_46 = x_45.permute(0, 2, 3, 1) + x_47 = torch.nn.functional.layer_norm( + x_46, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_5 = torch.nn.functional.relu(x_49, inplace=False) + x_49 = None + pow_6 = relu_5**2 + relu_5 = None + mul_5 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_6 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_6 + ) = None + x_50 = ( + mul_5 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = x_45 + x_53 + x_45 = x_53 = None + x_55 = x_54.permute(0, 2, 3, 1) + x_54 = None + x_56 = torch.nn.functional.layer_norm( + x_55, + (96,), + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) + x_57 = x_56.permute(0, 3, 1, 2) + x_56 = None + x_58 = torch.conv2d( + x_57, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_1 = sym_sum // 8 + sym_sum_2 = torch.sym_sum([1, floordiv_1]) + floordiv_1 = sym_sum_2 = None + x_59 = x_58.permute(0, 2, 3, 1) + x_60 = torch.nn.functional.layer_norm( + x_59, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_61 = x_60.permute(0, 3, 1, 2) + x_60 = None + x_62 = torch.conv2d( + x_61, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_6 = torch.nn.functional.relu(x_62, inplace=False) + x_62 = None + pow_7 = relu_6**2 + relu_6 = None + mul_6 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_7 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_7 + ) = None + x_63 = ( + mul_6 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_66 = x_58 + x_65 + x_58 = x_65 = None + x_67 = x_66.permute(0, 2, 3, 1) + x_68 = torch.nn.functional.layer_norm( + x_67, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + x_70 = torch.conv2d( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_7 = torch.nn.functional.relu(x_70, inplace=False) + x_70 = None + pow_8 = relu_7**2 + relu_7 = None + mul_7 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_8 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_8 + ) = None + x_71 = ( + mul_7 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + x_73 = torch.conv2d( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + x_75 = x_66 + x_74 + x_66 = x_74 = None + x_76 = x_75.permute(0, 2, 3, 1) + x_77 = torch.nn.functional.layer_norm( + x_76, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_8 = torch.nn.functional.relu(x_79, inplace=False) + x_79 = None + pow_9 = relu_8**2 + relu_8 = None + mul_8 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_9 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_9 + ) = None + x_80 = ( + mul_8 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_81 = torch.conv2d( + x_80, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_83 = x_75 + x_82 + x_75 = x_82 = None + x_84 = x_83.permute(0, 2, 3, 1) + x_85 = torch.nn.functional.layer_norm( + x_84, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_86 = x_85.permute(0, 3, 1, 2) + x_85 = None + x_87 = torch.conv2d( + x_86, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_9 = torch.nn.functional.relu(x_87, inplace=False) + x_87 = None + pow_10 = relu_9**2 + relu_9 = None + mul_9 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_10 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_10 + ) = None + x_88 = ( + mul_9 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = x_83 + x_91 + x_83 = x_91 = None + x_93 = x_92.permute(0, 2, 3, 1) + x_94 = torch.nn.functional.layer_norm( + x_93, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = torch.conv2d( + x_95, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_10 = torch.nn.functional.relu(x_96, inplace=False) + x_96 = None + pow_11 = relu_10**2 + relu_10 = None + mul_10 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_11 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_11 + ) = None + x_97 = ( + mul_10 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_98 = torch.conv2d( + x_97, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_99 = torch.conv2d( + x_98, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_100 = x_92 + x_99 + x_92 = x_99 = None + x_101 = x_100.permute(0, 2, 3, 1) + x_102 = torch.nn.functional.layer_norm( + x_101, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_103 = x_102.permute(0, 3, 1, 2) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_11 = torch.nn.functional.relu(x_104, inplace=False) + x_104 = None + pow_12 = relu_11**2 + relu_11 = None + mul_11 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_12 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_12 + ) = None + x_105 = ( + mul_11 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_100 + x_108 + x_100 = x_108 = None + x_110 = x_109.permute(0, 2, 3, 1) + x_111 = torch.nn.functional.layer_norm( + x_110, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_110 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) + x_112 = x_111.permute(0, 3, 1, 2) + x_111 = None + x_113 = torch.conv2d( + x_112, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_112 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_12 = torch.nn.functional.relu(x_113, inplace=False) + x_113 = None + pow_13 = relu_12**2 + relu_12 = None + mul_12 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ + * pow_13 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_13 + ) = None + x_114 = ( + mul_12 + + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_12 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_115 = torch.conv2d( + x_114, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_114 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_117 = x_109 + x_116 + x_109 = x_116 = None + x_118 = x_117.permute(0, 2, 3, 1) + x_119 = torch.nn.functional.layer_norm( + x_118, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_118 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) + x_120 = x_119.permute(0, 3, 1, 2) + x_119 = None + x_121 = torch.conv2d( + x_120, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_120 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_13 = torch.nn.functional.relu(x_121, inplace=False) + x_121 = None + pow_14 = relu_13**2 + relu_13 = None + mul_13 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + * pow_14 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( + pow_14 + ) = None + x_122 = ( + mul_13 + + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + ) + mul_13 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) + x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) + x_122 = None + x_124 = torch.conv2d( + x_123, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_123 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_126 = x_117 + x_125 + x_117 = x_125 = None + x_127 = x_126.permute(0, 2, 3, 1) + x_128 = torch.nn.functional.layer_norm( + x_127, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_127 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) + x_129 = x_128.permute(0, 3, 1, 2) + x_128 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_14 = torch.nn.functional.relu(x_130, inplace=False) + x_130 = None + pow_15 = relu_14**2 + relu_14 = None + mul_14 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ + * pow_15 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_15 + ) = None + x_131 = ( + mul_14 + + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_14 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_132 = torch.conv2d( + x_131, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_131 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_132 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_134 = x_126 + x_133 + x_126 = x_133 = None + x_135 = x_134.permute(0, 2, 3, 1) + x_136 = torch.nn.functional.layer_norm( + x_135, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_135 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) + x_137 = x_136.permute(0, 3, 1, 2) + x_136 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_15 = torch.nn.functional.relu(x_138, inplace=False) + x_138 = None + pow_16 = relu_15**2 + relu_15 = None + mul_15 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + * pow_16 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( + pow_16 + ) = None + x_139 = ( + mul_15 + + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + ) + mul_15 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) + x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) + x_139 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_140 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_134 + x_142 + x_134 = x_142 = None + x_144 = x_143.permute(0, 2, 3, 1) + x_145 = torch.nn.functional.layer_norm( + x_144, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_144 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) + x_146 = x_145.permute(0, 3, 1, 2) + x_145 = None + x_147 = torch.conv2d( + x_146, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_146 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_16 = torch.nn.functional.relu(x_147, inplace=False) + x_147 = None + pow_17 = relu_16**2 + relu_16 = None + mul_16 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ + * pow_17 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_17 + ) = None + x_148 = ( + mul_16 + + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_16 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_149 = torch.conv2d( + x_148, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_148 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_150 = torch.conv2d( + x_149, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_149 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_151 = x_143 + x_150 + x_143 = x_150 = None + x_152 = x_151.permute(0, 2, 3, 1) + x_153 = torch.nn.functional.layer_norm( + x_152, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_152 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) + x_154 = x_153.permute(0, 3, 1, 2) + x_153 = None + x_155 = torch.conv2d( + x_154, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_154 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_17 = torch.nn.functional.relu(x_155, inplace=False) + x_155 = None + pow_18 = relu_17**2 + relu_17 = None + mul_17 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + * pow_18 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( + pow_18 + ) = None + x_156 = ( + mul_17 + + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + ) + mul_17 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch.conv2d( + x_157, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_157 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_151 + x_159 + x_151 = x_159 = None + x_161 = x_160.permute(0, 2, 3, 1) + x_162 = torch.nn.functional.layer_norm( + x_161, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_161 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) + x_163 = x_162.permute(0, 3, 1, 2) + x_162 = None + x_164 = torch.conv2d( + x_163, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_163 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_18 = torch.nn.functional.relu(x_164, inplace=False) + x_164 = None + pow_19 = relu_18**2 + relu_18 = None + mul_18 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ + * pow_19 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_19 + ) = None + x_165 = ( + mul_18 + + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_18 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_166 = torch.conv2d( + x_165, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_165 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_167 = torch.conv2d( + x_166, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_166 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_168 = x_160 + x_167 + x_160 = x_167 = None + x_169 = x_168.permute(0, 2, 3, 1) + x_170 = torch.nn.functional.layer_norm( + x_169, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_169 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) + x_171 = x_170.permute(0, 3, 1, 2) + x_170 = None + x_172 = torch.conv2d( + x_171, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_171 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_19 = torch.nn.functional.relu(x_172, inplace=False) + x_172 = None + pow_20 = relu_19**2 + relu_19 = None + mul_19 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + * pow_20 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( + pow_20 + ) = None + x_173 = ( + mul_19 + + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + ) + mul_19 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) + x_174 = torch.nn.functional.dropout(x_173, 0.0, False, False) + x_173 = None + x_175 = torch.conv2d( + x_174, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_174 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = x_168 + x_176 + x_168 = x_176 = None + x_178 = x_177.permute(0, 2, 3, 1) + x_179 = torch.nn.functional.layer_norm( + x_178, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_178 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) + x_180 = x_179.permute(0, 3, 1, 2) + x_179 = None + x_181 = torch.conv2d( + x_180, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_180 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_20 = torch.nn.functional.relu(x_181, inplace=False) + x_181 = None + pow_21 = relu_20**2 + relu_20 = None + mul_20 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ + * pow_21 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_21 + ) = None + x_182 = ( + mul_20 + + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_20 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_183 = torch.conv2d( + x_182, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_182 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_184 = torch.conv2d( + x_183, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_183 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_185 = x_177 + x_184 + x_177 = x_184 = None + x_186 = x_185.permute(0, 2, 3, 1) + x_187 = torch.nn.functional.layer_norm( + x_186, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) + x_188 = x_187.permute(0, 3, 1, 2) + x_187 = None + x_189 = torch.conv2d( + x_188, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_188 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_21 = torch.nn.functional.relu(x_189, inplace=False) + x_189 = None + pow_22 = relu_21**2 + relu_21 = None + mul_21 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + * pow_22 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( + pow_22 + ) = None + x_190 = ( + mul_21 + + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + ) + mul_21 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) + x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) + x_190 = None + x_192 = torch.conv2d( + x_191, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_191 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_185 + x_193 + x_185 = x_193 = None + x_195 = x_194.permute(0, 2, 3, 1) + x_196 = torch.nn.functional.layer_norm( + x_195, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_195 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) + x_197 = x_196.permute(0, 3, 1, 2) + x_196 = None + x_198 = torch.conv2d( + x_197, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_197 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_22 = torch.nn.functional.relu(x_198, inplace=False) + x_198 = None + pow_23 = relu_22**2 + relu_22 = None + mul_22 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ + * pow_23 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_23 + ) = None + x_199 = ( + mul_22 + + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_22 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_200 = torch.conv2d( + x_199, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_199 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_201 = torch.conv2d( + x_200, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_200 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_202 = x_194 + x_201 + x_194 = x_201 = None + x_203 = x_202.permute(0, 2, 3, 1) + x_204 = torch.nn.functional.layer_norm( + x_203, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_203 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) + x_205 = x_204.permute(0, 3, 1, 2) + x_204 = None + x_206 = torch.conv2d( + x_205, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_205 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_23 = torch.nn.functional.relu(x_206, inplace=False) + x_206 = None + pow_24 = relu_23**2 + relu_23 = None + mul_23 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + * pow_24 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( + pow_24 + ) = None + x_207 = ( + mul_23 + + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + ) + mul_23 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) + x_208 = torch.nn.functional.dropout(x_207, 0.0, False, False) + x_207 = None + x_209 = torch.conv2d( + x_208, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_208 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) + x_209 = None + x_211 = x_202 + x_210 + x_202 = x_210 = None + x_212 = x_211.permute(0, 2, 3, 1) + x_213 = torch.nn.functional.layer_norm( + x_212, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_212 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) + x_214 = x_213.permute(0, 3, 1, 2) + x_213 = None + x_215 = torch.conv2d( + x_214, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_214 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_24 = torch.nn.functional.relu(x_215, inplace=False) + x_215 = None + pow_25 = relu_24**2 + relu_24 = None + mul_24 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ + * pow_25 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_25 + ) = None + x_216 = ( + mul_24 + + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_24 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_217 = torch.conv2d( + x_216, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_216 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_217 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_219 = x_211 + x_218 + x_211 = x_218 = None + x_220 = x_219.permute(0, 2, 3, 1) + x_221 = torch.nn.functional.layer_norm( + x_220, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_220 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) + x_222 = x_221.permute(0, 3, 1, 2) + x_221 = None + x_223 = torch.conv2d( + x_222, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_222 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_25 = torch.nn.functional.relu(x_223, inplace=False) + x_223 = None + pow_26 = relu_25**2 + relu_25 = None + mul_25 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + * pow_26 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( + pow_26 + ) = None + x_224 = ( + mul_25 + + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + ) + mul_25 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = torch.conv2d( + x_225, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_225 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) + x_226 = None + x_228 = x_219 + x_227 + x_219 = x_227 = None + x_229 = x_228.permute(0, 2, 3, 1) + x_230 = torch.nn.functional.layer_norm( + x_229, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_229 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) + x_231 = x_230.permute(0, 3, 1, 2) + x_230 = None + x_232 = torch.conv2d( + x_231, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_231 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_26 = torch.nn.functional.relu(x_232, inplace=False) + x_232 = None + pow_27 = relu_26**2 + relu_26 = None + mul_26 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ + * pow_27 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_27 + ) = None + x_233 = ( + mul_26 + + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_26 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_234 = torch.conv2d( + x_233, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_233 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_235 = torch.conv2d( + x_234, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_234 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_236 = x_228 + x_235 + x_228 = x_235 = None + x_237 = x_236.permute(0, 2, 3, 1) + x_238 = torch.nn.functional.layer_norm( + x_237, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_237 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) + x_239 = x_238.permute(0, 3, 1, 2) + x_238 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_239 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_27 = torch.nn.functional.relu(x_240, inplace=False) + x_240 = None + pow_28 = relu_27**2 + relu_27 = None + mul_27 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + * pow_28 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( + pow_28 + ) = None + x_241 = ( + mul_27 + + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + ) + mul_27 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) + x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) + x_241 = None + x_243 = torch.conv2d( + x_242, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_242 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_244 = torch.nn.functional.dropout(x_243, 0.0, False, False) + x_243 = None + x_245 = x_236 + x_244 + x_236 = x_244 = None + x_246 = x_245.permute(0, 2, 3, 1) + x_247 = torch.nn.functional.layer_norm( + x_246, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_246 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) + x_248 = x_247.permute(0, 3, 1, 2) + x_247 = None + x_249 = torch.conv2d( + x_248, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_248 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_28 = torch.nn.functional.relu(x_249, inplace=False) + x_249 = None + pow_29 = relu_28**2 + relu_28 = None + mul_28 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ + * pow_29 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_29 + ) = None + x_250 = ( + mul_28 + + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_28 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_250 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_252 = torch.conv2d( + x_251, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_251 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_253 = x_245 + x_252 + x_245 = x_252 = None + x_254 = x_253.permute(0, 2, 3, 1) + x_255 = torch.nn.functional.layer_norm( + x_254, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_254 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) + x_256 = x_255.permute(0, 3, 1, 2) + x_255 = None + x_257 = torch.conv2d( + x_256, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_256 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_29 = torch.nn.functional.relu(x_257, inplace=False) + x_257 = None + pow_30 = relu_29**2 + relu_29 = None + mul_29 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + * pow_30 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( + pow_30 + ) = None + x_258 = ( + mul_29 + + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + ) + mul_29 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = torch.conv2d( + x_259, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_259 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_261 = torch.nn.functional.dropout(x_260, 0.0, False, False) + x_260 = None + x_262 = x_253 + x_261 + x_253 = x_261 = None + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (192,), + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) + x_265 = x_264.permute(0, 3, 1, 2) + x_264 = None + x_266 = torch.conv2d( + x_265, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_265 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_2 = sym_sum // 16 + sym_sum_3 = torch.sym_sum([1, floordiv_2]) + floordiv_2 = None + reshape = x_266.reshape(1, 384, -1) + x_266 = None + x_267 = reshape.transpose(1, 2) + reshape = None + view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_30 = x_267 * view + view = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_267, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_52 = sym_sum_3**2 + linear = torch._C._nn.linear( + layer_norm_33, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_33 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_1 = linear.reshape(1, pow_52, 3, 12, 32) + linear = None + qkv = reshape_1.permute(2, 0, 3, 1, 4) + reshape_1 = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_268 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) + q = k = v = None + transpose_1 = x_268.transpose(1, 2) + x_268 = None + x_269 = transpose_1.reshape(1, pow_52, 384) + transpose_1 = None + x_270 = torch._C._nn.linear( + x_269, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_269 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_271 = torch.nn.functional.dropout(x_270, 0.0, False, False) + x_270 = None + x_272 = mul_30 + x_271 + mul_30 = x_271 = None + view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_31 = x_272 * view_1 + view_1 = None + layer_norm_34 = torch.nn.functional.layer_norm( + x_272, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_272 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_273 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_34 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_30 = torch.nn.functional.relu(x_273, inplace=False) + x_273 = None + pow_31 = relu_30**2 + relu_30 = None + mul_32 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_31 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_31 + ) = None + x_274 = ( + mul_32 + + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_275 = torch.nn.functional.dropout(x_274, 0.0, False, False) + x_274 = None + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_277 = torch.nn.functional.dropout(x_276, 0.0, False, False) + x_276 = None + x_278 = mul_31 + x_277 + mul_31 = x_277 = None + view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_33 = x_278 * view_2 + view_2 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_278, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_3 = linear_4.reshape(1, pow_52, 3, 12, 32) + linear_4 = None + qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) + reshape_3 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_279 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) + q_1 = k_1 = v_1 = None + transpose_2 = x_279.transpose(1, 2) + x_279 = None + x_280 = transpose_2.reshape(1, pow_52, 384) + transpose_2 = None + x_281 = torch._C._nn.linear( + x_280, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_280 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) + x_281 = None + x_283 = mul_33 + x_282 + mul_33 = x_282 = None + view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_34 = x_283 * view_3 + view_3 = None + layer_norm_36 = torch.nn.functional.layer_norm( + x_283, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_283 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_284 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_36 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_31 = torch.nn.functional.relu(x_284, inplace=False) + x_284 = None + pow_32 = relu_31**2 + relu_31 = None + mul_35 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_32 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_32 + ) = None + x_285 = ( + mul_35 + + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_286 = torch.nn.functional.dropout(x_285, 0.0, False, False) + x_285 = None + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) + x_287 = None + x_289 = mul_34 + x_288 + mul_34 = x_288 = None + view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_36 = x_289 * view_4 + view_4 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_289, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_37 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_5 = linear_8.reshape(1, pow_52, 3, 12, 32) + linear_8 = None + qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) + reshape_5 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_290 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) + q_2 = k_2 = v_2 = None + transpose_3 = x_290.transpose(1, 2) + x_290 = None + x_291 = transpose_3.reshape(1, pow_52, 384) + transpose_3 = None + x_292 = torch._C._nn.linear( + x_291, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_291 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_293 = torch.nn.functional.dropout(x_292, 0.0, False, False) + x_292 = None + x_294 = mul_36 + x_293 + mul_36 = x_293 = None + view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_37 = x_294 * view_5 + view_5 = None + layer_norm_38 = torch.nn.functional.layer_norm( + x_294, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_294 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_295 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_32 = torch.nn.functional.relu(x_295, inplace=False) + x_295 = None + pow_33 = relu_32**2 + relu_32 = None + mul_38 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_33 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_33 + ) = None + x_296 = ( + mul_38 + + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_297 = torch.nn.functional.dropout(x_296, 0.0, False, False) + x_296 = None + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_299 = torch.nn.functional.dropout(x_298, 0.0, False, False) + x_298 = None + x_300 = mul_37 + x_299 + mul_37 = x_299 = None + view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_39 = x_300 * view_6 + view_6 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_300, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_39 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_7 = linear_12.reshape(1, pow_52, 3, 12, 32) + linear_12 = None + qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) + reshape_7 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_301 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) + q_3 = k_3 = v_3 = None + transpose_4 = x_301.transpose(1, 2) + x_301 = None + x_302 = transpose_4.reshape(1, pow_52, 384) + transpose_4 = None + x_303 = torch._C._nn.linear( + x_302, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_302 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_304 = torch.nn.functional.dropout(x_303, 0.0, False, False) + x_303 = None + x_305 = mul_39 + x_304 + mul_39 = x_304 = None + view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_40 = x_305 * view_7 + view_7 = None + layer_norm_40 = torch.nn.functional.layer_norm( + x_305, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_305 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) + x_306 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_40 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_33 = torch.nn.functional.relu(x_306, inplace=False) + x_306 = None + pow_34 = relu_33**2 + relu_33 = None + mul_41 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + * pow_34 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( + pow_34 + ) = None + x_307 = ( + mul_41 + + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + ) + mul_41 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) + x_308 = torch.nn.functional.dropout(x_307, 0.0, False, False) + x_307 = None + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) + x_309 = None + x_311 = mul_40 + x_310 + mul_40 = x_310 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_42 = x_311 * view_8 + view_8 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_311, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_41 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_9 = linear_16.reshape(1, pow_52, 3, 12, 32) + linear_16 = None + qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) + reshape_9 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_312 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) + q_4 = k_4 = v_4 = None + transpose_5 = x_312.transpose(1, 2) + x_312 = None + x_313 = transpose_5.reshape(1, pow_52, 384) + transpose_5 = None + x_314 = torch._C._nn.linear( + x_313, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_313 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) + x_314 = None + x_316 = mul_42 + x_315 + mul_42 = x_315 = None + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_43 = x_316 * view_9 + view_9 = None + layer_norm_42 = torch.nn.functional.layer_norm( + x_316, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_316 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) + x_317 = torch._C._nn.linear( + layer_norm_42, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_42 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_34 = torch.nn.functional.relu(x_317, inplace=False) + x_317 = None + pow_35 = relu_34**2 + relu_34 = None + mul_44 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + * pow_35 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( + pow_35 + ) = None + x_318 = ( + mul_44 + + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + ) + mul_44 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) + x_319 = torch.nn.functional.dropout(x_318, 0.0, False, False) + x_318 = None + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_321 = torch.nn.functional.dropout(x_320, 0.0, False, False) + x_320 = None + x_322 = mul_43 + x_321 + mul_43 = x_321 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_45 = x_322 * view_10 + view_10 = None + layer_norm_43 = torch.nn.functional.layer_norm( + x_322, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) + linear_20 = torch._C._nn.linear( + layer_norm_43, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_43 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_11 = linear_20.reshape(1, pow_52, 3, 12, 32) + linear_20 = None + qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) + reshape_11 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_323 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) + q_5 = k_5 = v_5 = None + transpose_6 = x_323.transpose(1, 2) + x_323 = None + x_324 = transpose_6.reshape(1, pow_52, 384) + transpose_6 = None + x_325 = torch._C._nn.linear( + x_324, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_324 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_326 = torch.nn.functional.dropout(x_325, 0.0, False, False) + x_325 = None + x_327 = mul_45 + x_326 + mul_45 = x_326 = None + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_46 = x_327 * view_11 + view_11 = None + layer_norm_44 = torch.nn.functional.layer_norm( + x_327, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_327 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) + x_328 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_44 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_35 = torch.nn.functional.relu(x_328, inplace=False) + x_328 = None + pow_36 = relu_35**2 + relu_35 = None + mul_47 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + * pow_36 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( + pow_36 + ) = None + x_329 = ( + mul_47 + + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + ) + mul_47 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) + x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) + x_329 = None + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_332 = torch.nn.functional.dropout(x_331, 0.0, False, False) + x_331 = None + x_333 = mul_46 + x_332 + mul_46 = x_332 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_48 = x_333 * view_12 + view_12 = None + layer_norm_45 = torch.nn.functional.layer_norm( + x_333, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) + linear_24 = torch._C._nn.linear( + layer_norm_45, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_45 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_13 = linear_24.reshape(1, pow_52, 3, 12, 32) + linear_24 = None + qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) + reshape_13 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_334 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) + q_6 = k_6 = v_6 = None + transpose_7 = x_334.transpose(1, 2) + x_334 = None + x_335 = transpose_7.reshape(1, pow_52, 384) + transpose_7 = None + x_336 = torch._C._nn.linear( + x_335, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_335 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_337 = torch.nn.functional.dropout(x_336, 0.0, False, False) + x_336 = None + x_338 = mul_48 + x_337 + mul_48 = x_337 = None + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_49 = x_338 * view_13 + view_13 = None + layer_norm_46 = torch.nn.functional.layer_norm( + x_338, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_338 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) + x_339 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_46 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_36 = torch.nn.functional.relu(x_339, inplace=False) + x_339 = None + pow_37 = relu_36**2 + relu_36 = None + mul_50 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + * pow_37 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( + pow_37 + ) = None + x_340 = ( + mul_50 + + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + ) + mul_50 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) + x_341 = torch.nn.functional.dropout(x_340, 0.0, False, False) + x_340 = None + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_343 = torch.nn.functional.dropout(x_342, 0.0, False, False) + x_342 = None + x_344 = mul_49 + x_343 + mul_49 = x_343 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_51 = x_344 * view_14 + view_14 = None + layer_norm_47 = torch.nn.functional.layer_norm( + x_344, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) + linear_28 = torch._C._nn.linear( + layer_norm_47, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_47 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_15 = linear_28.reshape(1, pow_52, 3, 12, 32) + linear_28 = None + qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) + reshape_15 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_345 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) + q_7 = k_7 = v_7 = None + transpose_8 = x_345.transpose(1, 2) + x_345 = None + x_346 = transpose_8.reshape(1, pow_52, 384) + transpose_8 = None + x_347 = torch._C._nn.linear( + x_346, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_346 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) + x_347 = None + x_349 = mul_51 + x_348 + mul_51 = x_348 = None + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_52 = x_349 * view_15 + view_15 = None + layer_norm_48 = torch.nn.functional.layer_norm( + x_349, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_349 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) + x_350 = torch._C._nn.linear( + layer_norm_48, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_48 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_37 = torch.nn.functional.relu(x_350, inplace=False) + x_350 = None + pow_38 = relu_37**2 + relu_37 = None + mul_53 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + * pow_38 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( + pow_38 + ) = None + x_351 = ( + mul_53 + + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + ) + mul_53 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) + x_352 = torch.nn.functional.dropout(x_351, 0.0, False, False) + x_351 = None + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_354 = torch.nn.functional.dropout(x_353, 0.0, False, False) + x_353 = None + x_355 = mul_52 + x_354 + mul_52 = x_354 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_54 = x_355 * view_16 + view_16 = None + layer_norm_49 = torch.nn.functional.layer_norm( + x_355, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) + linear_32 = torch._C._nn.linear( + layer_norm_49, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_49 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_17 = linear_32.reshape(1, pow_52, 3, 12, 32) + linear_32 = None + qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) + reshape_17 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_356 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) + q_8 = k_8 = v_8 = None + transpose_9 = x_356.transpose(1, 2) + x_356 = None + x_357 = transpose_9.reshape(1, pow_52, 384) + transpose_9 = None + x_358 = torch._C._nn.linear( + x_357, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_357 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_359 = torch.nn.functional.dropout(x_358, 0.0, False, False) + x_358 = None + x_360 = mul_54 + x_359 + mul_54 = x_359 = None + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_55 = x_360 * view_17 + view_17 = None + layer_norm_50 = torch.nn.functional.layer_norm( + x_360, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_360 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) + x_361 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_50 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_38 = torch.nn.functional.relu(x_361, inplace=False) + x_361 = None + pow_39 = relu_38**2 + relu_38 = None + mul_56 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + * pow_39 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( + pow_39 + ) = None + x_362 = ( + mul_56 + + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + ) + mul_56 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) + x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) + x_362 = None + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_365 = torch.nn.functional.dropout(x_364, 0.0, False, False) + x_364 = None + x_366 = mul_55 + x_365 + mul_55 = x_365 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_57 = x_366 * view_18 + view_18 = None + layer_norm_51 = torch.nn.functional.layer_norm( + x_366, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_51, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_51 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_19 = linear_36.reshape(1, pow_52, 3, 12, 32) + linear_36 = None + qkv_9 = reshape_19.permute(2, 0, 3, 1, 4) + reshape_19 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_367 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) + q_9 = k_9 = v_9 = None + transpose_10 = x_367.transpose(1, 2) + x_367 = None + x_368 = transpose_10.reshape(1, pow_52, 384) + transpose_10 = None + x_369 = torch._C._nn.linear( + x_368, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_368 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_370 = torch.nn.functional.dropout(x_369, 0.0, False, False) + x_369 = None + x_371 = mul_57 + x_370 + mul_57 = x_370 = None + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_58 = x_371 * view_19 + view_19 = None + layer_norm_52 = torch.nn.functional.layer_norm( + x_371, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_371 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) + x_372 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_52 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_39 = torch.nn.functional.relu(x_372, inplace=False) + x_372 = None + pow_40 = relu_39**2 + relu_39 = None + mul_59 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + * pow_40 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( + pow_40 + ) = None + x_373 = ( + mul_59 + + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + ) + mul_59 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) + x_374 = torch.nn.functional.dropout(x_373, 0.0, False, False) + x_373 = None + x_375 = torch._C._nn.linear( + x_374, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_374 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_376 = torch.nn.functional.dropout(x_375, 0.0, False, False) + x_375 = None + x_377 = mul_58 + x_376 + mul_58 = x_376 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_60 = x_377 * view_20 + view_20 = None + layer_norm_53 = torch.nn.functional.layer_norm( + x_377, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_377 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_53, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_53 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_21 = linear_40.reshape(1, pow_52, 3, 12, 32) + linear_40 = None + qkv_10 = reshape_21.permute(2, 0, 3, 1, 4) + reshape_21 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_378 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_378.transpose(1, 2) + x_378 = None + x_379 = transpose_11.reshape(1, pow_52, 384) + transpose_11 = None + x_380 = torch._C._nn.linear( + x_379, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_379 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_381 = torch.nn.functional.dropout(x_380, 0.0, False, False) + x_380 = None + x_382 = mul_60 + x_381 + mul_60 = x_381 = None + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_61 = x_382 * view_21 + view_21 = None + layer_norm_54 = torch.nn.functional.layer_norm( + x_382, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_382 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) + x_383 = torch._C._nn.linear( + layer_norm_54, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_54 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_40 = torch.nn.functional.relu(x_383, inplace=False) + x_383 = None + pow_41 = relu_40**2 + relu_40 = None + mul_62 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + * pow_41 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( + pow_41 + ) = None + x_384 = ( + mul_62 + + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + ) + mul_62 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) + x_385 = torch.nn.functional.dropout(x_384, 0.0, False, False) + x_384 = None + x_386 = torch._C._nn.linear( + x_385, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_385 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_387 = torch.nn.functional.dropout(x_386, 0.0, False, False) + x_386 = None + x_388 = mul_61 + x_387 + mul_61 = x_387 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_63 = x_388 * view_22 + view_22 = None + layer_norm_55 = torch.nn.functional.layer_norm( + x_388, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_388 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_55, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_55 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_23 = linear_44.reshape(1, pow_52, 3, 12, 32) + linear_44 = None + qkv_11 = reshape_23.permute(2, 0, 3, 1, 4) + reshape_23 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_389 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_389.transpose(1, 2) + x_389 = None + x_390 = transpose_12.reshape(1, pow_52, 384) + transpose_12 = None + x_391 = torch._C._nn.linear( + x_390, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_390 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_392 = torch.nn.functional.dropout(x_391, 0.0, False, False) + x_391 = None + x_393 = mul_63 + x_392 + mul_63 = x_392 = None + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_64 = x_393 * view_23 + view_23 = None + layer_norm_56 = torch.nn.functional.layer_norm( + x_393, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_393 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) + x_394 = torch._C._nn.linear( + layer_norm_56, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_56 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_41 = torch.nn.functional.relu(x_394, inplace=False) + x_394 = None + pow_42 = relu_41**2 + relu_41 = None + mul_65 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + * pow_42 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( + pow_42 + ) = None + x_395 = ( + mul_65 + + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + ) + mul_65 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) + x_396 = torch.nn.functional.dropout(x_395, 0.0, False, False) + x_395 = None + x_397 = torch._C._nn.linear( + x_396, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_396 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_398 = torch.nn.functional.dropout(x_397, 0.0, False, False) + x_397 = None + x_399 = mul_64 + x_398 + mul_64 = x_398 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_66 = x_399 * view_24 + view_24 = None + layer_norm_57 = torch.nn.functional.layer_norm( + x_399, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_399 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_57, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_57 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_25 = linear_48.reshape(1, pow_52, 3, 12, 32) + linear_48 = None + qkv_12 = reshape_25.permute(2, 0, 3, 1, 4) + reshape_25 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_12 = unbind_12[0] + k_12 = unbind_12[1] + v_12 = unbind_12[2] + unbind_12 = None + x_400 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_13 = x_400.transpose(1, 2) + x_400 = None + x_401 = transpose_13.reshape(1, pow_52, 384) + transpose_13 = None + x_402 = torch._C._nn.linear( + x_401, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_401 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_403 = torch.nn.functional.dropout(x_402, 0.0, False, False) + x_402 = None + x_404 = mul_66 + x_403 + mul_66 = x_403 = None + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_67 = x_404 * view_25 + view_25 = None + layer_norm_58 = torch.nn.functional.layer_norm( + x_404, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_404 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = (None) + x_405 = torch._C._nn.linear( + layer_norm_58, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_58 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_42 = torch.nn.functional.relu(x_405, inplace=False) + x_405 = None + pow_43 = relu_42**2 + relu_42 = None + mul_68 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ + * pow_43 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = ( + pow_43 + ) = None + x_406 = ( + mul_68 + + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ + ) + mul_68 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = (None) + x_407 = torch.nn.functional.dropout(x_406, 0.0, False, False) + x_406 = None + x_408 = torch._C._nn.linear( + x_407, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_407 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_409 = torch.nn.functional.dropout(x_408, 0.0, False, False) + x_408 = None + x_410 = mul_67 + x_409 + mul_67 = x_409 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_69 = x_410 * view_26 + view_26 = None + layer_norm_59 = torch.nn.functional.layer_norm( + x_410, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_410 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = (None) + linear_52 = torch._C._nn.linear( + layer_norm_59, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_59 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_27 = linear_52.reshape(1, pow_52, 3, 12, 32) + linear_52 = None + qkv_13 = reshape_27.permute(2, 0, 3, 1, 4) + reshape_27 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_13 = unbind_13[0] + k_13 = unbind_13[1] + v_13 = unbind_13[2] + unbind_13 = None + x_411 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_14 = x_411.transpose(1, 2) + x_411 = None + x_412 = transpose_14.reshape(1, pow_52, 384) + transpose_14 = None + x_413 = torch._C._nn.linear( + x_412, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_412 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_414 = torch.nn.functional.dropout(x_413, 0.0, False, False) + x_413 = None + x_415 = mul_69 + x_414 + mul_69 = x_414 = None + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_70 = x_415 * view_27 + view_27 = None + layer_norm_60 = torch.nn.functional.layer_norm( + x_415, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_415 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = (None) + x_416 = torch._C._nn.linear( + layer_norm_60, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_60 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_43 = torch.nn.functional.relu(x_416, inplace=False) + x_416 = None + pow_44 = relu_43**2 + relu_43 = None + mul_71 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ + * pow_44 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = ( + pow_44 + ) = None + x_417 = ( + mul_71 + + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ + ) + mul_71 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = (None) + x_418 = torch.nn.functional.dropout(x_417, 0.0, False, False) + x_417 = None + x_419 = torch._C._nn.linear( + x_418, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_418 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_420 = torch.nn.functional.dropout(x_419, 0.0, False, False) + x_419 = None + x_421 = mul_70 + x_420 + mul_70 = x_420 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_72 = x_421 * view_28 + view_28 = None + layer_norm_61 = torch.nn.functional.layer_norm( + x_421, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_421 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_61, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_61 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_29 = linear_56.reshape(1, pow_52, 3, 12, 32) + linear_56 = None + qkv_14 = reshape_29.permute(2, 0, 3, 1, 4) + reshape_29 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_14 = unbind_14[0] + k_14 = unbind_14[1] + v_14 = unbind_14[2] + unbind_14 = None + x_422 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_15 = x_422.transpose(1, 2) + x_422 = None + x_423 = transpose_15.reshape(1, pow_52, 384) + transpose_15 = None + x_424 = torch._C._nn.linear( + x_423, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_423 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_425 = torch.nn.functional.dropout(x_424, 0.0, False, False) + x_424 = None + x_426 = mul_72 + x_425 + mul_72 = x_425 = None + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_73 = x_426 * view_29 + view_29 = None + layer_norm_62 = torch.nn.functional.layer_norm( + x_426, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_426 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = (None) + x_427 = torch._C._nn.linear( + layer_norm_62, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_62 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_44 = torch.nn.functional.relu(x_427, inplace=False) + x_427 = None + pow_45 = relu_44**2 + relu_44 = None + mul_74 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ + * pow_45 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = ( + pow_45 + ) = None + x_428 = ( + mul_74 + + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ + ) + mul_74 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = (None) + x_429 = torch.nn.functional.dropout(x_428, 0.0, False, False) + x_428 = None + x_430 = torch._C._nn.linear( + x_429, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_429 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_431 = torch.nn.functional.dropout(x_430, 0.0, False, False) + x_430 = None + x_432 = mul_73 + x_431 + mul_73 = x_431 = None + view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_75 = x_432 * view_30 + view_30 = None + layer_norm_63 = torch.nn.functional.layer_norm( + x_432, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_432 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = (None) + linear_60 = torch._C._nn.linear( + layer_norm_63, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_63 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_31 = linear_60.reshape(1, pow_52, 3, 12, 32) + linear_60 = None + qkv_15 = reshape_31.permute(2, 0, 3, 1, 4) + reshape_31 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_15 = unbind_15[0] + k_15 = unbind_15[1] + v_15 = unbind_15[2] + unbind_15 = None + x_433 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_16 = x_433.transpose(1, 2) + x_433 = None + x_434 = transpose_16.reshape(1, pow_52, 384) + transpose_16 = None + x_435 = torch._C._nn.linear( + x_434, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_434 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_436 = torch.nn.functional.dropout(x_435, 0.0, False, False) + x_435 = None + x_437 = mul_75 + x_436 + mul_75 = x_436 = None + view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_76 = x_437 * view_31 + view_31 = None + layer_norm_64 = torch.nn.functional.layer_norm( + x_437, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_437 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = (None) + x_438 = torch._C._nn.linear( + layer_norm_64, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_64 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_45 = torch.nn.functional.relu(x_438, inplace=False) + x_438 = None + pow_46 = relu_45**2 + relu_45 = None + mul_77 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ + * pow_46 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = ( + pow_46 + ) = None + x_439 = ( + mul_77 + + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ + ) + mul_77 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = (None) + x_440 = torch.nn.functional.dropout(x_439, 0.0, False, False) + x_439 = None + x_441 = torch._C._nn.linear( + x_440, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_440 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_442 = torch.nn.functional.dropout(x_441, 0.0, False, False) + x_441 = None + x_443 = mul_76 + x_442 + mul_76 = x_442 = None + view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_78 = x_443 * view_32 + view_32 = None + layer_norm_65 = torch.nn.functional.layer_norm( + x_443, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_443 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_65, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_65 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_33 = linear_64.reshape(1, pow_52, 3, 12, 32) + linear_64 = None + qkv_16 = reshape_33.permute(2, 0, 3, 1, 4) + reshape_33 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_16 = unbind_16[0] + k_16 = unbind_16[1] + v_16 = unbind_16[2] + unbind_16 = None + x_444 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_17 = x_444.transpose(1, 2) + x_444 = None + x_445 = transpose_17.reshape(1, pow_52, 384) + transpose_17 = None + x_446 = torch._C._nn.linear( + x_445, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_445 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_447 = torch.nn.functional.dropout(x_446, 0.0, False, False) + x_446 = None + x_448 = mul_78 + x_447 + mul_78 = x_447 = None + view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_79 = x_448 * view_33 + view_33 = None + layer_norm_66 = torch.nn.functional.layer_norm( + x_448, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_448 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = (None) + x_449 = torch._C._nn.linear( + layer_norm_66, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_66 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_46 = torch.nn.functional.relu(x_449, inplace=False) + x_449 = None + pow_47 = relu_46**2 + relu_46 = None + mul_80 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ + * pow_47 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = ( + pow_47 + ) = None + x_450 = ( + mul_80 + + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ + ) + mul_80 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = (None) + x_451 = torch.nn.functional.dropout(x_450, 0.0, False, False) + x_450 = None + x_452 = torch._C._nn.linear( + x_451, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_451 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_453 = torch.nn.functional.dropout(x_452, 0.0, False, False) + x_452 = None + x_454 = mul_79 + x_453 + mul_79 = x_453 = None + view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_81 = x_454 * view_34 + view_34 = None + layer_norm_67 = torch.nn.functional.layer_norm( + x_454, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_454 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = (None) + linear_68 = torch._C._nn.linear( + layer_norm_67, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_67 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_35 = linear_68.reshape(1, pow_52, 3, 12, 32) + linear_68 = None + qkv_17 = reshape_35.permute(2, 0, 3, 1, 4) + reshape_35 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_17 = unbind_17[0] + k_17 = unbind_17[1] + v_17 = unbind_17[2] + unbind_17 = None + x_455 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_18 = x_455.transpose(1, 2) + x_455 = None + x_456 = transpose_18.reshape(1, pow_52, 384) + transpose_18 = pow_52 = None + x_457 = torch._C._nn.linear( + x_456, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_456 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_458 = torch.nn.functional.dropout(x_457, 0.0, False, False) + x_457 = None + x_459 = mul_81 + x_458 + mul_81 = x_458 = None + view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_82 = x_459 * view_35 + view_35 = None + layer_norm_68 = torch.nn.functional.layer_norm( + x_459, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_459 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = (None) + x_460 = torch._C._nn.linear( + layer_norm_68, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_68 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_47 = torch.nn.functional.relu(x_460, inplace=False) + x_460 = None + pow_48 = relu_47**2 + relu_47 = None + mul_83 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ + * pow_48 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = ( + pow_48 + ) = None + x_461 = ( + mul_83 + + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ + ) + mul_83 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = (None) + x_462 = torch.nn.functional.dropout(x_461, 0.0, False, False) + x_461 = None + x_463 = torch._C._nn.linear( + x_462, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_462 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_464 = torch.nn.functional.dropout(x_463, 0.0, False, False) + x_463 = None + x_465 = mul_82 + x_464 + mul_82 = x_464 = None + transpose_19 = x_465.transpose(1, 2) + x_465 = None + x_466 = transpose_19.reshape(1, 384, sym_sum_3, sym_sum_3) + transpose_19 = sym_sum_3 = None + x_467 = x_466.permute(0, 2, 3, 1) + x_466 = None + x_468 = torch.nn.functional.layer_norm( + x_467, + (384,), + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_467 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) + x_469 = x_468.permute(0, 3, 1, 2) + x_468 = None + x_470 = torch.conv2d( + x_469, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_469 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_3 = sym_sum // 32 + sym_sum = None + sym_sum_4 = torch.sym_sum([1, floordiv_3]) + floordiv_3 = None + reshape_38 = x_470.reshape(1, 576, -1) + x_470 = None + x_471 = reshape_38.transpose(1, 2) + reshape_38 = None + view_36 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_84 = x_471 * view_36 + view_36 = None + layer_norm_70 = torch.nn.functional.layer_norm( + x_471, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_471 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_53 = sym_sum_4**2 + linear_72 = torch._C._nn.linear( + layer_norm_70, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_70 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_39 = linear_72.reshape(1, pow_53, 3, 18, 32) + linear_72 = None + qkv_18 = reshape_39.permute(2, 0, 3, 1, 4) + reshape_39 = None + unbind_18 = qkv_18.unbind(0) + qkv_18 = None + q_18 = unbind_18[0] + k_18 = unbind_18[1] + v_18 = unbind_18[2] + unbind_18 = None + x_472 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_21 = x_472.transpose(1, 2) + x_472 = None + x_473 = transpose_21.reshape(1, pow_53, 576) + transpose_21 = None + x_474 = torch._C._nn.linear( + x_473, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_473 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_475 = torch.nn.functional.dropout(x_474, 0.0, False, False) + x_474 = None + x_476 = mul_84 + x_475 + mul_84 = x_475 = None + view_37 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_85 = x_476 * view_37 + view_37 = None + layer_norm_71 = torch.nn.functional.layer_norm( + x_476, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_476 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_477 = torch._C._nn.linear( + layer_norm_71, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_71 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_48 = torch.nn.functional.relu(x_477, inplace=False) + x_477 = None + pow_49 = relu_48**2 + relu_48 = None + mul_86 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_49 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_49 + ) = None + x_478 = ( + mul_86 + + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_86 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_479 = torch.nn.functional.dropout(x_478, 0.0, False, False) + x_478 = None + x_480 = torch._C._nn.linear( + x_479, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_479 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_481 = torch.nn.functional.dropout(x_480, 0.0, False, False) + x_480 = None + x_482 = mul_85 + x_481 + mul_85 = x_481 = None + view_38 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_87 = x_482 * view_38 + view_38 = None + layer_norm_72 = torch.nn.functional.layer_norm( + x_482, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_482 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_76 = torch._C._nn.linear( + layer_norm_72, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_72 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_41 = linear_76.reshape(1, pow_53, 3, 18, 32) + linear_76 = None + qkv_19 = reshape_41.permute(2, 0, 3, 1, 4) + reshape_41 = None + unbind_19 = qkv_19.unbind(0) + qkv_19 = None + q_19 = unbind_19[0] + k_19 = unbind_19[1] + v_19 = unbind_19[2] + unbind_19 = None + x_483 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_22 = x_483.transpose(1, 2) + x_483 = None + x_484 = transpose_22.reshape(1, pow_53, 576) + transpose_22 = None + x_485 = torch._C._nn.linear( + x_484, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_484 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_486 = torch.nn.functional.dropout(x_485, 0.0, False, False) + x_485 = None + x_487 = mul_87 + x_486 + mul_87 = x_486 = None + view_39 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_88 = x_487 * view_39 + view_39 = None + layer_norm_73 = torch.nn.functional.layer_norm( + x_487, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_487 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_488 = torch._C._nn.linear( + layer_norm_73, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_73 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_49 = torch.nn.functional.relu(x_488, inplace=False) + x_488 = None + pow_50 = relu_49**2 + relu_49 = None + mul_89 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_50 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_50 + ) = None + x_489 = ( + mul_89 + + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_89 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_490 = torch.nn.functional.dropout(x_489, 0.0, False, False) + x_489 = None + x_491 = torch._C._nn.linear( + x_490, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_490 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_492 = torch.nn.functional.dropout(x_491, 0.0, False, False) + x_491 = None + x_493 = mul_88 + x_492 + mul_88 = x_492 = None + view_40 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_90 = x_493 * view_40 + view_40 = None + layer_norm_74 = torch.nn.functional.layer_norm( + x_493, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_493 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_74, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_74 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_43 = linear_80.reshape(1, pow_53, 3, 18, 32) + linear_80 = None + qkv_20 = reshape_43.permute(2, 0, 3, 1, 4) + reshape_43 = None + unbind_20 = qkv_20.unbind(0) + qkv_20 = None + q_20 = unbind_20[0] + k_20 = unbind_20[1] + v_20 = unbind_20[2] + unbind_20 = None + x_494 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_23 = x_494.transpose(1, 2) + x_494 = None + x_495 = transpose_23.reshape(1, pow_53, 576) + transpose_23 = pow_53 = None + x_496 = torch._C._nn.linear( + x_495, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_495 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_497 = torch.nn.functional.dropout(x_496, 0.0, False, False) + x_496 = None + x_498 = mul_90 + x_497 + mul_90 = x_497 = None + view_41 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_91 = x_498 * view_41 + view_41 = None + layer_norm_75 = torch.nn.functional.layer_norm( + x_498, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_498 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_499 = torch._C._nn.linear( + layer_norm_75, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_75 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_50 = torch.nn.functional.relu(x_499, inplace=False) + x_499 = None + pow_51 = relu_50**2 + relu_50 = None + mul_92 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_51 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_51 + ) = None + x_500 = ( + mul_92 + + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_92 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_501 = torch.nn.functional.dropout(x_500, 0.0, False, False) + x_500 = None + x_502 = torch._C._nn.linear( + x_501, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_501 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_503 = torch.nn.functional.dropout(x_502, 0.0, False, False) + x_502 = None + x_504 = mul_91 + x_503 + mul_91 = x_503 = None + transpose_24 = x_504.transpose(1, 2) + x_504 = None + x_505 = transpose_24.reshape(1, 576, sym_sum_4, sym_sum_4) + transpose_24 = sym_sum_4 = None + x_506 = torch.nn.functional.adaptive_avg_pool2d(x_505, 1) + x_505 = None + x_507 = x_506.permute(0, 2, 3, 1) + x_506 = None + x_508 = torch.nn.functional.layer_norm( + x_507, + (576,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_507 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_509 = x_508.permute(0, 3, 1, 2) + x_508 = None + x_510 = x_509.flatten(1, -1) + x_509 = None + x_511 = torch.nn.functional.dropout(x_510, 0.0, False, False) + x_510 = None + x_512 = torch._C._nn.linear( + x_511, + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, + ) + x_511 = ( + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None + relu_51 = torch.nn.functional.relu(x_512, inplace=False) + x_512 = None + x_513 = torch.square(relu_51) + relu_51 = None + x_514 = torch.nn.functional.layer_norm( + x_513, + (2304,), + l_self_modules_head_modules_fc_modules_norm_parameters_weight_, + l_self_modules_head_modules_fc_modules_norm_parameters_bias_, + 1e-06, + ) + x_513 = ( + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None + x_515 = torch.nn.functional.dropout(x_514, 0.0, False, False) + x_514 = None + x_516 = torch._C._nn.linear( + x_515, + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, + ) + x_515 = ( + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None + return (x_516,) diff --git a/samples/timm/caformer_m36.sail_in22k_ft_in1k/weight_meta.py b/samples/timm/caformer_m36.sail_in22k_ft_in1k/weight_meta.py new file mode 100644 index 0000000000..aa8ca0fde6 --- /dev/null +++ b/samples/timm/caformer_m36.sail_in22k_ft_in1k/weight_meta.py @@ -0,0 +1,3968 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv_parameters_weight_" + shape = [96, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: + name = "L_self_modules_stem_modules_conv_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.228 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: + name = "L_self_modules_stem_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [192, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [96, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [192, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [96, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [192, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [96, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" + shape = [192, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" + shape = [384, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" + shape = [576, 384, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1728, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [576, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2304, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [576, 2304] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1728, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [576, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2304, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [576, 2304] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1728, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [576, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2304, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [576, 2304] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" + shape = [2304, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" + shape = [1000, 2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/graph_net.json b/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/input_meta.py b/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/input_tensor_constraints.py b/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/model.py b/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/model.py new file mode 100644 index 0000000000..dcce286c5c --- /dev/null +++ b/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/model.py @@ -0,0 +1,4754 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv_parameters_bias_ = ( + L_self_modules_stem_modules_conv_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_norm_parameters_weight_ = ( + L_self_modules_stem_modules_norm_parameters_weight_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv_parameters_weight_, + l_self_modules_stem_modules_conv_parameters_bias_, + (4, 4), + (2, 2), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_conv_parameters_weight_ + ) = l_self_modules_stem_modules_conv_parameters_bias_ = None + x_1 = x.permute(0, 2, 3, 1) + x = None + x_2 = torch.nn.functional.layer_norm( + x_1, (96,), l_self_modules_stem_modules_norm_parameters_weight_, None, 1e-06 + ) + x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None + x_3 = x_2.permute(0, 3, 1, 2) + x_2 = None + sym_sum = torch.sym_sum([-3, s1]) + s1 = None + floordiv = sym_sum // 4 + sym_sum_1 = torch.sym_sum([1, floordiv]) + floordiv = sym_sum_1 = None + x_4 = x_3.permute(0, 2, 3, 1) + x_5 = torch.nn.functional.layer_norm( + x_4, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu = torch.nn.functional.relu(x_7, inplace=False) + x_7 = None + pow_1 = relu**2 + relu = None + mul = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_1 + ) = None + x_8 = ( + mul + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_11 = x_3 + x_10 + x_3 = x_10 = None + x_12 = x_11.permute(0, 2, 3, 1) + x_13 = torch.nn.functional.layer_norm( + x_12, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_14 = x_13.permute(0, 3, 1, 2) + x_13 = None + x_15 = torch.conv2d( + x_14, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_1 = torch.nn.functional.relu(x_15, inplace=False) + x_15 = None + pow_2 = relu_1**2 + relu_1 = None + mul_1 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_2 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_2 + ) = None + x_16 = ( + mul_1 + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = x_11 + x_19 + x_11 = x_19 = None + x_21 = x_20.permute(0, 2, 3, 1) + x_22 = torch.nn.functional.layer_norm( + x_21, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = torch.conv2d( + x_23, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_2 = torch.nn.functional.relu(x_24, inplace=False) + x_24 = None + pow_3 = relu_2**2 + relu_2 = None + mul_2 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_3 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_3 + ) = None + x_25 = ( + mul_2 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_27 = torch.conv2d( + x_26, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_28 = x_20 + x_27 + x_20 = x_27 = None + x_29 = x_28.permute(0, 2, 3, 1) + x_30 = torch.nn.functional.layer_norm( + x_29, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_3 = torch.nn.functional.relu(x_32, inplace=False) + x_32 = None + pow_4 = relu_3**2 + relu_3 = None + mul_3 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_4 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_4 + ) = None + x_33 = ( + mul_3 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + x_37 = x_28 + x_36 + x_28 = x_36 = None + x_38 = x_37.permute(0, 2, 3, 1) + x_39 = torch.nn.functional.layer_norm( + x_38, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_40 = x_39.permute(0, 3, 1, 2) + x_39 = None + x_41 = torch.conv2d( + x_40, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_4 = torch.nn.functional.relu(x_41, inplace=False) + x_41 = None + pow_5 = relu_4**2 + relu_4 = None + mul_4 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_5 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_5 + ) = None + x_42 = ( + mul_4 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_44 = torch.conv2d( + x_43, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_45 = x_37 + x_44 + x_37 = x_44 = None + x_46 = x_45.permute(0, 2, 3, 1) + x_47 = torch.nn.functional.layer_norm( + x_46, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_5 = torch.nn.functional.relu(x_49, inplace=False) + x_49 = None + pow_6 = relu_5**2 + relu_5 = None + mul_5 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_6 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_6 + ) = None + x_50 = ( + mul_5 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = x_45 + x_53 + x_45 = x_53 = None + x_55 = x_54.permute(0, 2, 3, 1) + x_54 = None + x_56 = torch.nn.functional.layer_norm( + x_55, + (96,), + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) + x_57 = x_56.permute(0, 3, 1, 2) + x_56 = None + x_58 = torch.conv2d( + x_57, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_1 = sym_sum // 8 + sym_sum_2 = torch.sym_sum([1, floordiv_1]) + floordiv_1 = sym_sum_2 = None + x_59 = x_58.permute(0, 2, 3, 1) + x_60 = torch.nn.functional.layer_norm( + x_59, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_61 = x_60.permute(0, 3, 1, 2) + x_60 = None + x_62 = torch.conv2d( + x_61, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_6 = torch.nn.functional.relu(x_62, inplace=False) + x_62 = None + pow_7 = relu_6**2 + relu_6 = None + mul_6 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_7 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_7 + ) = None + x_63 = ( + mul_6 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_66 = x_58 + x_65 + x_58 = x_65 = None + x_67 = x_66.permute(0, 2, 3, 1) + x_68 = torch.nn.functional.layer_norm( + x_67, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + x_70 = torch.conv2d( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_7 = torch.nn.functional.relu(x_70, inplace=False) + x_70 = None + pow_8 = relu_7**2 + relu_7 = None + mul_7 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_8 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_8 + ) = None + x_71 = ( + mul_7 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + x_73 = torch.conv2d( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + x_75 = x_66 + x_74 + x_66 = x_74 = None + x_76 = x_75.permute(0, 2, 3, 1) + x_77 = torch.nn.functional.layer_norm( + x_76, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_8 = torch.nn.functional.relu(x_79, inplace=False) + x_79 = None + pow_9 = relu_8**2 + relu_8 = None + mul_8 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_9 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_9 + ) = None + x_80 = ( + mul_8 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_81 = torch.conv2d( + x_80, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_83 = x_75 + x_82 + x_75 = x_82 = None + x_84 = x_83.permute(0, 2, 3, 1) + x_85 = torch.nn.functional.layer_norm( + x_84, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_86 = x_85.permute(0, 3, 1, 2) + x_85 = None + x_87 = torch.conv2d( + x_86, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_9 = torch.nn.functional.relu(x_87, inplace=False) + x_87 = None + pow_10 = relu_9**2 + relu_9 = None + mul_9 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_10 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_10 + ) = None + x_88 = ( + mul_9 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = x_83 + x_91 + x_83 = x_91 = None + x_93 = x_92.permute(0, 2, 3, 1) + x_94 = torch.nn.functional.layer_norm( + x_93, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = torch.conv2d( + x_95, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_10 = torch.nn.functional.relu(x_96, inplace=False) + x_96 = None + pow_11 = relu_10**2 + relu_10 = None + mul_10 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_11 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_11 + ) = None + x_97 = ( + mul_10 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_98 = torch.conv2d( + x_97, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_99 = torch.conv2d( + x_98, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_100 = x_92 + x_99 + x_92 = x_99 = None + x_101 = x_100.permute(0, 2, 3, 1) + x_102 = torch.nn.functional.layer_norm( + x_101, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_103 = x_102.permute(0, 3, 1, 2) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_11 = torch.nn.functional.relu(x_104, inplace=False) + x_104 = None + pow_12 = relu_11**2 + relu_11 = None + mul_11 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_12 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_12 + ) = None + x_105 = ( + mul_11 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_100 + x_108 + x_100 = x_108 = None + x_110 = x_109.permute(0, 2, 3, 1) + x_111 = torch.nn.functional.layer_norm( + x_110, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_110 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) + x_112 = x_111.permute(0, 3, 1, 2) + x_111 = None + x_113 = torch.conv2d( + x_112, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_112 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_12 = torch.nn.functional.relu(x_113, inplace=False) + x_113 = None + pow_13 = relu_12**2 + relu_12 = None + mul_12 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ + * pow_13 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_13 + ) = None + x_114 = ( + mul_12 + + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_12 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_115 = torch.conv2d( + x_114, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_114 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_117 = x_109 + x_116 + x_109 = x_116 = None + x_118 = x_117.permute(0, 2, 3, 1) + x_119 = torch.nn.functional.layer_norm( + x_118, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_118 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) + x_120 = x_119.permute(0, 3, 1, 2) + x_119 = None + x_121 = torch.conv2d( + x_120, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_120 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_13 = torch.nn.functional.relu(x_121, inplace=False) + x_121 = None + pow_14 = relu_13**2 + relu_13 = None + mul_13 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + * pow_14 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( + pow_14 + ) = None + x_122 = ( + mul_13 + + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + ) + mul_13 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) + x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) + x_122 = None + x_124 = torch.conv2d( + x_123, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_123 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_126 = x_117 + x_125 + x_117 = x_125 = None + x_127 = x_126.permute(0, 2, 3, 1) + x_128 = torch.nn.functional.layer_norm( + x_127, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_127 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) + x_129 = x_128.permute(0, 3, 1, 2) + x_128 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_14 = torch.nn.functional.relu(x_130, inplace=False) + x_130 = None + pow_15 = relu_14**2 + relu_14 = None + mul_14 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ + * pow_15 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_15 + ) = None + x_131 = ( + mul_14 + + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_14 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_132 = torch.conv2d( + x_131, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_131 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_132 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_134 = x_126 + x_133 + x_126 = x_133 = None + x_135 = x_134.permute(0, 2, 3, 1) + x_136 = torch.nn.functional.layer_norm( + x_135, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_135 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) + x_137 = x_136.permute(0, 3, 1, 2) + x_136 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_15 = torch.nn.functional.relu(x_138, inplace=False) + x_138 = None + pow_16 = relu_15**2 + relu_15 = None + mul_15 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + * pow_16 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( + pow_16 + ) = None + x_139 = ( + mul_15 + + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + ) + mul_15 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) + x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) + x_139 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_140 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_134 + x_142 + x_134 = x_142 = None + x_144 = x_143.permute(0, 2, 3, 1) + x_145 = torch.nn.functional.layer_norm( + x_144, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_144 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) + x_146 = x_145.permute(0, 3, 1, 2) + x_145 = None + x_147 = torch.conv2d( + x_146, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_146 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_16 = torch.nn.functional.relu(x_147, inplace=False) + x_147 = None + pow_17 = relu_16**2 + relu_16 = None + mul_16 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ + * pow_17 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_17 + ) = None + x_148 = ( + mul_16 + + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_16 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_149 = torch.conv2d( + x_148, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_148 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_150 = torch.conv2d( + x_149, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_149 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_151 = x_143 + x_150 + x_143 = x_150 = None + x_152 = x_151.permute(0, 2, 3, 1) + x_153 = torch.nn.functional.layer_norm( + x_152, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_152 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) + x_154 = x_153.permute(0, 3, 1, 2) + x_153 = None + x_155 = torch.conv2d( + x_154, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_154 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_17 = torch.nn.functional.relu(x_155, inplace=False) + x_155 = None + pow_18 = relu_17**2 + relu_17 = None + mul_17 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + * pow_18 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( + pow_18 + ) = None + x_156 = ( + mul_17 + + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + ) + mul_17 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch.conv2d( + x_157, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_157 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_151 + x_159 + x_151 = x_159 = None + x_161 = x_160.permute(0, 2, 3, 1) + x_162 = torch.nn.functional.layer_norm( + x_161, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_161 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) + x_163 = x_162.permute(0, 3, 1, 2) + x_162 = None + x_164 = torch.conv2d( + x_163, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_163 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_18 = torch.nn.functional.relu(x_164, inplace=False) + x_164 = None + pow_19 = relu_18**2 + relu_18 = None + mul_18 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ + * pow_19 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_19 + ) = None + x_165 = ( + mul_18 + + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_18 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_166 = torch.conv2d( + x_165, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_165 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_167 = torch.conv2d( + x_166, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_166 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_168 = x_160 + x_167 + x_160 = x_167 = None + x_169 = x_168.permute(0, 2, 3, 1) + x_170 = torch.nn.functional.layer_norm( + x_169, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_169 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) + x_171 = x_170.permute(0, 3, 1, 2) + x_170 = None + x_172 = torch.conv2d( + x_171, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_171 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_19 = torch.nn.functional.relu(x_172, inplace=False) + x_172 = None + pow_20 = relu_19**2 + relu_19 = None + mul_19 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + * pow_20 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( + pow_20 + ) = None + x_173 = ( + mul_19 + + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + ) + mul_19 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) + x_174 = torch.nn.functional.dropout(x_173, 0.0, False, False) + x_173 = None + x_175 = torch.conv2d( + x_174, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_174 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = x_168 + x_176 + x_168 = x_176 = None + x_178 = x_177.permute(0, 2, 3, 1) + x_179 = torch.nn.functional.layer_norm( + x_178, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_178 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) + x_180 = x_179.permute(0, 3, 1, 2) + x_179 = None + x_181 = torch.conv2d( + x_180, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_180 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_20 = torch.nn.functional.relu(x_181, inplace=False) + x_181 = None + pow_21 = relu_20**2 + relu_20 = None + mul_20 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ + * pow_21 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_21 + ) = None + x_182 = ( + mul_20 + + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_20 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_183 = torch.conv2d( + x_182, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_182 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_184 = torch.conv2d( + x_183, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_183 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_185 = x_177 + x_184 + x_177 = x_184 = None + x_186 = x_185.permute(0, 2, 3, 1) + x_187 = torch.nn.functional.layer_norm( + x_186, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) + x_188 = x_187.permute(0, 3, 1, 2) + x_187 = None + x_189 = torch.conv2d( + x_188, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_188 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_21 = torch.nn.functional.relu(x_189, inplace=False) + x_189 = None + pow_22 = relu_21**2 + relu_21 = None + mul_21 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + * pow_22 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( + pow_22 + ) = None + x_190 = ( + mul_21 + + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + ) + mul_21 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) + x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) + x_190 = None + x_192 = torch.conv2d( + x_191, + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_191 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_185 + x_193 + x_185 = x_193 = None + x_195 = x_194.permute(0, 2, 3, 1) + x_196 = torch.nn.functional.layer_norm( + x_195, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_195 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) + x_197 = x_196.permute(0, 3, 1, 2) + x_196 = None + x_198 = torch.conv2d( + x_197, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_197 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_22 = torch.nn.functional.relu(x_198, inplace=False) + x_198 = None + pow_23 = relu_22**2 + relu_22 = None + mul_22 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ + * pow_23 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_23 + ) = None + x_199 = ( + mul_22 + + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_22 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_200 = torch.conv2d( + x_199, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_199 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_201 = torch.conv2d( + x_200, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_200 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_202 = x_194 + x_201 + x_194 = x_201 = None + x_203 = x_202.permute(0, 2, 3, 1) + x_204 = torch.nn.functional.layer_norm( + x_203, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_203 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) + x_205 = x_204.permute(0, 3, 1, 2) + x_204 = None + x_206 = torch.conv2d( + x_205, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_205 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_23 = torch.nn.functional.relu(x_206, inplace=False) + x_206 = None + pow_24 = relu_23**2 + relu_23 = None + mul_23 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + * pow_24 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( + pow_24 + ) = None + x_207 = ( + mul_23 + + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + ) + mul_23 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) + x_208 = torch.nn.functional.dropout(x_207, 0.0, False, False) + x_207 = None + x_209 = torch.conv2d( + x_208, + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_208 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) + x_209 = None + x_211 = x_202 + x_210 + x_202 = x_210 = None + x_212 = x_211.permute(0, 2, 3, 1) + x_213 = torch.nn.functional.layer_norm( + x_212, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_212 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) + x_214 = x_213.permute(0, 3, 1, 2) + x_213 = None + x_215 = torch.conv2d( + x_214, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_214 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_24 = torch.nn.functional.relu(x_215, inplace=False) + x_215 = None + pow_25 = relu_24**2 + relu_24 = None + mul_24 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ + * pow_25 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_25 + ) = None + x_216 = ( + mul_24 + + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_24 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_217 = torch.conv2d( + x_216, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_216 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_217 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_219 = x_211 + x_218 + x_211 = x_218 = None + x_220 = x_219.permute(0, 2, 3, 1) + x_221 = torch.nn.functional.layer_norm( + x_220, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_220 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) + x_222 = x_221.permute(0, 3, 1, 2) + x_221 = None + x_223 = torch.conv2d( + x_222, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_222 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_25 = torch.nn.functional.relu(x_223, inplace=False) + x_223 = None + pow_26 = relu_25**2 + relu_25 = None + mul_25 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + * pow_26 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( + pow_26 + ) = None + x_224 = ( + mul_25 + + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + ) + mul_25 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = torch.conv2d( + x_225, + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_225 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) + x_226 = None + x_228 = x_219 + x_227 + x_219 = x_227 = None + x_229 = x_228.permute(0, 2, 3, 1) + x_230 = torch.nn.functional.layer_norm( + x_229, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_229 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) + x_231 = x_230.permute(0, 3, 1, 2) + x_230 = None + x_232 = torch.conv2d( + x_231, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_231 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_26 = torch.nn.functional.relu(x_232, inplace=False) + x_232 = None + pow_27 = relu_26**2 + relu_26 = None + mul_26 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ + * pow_27 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_27 + ) = None + x_233 = ( + mul_26 + + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_26 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_234 = torch.conv2d( + x_233, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_233 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_235 = torch.conv2d( + x_234, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_234 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_236 = x_228 + x_235 + x_228 = x_235 = None + x_237 = x_236.permute(0, 2, 3, 1) + x_238 = torch.nn.functional.layer_norm( + x_237, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_237 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) + x_239 = x_238.permute(0, 3, 1, 2) + x_238 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_239 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_27 = torch.nn.functional.relu(x_240, inplace=False) + x_240 = None + pow_28 = relu_27**2 + relu_27 = None + mul_27 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + * pow_28 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( + pow_28 + ) = None + x_241 = ( + mul_27 + + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + ) + mul_27 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) + x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) + x_241 = None + x_243 = torch.conv2d( + x_242, + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_242 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_244 = torch.nn.functional.dropout(x_243, 0.0, False, False) + x_243 = None + x_245 = x_236 + x_244 + x_236 = x_244 = None + x_246 = x_245.permute(0, 2, 3, 1) + x_247 = torch.nn.functional.layer_norm( + x_246, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_246 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) + x_248 = x_247.permute(0, 3, 1, 2) + x_247 = None + x_249 = torch.conv2d( + x_248, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_248 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_28 = torch.nn.functional.relu(x_249, inplace=False) + x_249 = None + pow_29 = relu_28**2 + relu_28 = None + mul_28 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ + * pow_29 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_29 + ) = None + x_250 = ( + mul_28 + + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_28 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + x_250 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_252 = torch.conv2d( + x_251, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_251 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_253 = x_245 + x_252 + x_245 = x_252 = None + x_254 = x_253.permute(0, 2, 3, 1) + x_255 = torch.nn.functional.layer_norm( + x_254, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_254 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) + x_256 = x_255.permute(0, 3, 1, 2) + x_255 = None + x_257 = torch.conv2d( + x_256, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_256 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_29 = torch.nn.functional.relu(x_257, inplace=False) + x_257 = None + pow_30 = relu_29**2 + relu_29 = None + mul_29 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + * pow_30 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( + pow_30 + ) = None + x_258 = ( + mul_29 + + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + ) + mul_29 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = torch.conv2d( + x_259, + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_259 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_261 = torch.nn.functional.dropout(x_260, 0.0, False, False) + x_260 = None + x_262 = x_253 + x_261 + x_253 = x_261 = None + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (192,), + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) + x_265 = x_264.permute(0, 3, 1, 2) + x_264 = None + x_266 = torch.conv2d( + x_265, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_265 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_2 = sym_sum // 16 + sym_sum_3 = torch.sym_sum([1, floordiv_2]) + floordiv_2 = None + reshape = x_266.reshape(1, 384, -1) + x_266 = None + x_267 = reshape.transpose(1, 2) + reshape = None + view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_30 = x_267 * view + view = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_267, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_52 = sym_sum_3**2 + linear = torch._C._nn.linear( + layer_norm_33, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_33 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_1 = linear.reshape(1, pow_52, 3, 12, 32) + linear = None + qkv = reshape_1.permute(2, 0, 3, 1, 4) + reshape_1 = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_268 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) + q = k = v = None + transpose_1 = x_268.transpose(1, 2) + x_268 = None + x_269 = transpose_1.reshape(1, pow_52, 384) + transpose_1 = None + x_270 = torch._C._nn.linear( + x_269, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_269 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_271 = torch.nn.functional.dropout(x_270, 0.0, False, False) + x_270 = None + x_272 = mul_30 + x_271 + mul_30 = x_271 = None + view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_31 = x_272 * view_1 + view_1 = None + layer_norm_34 = torch.nn.functional.layer_norm( + x_272, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_272 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_273 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_34 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_30 = torch.nn.functional.relu(x_273, inplace=False) + x_273 = None + pow_31 = relu_30**2 + relu_30 = None + mul_32 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_31 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_31 + ) = None + x_274 = ( + mul_32 + + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_275 = torch.nn.functional.dropout(x_274, 0.0, False, False) + x_274 = None + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_277 = torch.nn.functional.dropout(x_276, 0.0, False, False) + x_276 = None + x_278 = mul_31 + x_277 + mul_31 = x_277 = None + view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_33 = x_278 * view_2 + view_2 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_278, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_3 = linear_4.reshape(1, pow_52, 3, 12, 32) + linear_4 = None + qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) + reshape_3 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_279 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) + q_1 = k_1 = v_1 = None + transpose_2 = x_279.transpose(1, 2) + x_279 = None + x_280 = transpose_2.reshape(1, pow_52, 384) + transpose_2 = None + x_281 = torch._C._nn.linear( + x_280, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_280 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) + x_281 = None + x_283 = mul_33 + x_282 + mul_33 = x_282 = None + view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_34 = x_283 * view_3 + view_3 = None + layer_norm_36 = torch.nn.functional.layer_norm( + x_283, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_283 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_284 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_36 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_31 = torch.nn.functional.relu(x_284, inplace=False) + x_284 = None + pow_32 = relu_31**2 + relu_31 = None + mul_35 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_32 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_32 + ) = None + x_285 = ( + mul_35 + + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_286 = torch.nn.functional.dropout(x_285, 0.0, False, False) + x_285 = None + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) + x_287 = None + x_289 = mul_34 + x_288 + mul_34 = x_288 = None + view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_36 = x_289 * view_4 + view_4 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_289, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_37 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_5 = linear_8.reshape(1, pow_52, 3, 12, 32) + linear_8 = None + qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) + reshape_5 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_290 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) + q_2 = k_2 = v_2 = None + transpose_3 = x_290.transpose(1, 2) + x_290 = None + x_291 = transpose_3.reshape(1, pow_52, 384) + transpose_3 = None + x_292 = torch._C._nn.linear( + x_291, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_291 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_293 = torch.nn.functional.dropout(x_292, 0.0, False, False) + x_292 = None + x_294 = mul_36 + x_293 + mul_36 = x_293 = None + view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_37 = x_294 * view_5 + view_5 = None + layer_norm_38 = torch.nn.functional.layer_norm( + x_294, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_294 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_295 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_32 = torch.nn.functional.relu(x_295, inplace=False) + x_295 = None + pow_33 = relu_32**2 + relu_32 = None + mul_38 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_33 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_33 + ) = None + x_296 = ( + mul_38 + + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_297 = torch.nn.functional.dropout(x_296, 0.0, False, False) + x_296 = None + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_299 = torch.nn.functional.dropout(x_298, 0.0, False, False) + x_298 = None + x_300 = mul_37 + x_299 + mul_37 = x_299 = None + view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_39 = x_300 * view_6 + view_6 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_300, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_39 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_7 = linear_12.reshape(1, pow_52, 3, 12, 32) + linear_12 = None + qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) + reshape_7 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_301 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) + q_3 = k_3 = v_3 = None + transpose_4 = x_301.transpose(1, 2) + x_301 = None + x_302 = transpose_4.reshape(1, pow_52, 384) + transpose_4 = None + x_303 = torch._C._nn.linear( + x_302, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_302 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_304 = torch.nn.functional.dropout(x_303, 0.0, False, False) + x_303 = None + x_305 = mul_39 + x_304 + mul_39 = x_304 = None + view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_40 = x_305 * view_7 + view_7 = None + layer_norm_40 = torch.nn.functional.layer_norm( + x_305, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_305 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) + x_306 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_40 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_33 = torch.nn.functional.relu(x_306, inplace=False) + x_306 = None + pow_34 = relu_33**2 + relu_33 = None + mul_41 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + * pow_34 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( + pow_34 + ) = None + x_307 = ( + mul_41 + + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + ) + mul_41 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) + x_308 = torch.nn.functional.dropout(x_307, 0.0, False, False) + x_307 = None + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) + x_309 = None + x_311 = mul_40 + x_310 + mul_40 = x_310 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_42 = x_311 * view_8 + view_8 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_311, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_41 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_9 = linear_16.reshape(1, pow_52, 3, 12, 32) + linear_16 = None + qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) + reshape_9 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_312 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) + q_4 = k_4 = v_4 = None + transpose_5 = x_312.transpose(1, 2) + x_312 = None + x_313 = transpose_5.reshape(1, pow_52, 384) + transpose_5 = None + x_314 = torch._C._nn.linear( + x_313, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_313 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) + x_314 = None + x_316 = mul_42 + x_315 + mul_42 = x_315 = None + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_43 = x_316 * view_9 + view_9 = None + layer_norm_42 = torch.nn.functional.layer_norm( + x_316, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_316 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) + x_317 = torch._C._nn.linear( + layer_norm_42, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_42 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_34 = torch.nn.functional.relu(x_317, inplace=False) + x_317 = None + pow_35 = relu_34**2 + relu_34 = None + mul_44 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + * pow_35 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( + pow_35 + ) = None + x_318 = ( + mul_44 + + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + ) + mul_44 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) + x_319 = torch.nn.functional.dropout(x_318, 0.0, False, False) + x_318 = None + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_321 = torch.nn.functional.dropout(x_320, 0.0, False, False) + x_320 = None + x_322 = mul_43 + x_321 + mul_43 = x_321 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_45 = x_322 * view_10 + view_10 = None + layer_norm_43 = torch.nn.functional.layer_norm( + x_322, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) + linear_20 = torch._C._nn.linear( + layer_norm_43, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_43 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_11 = linear_20.reshape(1, pow_52, 3, 12, 32) + linear_20 = None + qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) + reshape_11 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_323 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) + q_5 = k_5 = v_5 = None + transpose_6 = x_323.transpose(1, 2) + x_323 = None + x_324 = transpose_6.reshape(1, pow_52, 384) + transpose_6 = None + x_325 = torch._C._nn.linear( + x_324, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_324 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_326 = torch.nn.functional.dropout(x_325, 0.0, False, False) + x_325 = None + x_327 = mul_45 + x_326 + mul_45 = x_326 = None + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_46 = x_327 * view_11 + view_11 = None + layer_norm_44 = torch.nn.functional.layer_norm( + x_327, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_327 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) + x_328 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_44 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_35 = torch.nn.functional.relu(x_328, inplace=False) + x_328 = None + pow_36 = relu_35**2 + relu_35 = None + mul_47 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + * pow_36 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( + pow_36 + ) = None + x_329 = ( + mul_47 + + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + ) + mul_47 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) + x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) + x_329 = None + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_332 = torch.nn.functional.dropout(x_331, 0.0, False, False) + x_331 = None + x_333 = mul_46 + x_332 + mul_46 = x_332 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_48 = x_333 * view_12 + view_12 = None + layer_norm_45 = torch.nn.functional.layer_norm( + x_333, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) + linear_24 = torch._C._nn.linear( + layer_norm_45, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_45 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_13 = linear_24.reshape(1, pow_52, 3, 12, 32) + linear_24 = None + qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) + reshape_13 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_334 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) + q_6 = k_6 = v_6 = None + transpose_7 = x_334.transpose(1, 2) + x_334 = None + x_335 = transpose_7.reshape(1, pow_52, 384) + transpose_7 = None + x_336 = torch._C._nn.linear( + x_335, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_335 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_337 = torch.nn.functional.dropout(x_336, 0.0, False, False) + x_336 = None + x_338 = mul_48 + x_337 + mul_48 = x_337 = None + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_49 = x_338 * view_13 + view_13 = None + layer_norm_46 = torch.nn.functional.layer_norm( + x_338, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_338 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) + x_339 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_46 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_36 = torch.nn.functional.relu(x_339, inplace=False) + x_339 = None + pow_37 = relu_36**2 + relu_36 = None + mul_50 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + * pow_37 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( + pow_37 + ) = None + x_340 = ( + mul_50 + + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + ) + mul_50 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) + x_341 = torch.nn.functional.dropout(x_340, 0.0, False, False) + x_340 = None + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_343 = torch.nn.functional.dropout(x_342, 0.0, False, False) + x_342 = None + x_344 = mul_49 + x_343 + mul_49 = x_343 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_51 = x_344 * view_14 + view_14 = None + layer_norm_47 = torch.nn.functional.layer_norm( + x_344, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) + linear_28 = torch._C._nn.linear( + layer_norm_47, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_47 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_15 = linear_28.reshape(1, pow_52, 3, 12, 32) + linear_28 = None + qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) + reshape_15 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_345 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) + q_7 = k_7 = v_7 = None + transpose_8 = x_345.transpose(1, 2) + x_345 = None + x_346 = transpose_8.reshape(1, pow_52, 384) + transpose_8 = None + x_347 = torch._C._nn.linear( + x_346, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_346 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) + x_347 = None + x_349 = mul_51 + x_348 + mul_51 = x_348 = None + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_52 = x_349 * view_15 + view_15 = None + layer_norm_48 = torch.nn.functional.layer_norm( + x_349, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_349 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) + x_350 = torch._C._nn.linear( + layer_norm_48, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_48 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_37 = torch.nn.functional.relu(x_350, inplace=False) + x_350 = None + pow_38 = relu_37**2 + relu_37 = None + mul_53 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + * pow_38 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( + pow_38 + ) = None + x_351 = ( + mul_53 + + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + ) + mul_53 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) + x_352 = torch.nn.functional.dropout(x_351, 0.0, False, False) + x_351 = None + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_354 = torch.nn.functional.dropout(x_353, 0.0, False, False) + x_353 = None + x_355 = mul_52 + x_354 + mul_52 = x_354 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_54 = x_355 * view_16 + view_16 = None + layer_norm_49 = torch.nn.functional.layer_norm( + x_355, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) + linear_32 = torch._C._nn.linear( + layer_norm_49, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_49 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_17 = linear_32.reshape(1, pow_52, 3, 12, 32) + linear_32 = None + qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) + reshape_17 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_356 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) + q_8 = k_8 = v_8 = None + transpose_9 = x_356.transpose(1, 2) + x_356 = None + x_357 = transpose_9.reshape(1, pow_52, 384) + transpose_9 = None + x_358 = torch._C._nn.linear( + x_357, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_357 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_359 = torch.nn.functional.dropout(x_358, 0.0, False, False) + x_358 = None + x_360 = mul_54 + x_359 + mul_54 = x_359 = None + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_55 = x_360 * view_17 + view_17 = None + layer_norm_50 = torch.nn.functional.layer_norm( + x_360, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_360 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) + x_361 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_50 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_38 = torch.nn.functional.relu(x_361, inplace=False) + x_361 = None + pow_39 = relu_38**2 + relu_38 = None + mul_56 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + * pow_39 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( + pow_39 + ) = None + x_362 = ( + mul_56 + + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + ) + mul_56 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) + x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) + x_362 = None + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_365 = torch.nn.functional.dropout(x_364, 0.0, False, False) + x_364 = None + x_366 = mul_55 + x_365 + mul_55 = x_365 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_57 = x_366 * view_18 + view_18 = None + layer_norm_51 = torch.nn.functional.layer_norm( + x_366, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_51, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_51 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_19 = linear_36.reshape(1, pow_52, 3, 12, 32) + linear_36 = None + qkv_9 = reshape_19.permute(2, 0, 3, 1, 4) + reshape_19 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_367 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) + q_9 = k_9 = v_9 = None + transpose_10 = x_367.transpose(1, 2) + x_367 = None + x_368 = transpose_10.reshape(1, pow_52, 384) + transpose_10 = None + x_369 = torch._C._nn.linear( + x_368, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_368 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_370 = torch.nn.functional.dropout(x_369, 0.0, False, False) + x_369 = None + x_371 = mul_57 + x_370 + mul_57 = x_370 = None + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_58 = x_371 * view_19 + view_19 = None + layer_norm_52 = torch.nn.functional.layer_norm( + x_371, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_371 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) + x_372 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_52 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_39 = torch.nn.functional.relu(x_372, inplace=False) + x_372 = None + pow_40 = relu_39**2 + relu_39 = None + mul_59 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ + * pow_40 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( + pow_40 + ) = None + x_373 = ( + mul_59 + + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ + ) + mul_59 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) + x_374 = torch.nn.functional.dropout(x_373, 0.0, False, False) + x_373 = None + x_375 = torch._C._nn.linear( + x_374, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_374 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_376 = torch.nn.functional.dropout(x_375, 0.0, False, False) + x_375 = None + x_377 = mul_58 + x_376 + mul_58 = x_376 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_60 = x_377 * view_20 + view_20 = None + layer_norm_53 = torch.nn.functional.layer_norm( + x_377, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_377 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_53, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_53 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_21 = linear_40.reshape(1, pow_52, 3, 12, 32) + linear_40 = None + qkv_10 = reshape_21.permute(2, 0, 3, 1, 4) + reshape_21 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_378 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_378.transpose(1, 2) + x_378 = None + x_379 = transpose_11.reshape(1, pow_52, 384) + transpose_11 = None + x_380 = torch._C._nn.linear( + x_379, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_379 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_381 = torch.nn.functional.dropout(x_380, 0.0, False, False) + x_380 = None + x_382 = mul_60 + x_381 + mul_60 = x_381 = None + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_61 = x_382 * view_21 + view_21 = None + layer_norm_54 = torch.nn.functional.layer_norm( + x_382, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_382 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) + x_383 = torch._C._nn.linear( + layer_norm_54, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_54 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_40 = torch.nn.functional.relu(x_383, inplace=False) + x_383 = None + pow_41 = relu_40**2 + relu_40 = None + mul_62 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ + * pow_41 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( + pow_41 + ) = None + x_384 = ( + mul_62 + + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ + ) + mul_62 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) + x_385 = torch.nn.functional.dropout(x_384, 0.0, False, False) + x_384 = None + x_386 = torch._C._nn.linear( + x_385, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_385 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_387 = torch.nn.functional.dropout(x_386, 0.0, False, False) + x_386 = None + x_388 = mul_61 + x_387 + mul_61 = x_387 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_63 = x_388 * view_22 + view_22 = None + layer_norm_55 = torch.nn.functional.layer_norm( + x_388, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_388 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_55, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_55 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_23 = linear_44.reshape(1, pow_52, 3, 12, 32) + linear_44 = None + qkv_11 = reshape_23.permute(2, 0, 3, 1, 4) + reshape_23 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_389 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_389.transpose(1, 2) + x_389 = None + x_390 = transpose_12.reshape(1, pow_52, 384) + transpose_12 = None + x_391 = torch._C._nn.linear( + x_390, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_390 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_392 = torch.nn.functional.dropout(x_391, 0.0, False, False) + x_391 = None + x_393 = mul_63 + x_392 + mul_63 = x_392 = None + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_64 = x_393 * view_23 + view_23 = None + layer_norm_56 = torch.nn.functional.layer_norm( + x_393, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_393 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) + x_394 = torch._C._nn.linear( + layer_norm_56, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_56 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_41 = torch.nn.functional.relu(x_394, inplace=False) + x_394 = None + pow_42 = relu_41**2 + relu_41 = None + mul_65 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ + * pow_42 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( + pow_42 + ) = None + x_395 = ( + mul_65 + + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ + ) + mul_65 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) + x_396 = torch.nn.functional.dropout(x_395, 0.0, False, False) + x_395 = None + x_397 = torch._C._nn.linear( + x_396, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_396 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_398 = torch.nn.functional.dropout(x_397, 0.0, False, False) + x_397 = None + x_399 = mul_64 + x_398 + mul_64 = x_398 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_66 = x_399 * view_24 + view_24 = None + layer_norm_57 = torch.nn.functional.layer_norm( + x_399, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_399 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_57, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_57 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_25 = linear_48.reshape(1, pow_52, 3, 12, 32) + linear_48 = None + qkv_12 = reshape_25.permute(2, 0, 3, 1, 4) + reshape_25 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_12 = unbind_12[0] + k_12 = unbind_12[1] + v_12 = unbind_12[2] + unbind_12 = None + x_400 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_13 = x_400.transpose(1, 2) + x_400 = None + x_401 = transpose_13.reshape(1, pow_52, 384) + transpose_13 = None + x_402 = torch._C._nn.linear( + x_401, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_401 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_403 = torch.nn.functional.dropout(x_402, 0.0, False, False) + x_402 = None + x_404 = mul_66 + x_403 + mul_66 = x_403 = None + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_67 = x_404 * view_25 + view_25 = None + layer_norm_58 = torch.nn.functional.layer_norm( + x_404, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_404 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = (None) + x_405 = torch._C._nn.linear( + layer_norm_58, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_58 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_42 = torch.nn.functional.relu(x_405, inplace=False) + x_405 = None + pow_43 = relu_42**2 + relu_42 = None + mul_68 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ + * pow_43 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = ( + pow_43 + ) = None + x_406 = ( + mul_68 + + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ + ) + mul_68 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = (None) + x_407 = torch.nn.functional.dropout(x_406, 0.0, False, False) + x_406 = None + x_408 = torch._C._nn.linear( + x_407, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_407 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_409 = torch.nn.functional.dropout(x_408, 0.0, False, False) + x_408 = None + x_410 = mul_67 + x_409 + mul_67 = x_409 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_69 = x_410 * view_26 + view_26 = None + layer_norm_59 = torch.nn.functional.layer_norm( + x_410, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_410 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = (None) + linear_52 = torch._C._nn.linear( + layer_norm_59, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_59 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_27 = linear_52.reshape(1, pow_52, 3, 12, 32) + linear_52 = None + qkv_13 = reshape_27.permute(2, 0, 3, 1, 4) + reshape_27 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_13 = unbind_13[0] + k_13 = unbind_13[1] + v_13 = unbind_13[2] + unbind_13 = None + x_411 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_14 = x_411.transpose(1, 2) + x_411 = None + x_412 = transpose_14.reshape(1, pow_52, 384) + transpose_14 = None + x_413 = torch._C._nn.linear( + x_412, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_412 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_414 = torch.nn.functional.dropout(x_413, 0.0, False, False) + x_413 = None + x_415 = mul_69 + x_414 + mul_69 = x_414 = None + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_70 = x_415 * view_27 + view_27 = None + layer_norm_60 = torch.nn.functional.layer_norm( + x_415, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_415 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = (None) + x_416 = torch._C._nn.linear( + layer_norm_60, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_60 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_43 = torch.nn.functional.relu(x_416, inplace=False) + x_416 = None + pow_44 = relu_43**2 + relu_43 = None + mul_71 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ + * pow_44 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = ( + pow_44 + ) = None + x_417 = ( + mul_71 + + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ + ) + mul_71 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = (None) + x_418 = torch.nn.functional.dropout(x_417, 0.0, False, False) + x_417 = None + x_419 = torch._C._nn.linear( + x_418, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_418 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_420 = torch.nn.functional.dropout(x_419, 0.0, False, False) + x_419 = None + x_421 = mul_70 + x_420 + mul_70 = x_420 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_72 = x_421 * view_28 + view_28 = None + layer_norm_61 = torch.nn.functional.layer_norm( + x_421, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_421 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_61, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_61 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_29 = linear_56.reshape(1, pow_52, 3, 12, 32) + linear_56 = None + qkv_14 = reshape_29.permute(2, 0, 3, 1, 4) + reshape_29 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_14 = unbind_14[0] + k_14 = unbind_14[1] + v_14 = unbind_14[2] + unbind_14 = None + x_422 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_15 = x_422.transpose(1, 2) + x_422 = None + x_423 = transpose_15.reshape(1, pow_52, 384) + transpose_15 = None + x_424 = torch._C._nn.linear( + x_423, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_423 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_425 = torch.nn.functional.dropout(x_424, 0.0, False, False) + x_424 = None + x_426 = mul_72 + x_425 + mul_72 = x_425 = None + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_73 = x_426 * view_29 + view_29 = None + layer_norm_62 = torch.nn.functional.layer_norm( + x_426, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_426 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = (None) + x_427 = torch._C._nn.linear( + layer_norm_62, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_62 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_44 = torch.nn.functional.relu(x_427, inplace=False) + x_427 = None + pow_45 = relu_44**2 + relu_44 = None + mul_74 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ + * pow_45 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = ( + pow_45 + ) = None + x_428 = ( + mul_74 + + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ + ) + mul_74 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = (None) + x_429 = torch.nn.functional.dropout(x_428, 0.0, False, False) + x_428 = None + x_430 = torch._C._nn.linear( + x_429, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_429 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_431 = torch.nn.functional.dropout(x_430, 0.0, False, False) + x_430 = None + x_432 = mul_73 + x_431 + mul_73 = x_431 = None + view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_75 = x_432 * view_30 + view_30 = None + layer_norm_63 = torch.nn.functional.layer_norm( + x_432, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_432 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = (None) + linear_60 = torch._C._nn.linear( + layer_norm_63, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_63 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_31 = linear_60.reshape(1, pow_52, 3, 12, 32) + linear_60 = None + qkv_15 = reshape_31.permute(2, 0, 3, 1, 4) + reshape_31 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_15 = unbind_15[0] + k_15 = unbind_15[1] + v_15 = unbind_15[2] + unbind_15 = None + x_433 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_16 = x_433.transpose(1, 2) + x_433 = None + x_434 = transpose_16.reshape(1, pow_52, 384) + transpose_16 = None + x_435 = torch._C._nn.linear( + x_434, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_434 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_436 = torch.nn.functional.dropout(x_435, 0.0, False, False) + x_435 = None + x_437 = mul_75 + x_436 + mul_75 = x_436 = None + view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_76 = x_437 * view_31 + view_31 = None + layer_norm_64 = torch.nn.functional.layer_norm( + x_437, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_437 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = (None) + x_438 = torch._C._nn.linear( + layer_norm_64, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_64 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_45 = torch.nn.functional.relu(x_438, inplace=False) + x_438 = None + pow_46 = relu_45**2 + relu_45 = None + mul_77 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ + * pow_46 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = ( + pow_46 + ) = None + x_439 = ( + mul_77 + + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ + ) + mul_77 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = (None) + x_440 = torch.nn.functional.dropout(x_439, 0.0, False, False) + x_439 = None + x_441 = torch._C._nn.linear( + x_440, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_440 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_442 = torch.nn.functional.dropout(x_441, 0.0, False, False) + x_441 = None + x_443 = mul_76 + x_442 + mul_76 = x_442 = None + view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_78 = x_443 * view_32 + view_32 = None + layer_norm_65 = torch.nn.functional.layer_norm( + x_443, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_443 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_65, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_65 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_33 = linear_64.reshape(1, pow_52, 3, 12, 32) + linear_64 = None + qkv_16 = reshape_33.permute(2, 0, 3, 1, 4) + reshape_33 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_16 = unbind_16[0] + k_16 = unbind_16[1] + v_16 = unbind_16[2] + unbind_16 = None + x_444 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_17 = x_444.transpose(1, 2) + x_444 = None + x_445 = transpose_17.reshape(1, pow_52, 384) + transpose_17 = None + x_446 = torch._C._nn.linear( + x_445, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_445 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_447 = torch.nn.functional.dropout(x_446, 0.0, False, False) + x_446 = None + x_448 = mul_78 + x_447 + mul_78 = x_447 = None + view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_79 = x_448 * view_33 + view_33 = None + layer_norm_66 = torch.nn.functional.layer_norm( + x_448, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_448 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = (None) + x_449 = torch._C._nn.linear( + layer_norm_66, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_66 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_46 = torch.nn.functional.relu(x_449, inplace=False) + x_449 = None + pow_47 = relu_46**2 + relu_46 = None + mul_80 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ + * pow_47 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = ( + pow_47 + ) = None + x_450 = ( + mul_80 + + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ + ) + mul_80 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = (None) + x_451 = torch.nn.functional.dropout(x_450, 0.0, False, False) + x_450 = None + x_452 = torch._C._nn.linear( + x_451, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_451 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_453 = torch.nn.functional.dropout(x_452, 0.0, False, False) + x_452 = None + x_454 = mul_79 + x_453 + mul_79 = x_453 = None + view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_81 = x_454 * view_34 + view_34 = None + layer_norm_67 = torch.nn.functional.layer_norm( + x_454, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_454 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = (None) + linear_68 = torch._C._nn.linear( + layer_norm_67, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_67 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_35 = linear_68.reshape(1, pow_52, 3, 12, 32) + linear_68 = None + qkv_17 = reshape_35.permute(2, 0, 3, 1, 4) + reshape_35 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_17 = unbind_17[0] + k_17 = unbind_17[1] + v_17 = unbind_17[2] + unbind_17 = None + x_455 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_18 = x_455.transpose(1, 2) + x_455 = None + x_456 = transpose_18.reshape(1, pow_52, 384) + transpose_18 = pow_52 = None + x_457 = torch._C._nn.linear( + x_456, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_456 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_458 = torch.nn.functional.dropout(x_457, 0.0, False, False) + x_457 = None + x_459 = mul_81 + x_458 + mul_81 = x_458 = None + view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_.view( + (384,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_82 = x_459 * view_35 + view_35 = None + layer_norm_68 = torch.nn.functional.layer_norm( + x_459, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_459 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = (None) + x_460 = torch._C._nn.linear( + layer_norm_68, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_68 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_47 = torch.nn.functional.relu(x_460, inplace=False) + x_460 = None + pow_48 = relu_47**2 + relu_47 = None + mul_83 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ + * pow_48 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = ( + pow_48 + ) = None + x_461 = ( + mul_83 + + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ + ) + mul_83 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = (None) + x_462 = torch.nn.functional.dropout(x_461, 0.0, False, False) + x_461 = None + x_463 = torch._C._nn.linear( + x_462, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_462 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_464 = torch.nn.functional.dropout(x_463, 0.0, False, False) + x_463 = None + x_465 = mul_82 + x_464 + mul_82 = x_464 = None + transpose_19 = x_465.transpose(1, 2) + x_465 = None + x_466 = transpose_19.reshape(1, 384, sym_sum_3, sym_sum_3) + transpose_19 = sym_sum_3 = None + x_467 = x_466.permute(0, 2, 3, 1) + x_466 = None + x_468 = torch.nn.functional.layer_norm( + x_467, + (384,), + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_467 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) + x_469 = x_468.permute(0, 3, 1, 2) + x_468 = None + x_470 = torch.conv2d( + x_469, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_469 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_3 = sym_sum // 32 + sym_sum = None + sym_sum_4 = torch.sym_sum([1, floordiv_3]) + floordiv_3 = None + reshape_38 = x_470.reshape(1, 576, -1) + x_470 = None + x_471 = reshape_38.transpose(1, 2) + reshape_38 = None + view_36 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_84 = x_471 * view_36 + view_36 = None + layer_norm_70 = torch.nn.functional.layer_norm( + x_471, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_471 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_53 = sym_sum_4**2 + linear_72 = torch._C._nn.linear( + layer_norm_70, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_70 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_39 = linear_72.reshape(1, pow_53, 3, 18, 32) + linear_72 = None + qkv_18 = reshape_39.permute(2, 0, 3, 1, 4) + reshape_39 = None + unbind_18 = qkv_18.unbind(0) + qkv_18 = None + q_18 = unbind_18[0] + k_18 = unbind_18[1] + v_18 = unbind_18[2] + unbind_18 = None + x_472 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_21 = x_472.transpose(1, 2) + x_472 = None + x_473 = transpose_21.reshape(1, pow_53, 576) + transpose_21 = None + x_474 = torch._C._nn.linear( + x_473, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_473 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_475 = torch.nn.functional.dropout(x_474, 0.0, False, False) + x_474 = None + x_476 = mul_84 + x_475 + mul_84 = x_475 = None + view_37 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_85 = x_476 * view_37 + view_37 = None + layer_norm_71 = torch.nn.functional.layer_norm( + x_476, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_476 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_477 = torch._C._nn.linear( + layer_norm_71, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_71 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_48 = torch.nn.functional.relu(x_477, inplace=False) + x_477 = None + pow_49 = relu_48**2 + relu_48 = None + mul_86 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_49 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_49 + ) = None + x_478 = ( + mul_86 + + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_86 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_479 = torch.nn.functional.dropout(x_478, 0.0, False, False) + x_478 = None + x_480 = torch._C._nn.linear( + x_479, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_479 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_481 = torch.nn.functional.dropout(x_480, 0.0, False, False) + x_480 = None + x_482 = mul_85 + x_481 + mul_85 = x_481 = None + view_38 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_87 = x_482 * view_38 + view_38 = None + layer_norm_72 = torch.nn.functional.layer_norm( + x_482, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_482 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_76 = torch._C._nn.linear( + layer_norm_72, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_72 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_41 = linear_76.reshape(1, pow_53, 3, 18, 32) + linear_76 = None + qkv_19 = reshape_41.permute(2, 0, 3, 1, 4) + reshape_41 = None + unbind_19 = qkv_19.unbind(0) + qkv_19 = None + q_19 = unbind_19[0] + k_19 = unbind_19[1] + v_19 = unbind_19[2] + unbind_19 = None + x_483 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_22 = x_483.transpose(1, 2) + x_483 = None + x_484 = transpose_22.reshape(1, pow_53, 576) + transpose_22 = None + x_485 = torch._C._nn.linear( + x_484, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_484 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_486 = torch.nn.functional.dropout(x_485, 0.0, False, False) + x_485 = None + x_487 = mul_87 + x_486 + mul_87 = x_486 = None + view_39 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_88 = x_487 * view_39 + view_39 = None + layer_norm_73 = torch.nn.functional.layer_norm( + x_487, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_487 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_488 = torch._C._nn.linear( + layer_norm_73, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_73 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_49 = torch.nn.functional.relu(x_488, inplace=False) + x_488 = None + pow_50 = relu_49**2 + relu_49 = None + mul_89 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_50 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_50 + ) = None + x_489 = ( + mul_89 + + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_89 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_490 = torch.nn.functional.dropout(x_489, 0.0, False, False) + x_489 = None + x_491 = torch._C._nn.linear( + x_490, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_490 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_492 = torch.nn.functional.dropout(x_491, 0.0, False, False) + x_491 = None + x_493 = mul_88 + x_492 + mul_88 = x_492 = None + view_40 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_90 = x_493 * view_40 + view_40 = None + layer_norm_74 = torch.nn.functional.layer_norm( + x_493, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_493 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_74, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_74 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_43 = linear_80.reshape(1, pow_53, 3, 18, 32) + linear_80 = None + qkv_20 = reshape_43.permute(2, 0, 3, 1, 4) + reshape_43 = None + unbind_20 = qkv_20.unbind(0) + qkv_20 = None + q_20 = unbind_20[0] + k_20 = unbind_20[1] + v_20 = unbind_20[2] + unbind_20 = None + x_494 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_23 = x_494.transpose(1, 2) + x_494 = None + x_495 = transpose_23.reshape(1, pow_53, 576) + transpose_23 = pow_53 = None + x_496 = torch._C._nn.linear( + x_495, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_495 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_497 = torch.nn.functional.dropout(x_496, 0.0, False, False) + x_496 = None + x_498 = mul_90 + x_497 + mul_90 = x_497 = None + view_41 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (576,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_91 = x_498 * view_41 + view_41 = None + layer_norm_75 = torch.nn.functional.layer_norm( + x_498, + (576,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_498 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_499 = torch._C._nn.linear( + layer_norm_75, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_75 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_50 = torch.nn.functional.relu(x_499, inplace=False) + x_499 = None + pow_51 = relu_50**2 + relu_50 = None + mul_92 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_51 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_51 + ) = None + x_500 = ( + mul_92 + + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_92 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_501 = torch.nn.functional.dropout(x_500, 0.0, False, False) + x_500 = None + x_502 = torch._C._nn.linear( + x_501, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_501 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_503 = torch.nn.functional.dropout(x_502, 0.0, False, False) + x_502 = None + x_504 = mul_91 + x_503 + mul_91 = x_503 = None + transpose_24 = x_504.transpose(1, 2) + x_504 = None + x_505 = transpose_24.reshape(1, 576, sym_sum_4, sym_sum_4) + transpose_24 = sym_sum_4 = None + x_506 = torch.nn.functional.adaptive_avg_pool2d(x_505, 1) + x_505 = None + x_507 = x_506.permute(0, 2, 3, 1) + x_506 = None + x_508 = torch.nn.functional.layer_norm( + x_507, + (576,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_507 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_509 = x_508.permute(0, 3, 1, 2) + x_508 = None + x_510 = x_509.flatten(1, -1) + x_509 = None + x_511 = torch.nn.functional.dropout(x_510, 0.0, False, False) + x_510 = None + x_512 = torch._C._nn.linear( + x_511, + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, + ) + x_511 = ( + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None + relu_51 = torch.nn.functional.relu(x_512, inplace=False) + x_512 = None + x_513 = torch.square(relu_51) + relu_51 = None + x_514 = torch.nn.functional.layer_norm( + x_513, + (2304,), + l_self_modules_head_modules_fc_modules_norm_parameters_weight_, + l_self_modules_head_modules_fc_modules_norm_parameters_bias_, + 1e-06, + ) + x_513 = ( + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None + x_515 = torch.nn.functional.dropout(x_514, 0.0, False, False) + x_514 = None + x_516 = torch._C._nn.linear( + x_515, + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, + ) + x_515 = ( + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None + return (x_516,) diff --git a/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/weight_meta.py b/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/weight_meta.py new file mode 100644 index 0000000000..8e5c0d0fc0 --- /dev/null +++ b/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/weight_meta.py @@ -0,0 +1,3968 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv_parameters_weight_" + shape = [96, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: + name = "L_self_modules_stem_modules_conv_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.230 + std = 1.288 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: + name = "L_self_modules_stem_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [192, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [96, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [192, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [96, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [192, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [96, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" + shape = [192, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" + shape = [384, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" + shape = [576, 384, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1728, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [576, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2304, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [576, 2304] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1728, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [576, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2304, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [576, 2304] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1728, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [576, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2304, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [576, 2304] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" + shape = [2304, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" + shape = [1000, 2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/caformer_s18.sail_in1k/graph_net.json b/samples/timm/caformer_s18.sail_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/caformer_s18.sail_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/caformer_s18.sail_in1k/input_meta.py b/samples/timm/caformer_s18.sail_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_s18.sail_in1k/input_tensor_constraints.py b/samples/timm/caformer_s18.sail_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_s18.sail_in1k/model.py b/samples/timm/caformer_s18.sail_in1k/model.py new file mode 100644 index 0000000000..ba7a8ecdba --- /dev/null +++ b/samples/timm/caformer_s18.sail_in1k/model.py @@ -0,0 +1,2468 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv_parameters_bias_ = ( + L_self_modules_stem_modules_conv_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_norm_parameters_weight_ = ( + L_self_modules_stem_modules_norm_parameters_weight_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv_parameters_weight_, + l_self_modules_stem_modules_conv_parameters_bias_, + (4, 4), + (2, 2), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_conv_parameters_weight_ + ) = l_self_modules_stem_modules_conv_parameters_bias_ = None + x_1 = x.permute(0, 2, 3, 1) + x = None + x_2 = torch.nn.functional.layer_norm( + x_1, (64,), l_self_modules_stem_modules_norm_parameters_weight_, None, 1e-06 + ) + x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None + x_3 = x_2.permute(0, 3, 1, 2) + x_2 = None + sym_sum = torch.sym_sum([-3, s1]) + s1 = None + floordiv = sym_sum // 4 + sym_sum_1 = torch.sym_sum([1, floordiv]) + floordiv = sym_sum_1 = None + x_4 = x_3.permute(0, 2, 3, 1) + x_5 = torch.nn.functional.layer_norm( + x_4, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu = torch.nn.functional.relu(x_7, inplace=False) + x_7 = None + pow_1 = relu**2 + relu = None + mul = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_1 + ) = None + x_8 = ( + mul + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_11 = x_3 + x_10 + x_3 = x_10 = None + x_12 = x_11.permute(0, 2, 3, 1) + x_13 = torch.nn.functional.layer_norm( + x_12, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_14 = x_13.permute(0, 3, 1, 2) + x_13 = None + x_15 = torch.conv2d( + x_14, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_1 = torch.nn.functional.relu(x_15, inplace=False) + x_15 = None + pow_2 = relu_1**2 + relu_1 = None + mul_1 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_2 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_2 + ) = None + x_16 = ( + mul_1 + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = x_11 + x_19 + x_11 = x_19 = None + x_21 = x_20.permute(0, 2, 3, 1) + x_22 = torch.nn.functional.layer_norm( + x_21, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = torch.conv2d( + x_23, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_2 = torch.nn.functional.relu(x_24, inplace=False) + x_24 = None + pow_3 = relu_2**2 + relu_2 = None + mul_2 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_3 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_3 + ) = None + x_25 = ( + mul_2 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_27 = torch.conv2d( + x_26, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_28 = x_20 + x_27 + x_20 = x_27 = None + x_29 = x_28.permute(0, 2, 3, 1) + x_30 = torch.nn.functional.layer_norm( + x_29, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_3 = torch.nn.functional.relu(x_32, inplace=False) + x_32 = None + pow_4 = relu_3**2 + relu_3 = None + mul_3 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_4 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_4 + ) = None + x_33 = ( + mul_3 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + x_37 = x_28 + x_36 + x_28 = x_36 = None + x_38 = x_37.permute(0, 2, 3, 1) + x_39 = torch.nn.functional.layer_norm( + x_38, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_40 = x_39.permute(0, 3, 1, 2) + x_39 = None + x_41 = torch.conv2d( + x_40, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_4 = torch.nn.functional.relu(x_41, inplace=False) + x_41 = None + pow_5 = relu_4**2 + relu_4 = None + mul_4 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_5 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_5 + ) = None + x_42 = ( + mul_4 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_44 = torch.conv2d( + x_43, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_45 = x_37 + x_44 + x_37 = x_44 = None + x_46 = x_45.permute(0, 2, 3, 1) + x_47 = torch.nn.functional.layer_norm( + x_46, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_5 = torch.nn.functional.relu(x_49, inplace=False) + x_49 = None + pow_6 = relu_5**2 + relu_5 = None + mul_5 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_6 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_6 + ) = None + x_50 = ( + mul_5 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = x_45 + x_53 + x_45 = x_53 = None + x_55 = x_54.permute(0, 2, 3, 1) + x_54 = None + x_56 = torch.nn.functional.layer_norm( + x_55, + (64,), + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) + x_57 = x_56.permute(0, 3, 1, 2) + x_56 = None + x_58 = torch.conv2d( + x_57, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_1 = sym_sum // 8 + sym_sum_2 = torch.sym_sum([1, floordiv_1]) + floordiv_1 = sym_sum_2 = None + x_59 = x_58.permute(0, 2, 3, 1) + x_60 = torch.nn.functional.layer_norm( + x_59, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_61 = x_60.permute(0, 3, 1, 2) + x_60 = None + x_62 = torch.conv2d( + x_61, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_6 = torch.nn.functional.relu(x_62, inplace=False) + x_62 = None + pow_7 = relu_6**2 + relu_6 = None + mul_6 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_7 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_7 + ) = None + x_63 = ( + mul_6 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_66 = x_58 + x_65 + x_58 = x_65 = None + x_67 = x_66.permute(0, 2, 3, 1) + x_68 = torch.nn.functional.layer_norm( + x_67, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + x_70 = torch.conv2d( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_7 = torch.nn.functional.relu(x_70, inplace=False) + x_70 = None + pow_8 = relu_7**2 + relu_7 = None + mul_7 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_8 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_8 + ) = None + x_71 = ( + mul_7 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + x_73 = torch.conv2d( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + x_75 = x_66 + x_74 + x_66 = x_74 = None + x_76 = x_75.permute(0, 2, 3, 1) + x_77 = torch.nn.functional.layer_norm( + x_76, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_8 = torch.nn.functional.relu(x_79, inplace=False) + x_79 = None + pow_9 = relu_8**2 + relu_8 = None + mul_8 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_9 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_9 + ) = None + x_80 = ( + mul_8 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_81 = torch.conv2d( + x_80, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_83 = x_75 + x_82 + x_75 = x_82 = None + x_84 = x_83.permute(0, 2, 3, 1) + x_85 = torch.nn.functional.layer_norm( + x_84, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_86 = x_85.permute(0, 3, 1, 2) + x_85 = None + x_87 = torch.conv2d( + x_86, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_9 = torch.nn.functional.relu(x_87, inplace=False) + x_87 = None + pow_10 = relu_9**2 + relu_9 = None + mul_9 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_10 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_10 + ) = None + x_88 = ( + mul_9 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = x_83 + x_91 + x_83 = x_91 = None + x_93 = x_92.permute(0, 2, 3, 1) + x_94 = torch.nn.functional.layer_norm( + x_93, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = torch.conv2d( + x_95, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_10 = torch.nn.functional.relu(x_96, inplace=False) + x_96 = None + pow_11 = relu_10**2 + relu_10 = None + mul_10 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_11 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_11 + ) = None + x_97 = ( + mul_10 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_98 = torch.conv2d( + x_97, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_99 = torch.conv2d( + x_98, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_100 = x_92 + x_99 + x_92 = x_99 = None + x_101 = x_100.permute(0, 2, 3, 1) + x_102 = torch.nn.functional.layer_norm( + x_101, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_103 = x_102.permute(0, 3, 1, 2) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_11 = torch.nn.functional.relu(x_104, inplace=False) + x_104 = None + pow_12 = relu_11**2 + relu_11 = None + mul_11 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_12 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_12 + ) = None + x_105 = ( + mul_11 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_100 + x_108 + x_100 = x_108 = None + x_110 = x_109.permute(0, 2, 3, 1) + x_109 = None + x_111 = torch.nn.functional.layer_norm( + x_110, + (128,), + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_110 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) + x_112 = x_111.permute(0, 3, 1, 2) + x_111 = None + x_113 = torch.conv2d( + x_112, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_112 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_2 = sym_sum // 16 + sym_sum_3 = torch.sym_sum([1, floordiv_2]) + floordiv_2 = None + reshape = x_113.reshape(1, 320, -1) + x_113 = None + x_114 = reshape.transpose(1, 2) + reshape = None + view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_12 = x_114 * view + view = None + layer_norm_15 = torch.nn.functional.layer_norm( + x_114, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_25 = sym_sum_3**2 + linear = torch._C._nn.linear( + layer_norm_15, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_15 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_1 = linear.reshape(1, pow_25, 3, 10, 32) + linear = None + qkv = reshape_1.permute(2, 0, 3, 1, 4) + reshape_1 = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_115 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) + q = k = v = None + transpose_1 = x_115.transpose(1, 2) + x_115 = None + x_116 = transpose_1.reshape(1, pow_25, 320) + transpose_1 = None + x_117 = torch._C._nn.linear( + x_116, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_116 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = mul_12 + x_118 + mul_12 = x_118 = None + view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_13 = x_119 * view_1 + view_1 = None + layer_norm_16 = torch.nn.functional.layer_norm( + x_119, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_119 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_120 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_16 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_12 = torch.nn.functional.relu(x_120, inplace=False) + x_120 = None + pow_13 = relu_12**2 + relu_12 = None + mul_14 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_13 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_13 + ) = None + x_121 = ( + mul_14 + + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_14 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_122 = torch.nn.functional.dropout(x_121, 0.0, False, False) + x_121 = None + x_123 = torch._C._nn.linear( + x_122, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = mul_13 + x_124 + mul_13 = x_124 = None + view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_15 = x_125 * view_2 + view_2 = None + layer_norm_17 = torch.nn.functional.layer_norm( + x_125, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_17, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_17 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_3 = linear_4.reshape(1, pow_25, 3, 10, 32) + linear_4 = None + qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) + reshape_3 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_126 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) + q_1 = k_1 = v_1 = None + transpose_2 = x_126.transpose(1, 2) + x_126 = None + x_127 = transpose_2.reshape(1, pow_25, 320) + transpose_2 = None + x_128 = torch._C._nn.linear( + x_127, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_127 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_129 = torch.nn.functional.dropout(x_128, 0.0, False, False) + x_128 = None + x_130 = mul_15 + x_129 + mul_15 = x_129 = None + view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_16 = x_130 * view_3 + view_3 = None + layer_norm_18 = torch.nn.functional.layer_norm( + x_130, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_130 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_131 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_18 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_13 = torch.nn.functional.relu(x_131, inplace=False) + x_131 = None + pow_14 = relu_13**2 + relu_13 = None + mul_17 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_14 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_14 + ) = None + x_132 = ( + mul_17 + + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_17 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + x_134 = torch._C._nn.linear( + x_133, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_133 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = mul_16 + x_135 + mul_16 = x_135 = None + view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_18 = x_136 * view_4 + view_4 = None + layer_norm_19 = torch.nn.functional.layer_norm( + x_136, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_19, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_19 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_5 = linear_8.reshape(1, pow_25, 3, 10, 32) + linear_8 = None + qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) + reshape_5 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_137 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) + q_2 = k_2 = v_2 = None + transpose_3 = x_137.transpose(1, 2) + x_137 = None + x_138 = transpose_3.reshape(1, pow_25, 320) + transpose_3 = None + x_139 = torch._C._nn.linear( + x_138, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_138 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) + x_139 = None + x_141 = mul_18 + x_140 + mul_18 = x_140 = None + view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_19 = x_141 * view_5 + view_5 = None + layer_norm_20 = torch.nn.functional.layer_norm( + x_141, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_141 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_142 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_20 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_14 = torch.nn.functional.relu(x_142, inplace=False) + x_142 = None + pow_15 = relu_14**2 + relu_14 = None + mul_20 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_15 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_15 + ) = None + x_143 = ( + mul_20 + + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_20 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) + x_143 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_144 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = mul_19 + x_146 + mul_19 = x_146 = None + view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_21 = x_147 * view_6 + view_6 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_147, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_147 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_21 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_7 = linear_12.reshape(1, pow_25, 3, 10, 32) + linear_12 = None + qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) + reshape_7 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_148 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) + q_3 = k_3 = v_3 = None + transpose_4 = x_148.transpose(1, 2) + x_148 = None + x_149 = transpose_4.reshape(1, pow_25, 320) + transpose_4 = None + x_150 = torch._C._nn.linear( + x_149, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_149 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_151 = torch.nn.functional.dropout(x_150, 0.0, False, False) + x_150 = None + x_152 = mul_21 + x_151 + mul_21 = x_151 = None + view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_22 = x_152 * view_7 + view_7 = None + layer_norm_22 = torch.nn.functional.layer_norm( + x_152, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_152 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) + x_153 = torch._C._nn.linear( + layer_norm_22, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_22 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_15 = torch.nn.functional.relu(x_153, inplace=False) + x_153 = None + pow_16 = relu_15**2 + relu_15 = None + mul_23 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + * pow_16 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( + pow_16 + ) = None + x_154 = ( + mul_23 + + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + ) + mul_23 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) + x_155 = torch.nn.functional.dropout(x_154, 0.0, False, False) + x_154 = None + x_156 = torch._C._nn.linear( + x_155, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_155 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = mul_22 + x_157 + mul_22 = x_157 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_24 = x_158 * view_8 + view_8 = None + layer_norm_23 = torch.nn.functional.layer_norm( + x_158, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_23, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_23 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_9 = linear_16.reshape(1, pow_25, 3, 10, 32) + linear_16 = None + qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) + reshape_9 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_159 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) + q_4 = k_4 = v_4 = None + transpose_5 = x_159.transpose(1, 2) + x_159 = None + x_160 = transpose_5.reshape(1, pow_25, 320) + transpose_5 = None + x_161 = torch._C._nn.linear( + x_160, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_160 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_162 = torch.nn.functional.dropout(x_161, 0.0, False, False) + x_161 = None + x_163 = mul_24 + x_162 + mul_24 = x_162 = None + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_25 = x_163 * view_9 + view_9 = None + layer_norm_24 = torch.nn.functional.layer_norm( + x_163, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_163 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) + x_164 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_24 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_16 = torch.nn.functional.relu(x_164, inplace=False) + x_164 = None + pow_17 = relu_16**2 + relu_16 = None + mul_26 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + * pow_17 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( + pow_17 + ) = None + x_165 = ( + mul_26 + + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + ) + mul_26 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) + x_166 = torch.nn.functional.dropout(x_165, 0.0, False, False) + x_165 = None + x_167 = torch._C._nn.linear( + x_166, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_166 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_169 = mul_25 + x_168 + mul_25 = x_168 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_27 = x_169 * view_10 + view_10 = None + layer_norm_25 = torch.nn.functional.layer_norm( + x_169, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) + linear_20 = torch._C._nn.linear( + layer_norm_25, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_25 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_11 = linear_20.reshape(1, pow_25, 3, 10, 32) + linear_20 = None + qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) + reshape_11 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_170 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) + q_5 = k_5 = v_5 = None + transpose_6 = x_170.transpose(1, 2) + x_170 = None + x_171 = transpose_6.reshape(1, pow_25, 320) + transpose_6 = None + x_172 = torch._C._nn.linear( + x_171, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_171 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_173 = torch.nn.functional.dropout(x_172, 0.0, False, False) + x_172 = None + x_174 = mul_27 + x_173 + mul_27 = x_173 = None + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_28 = x_174 * view_11 + view_11 = None + layer_norm_26 = torch.nn.functional.layer_norm( + x_174, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_174 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) + x_175 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_26 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_17 = torch.nn.functional.relu(x_175, inplace=False) + x_175 = None + pow_18 = relu_17**2 + relu_17 = None + mul_29 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + * pow_18 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( + pow_18 + ) = None + x_176 = ( + mul_29 + + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + ) + mul_29 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) + x_177 = torch.nn.functional.dropout(x_176, 0.0, False, False) + x_176 = None + x_178 = torch._C._nn.linear( + x_177, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_177 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_180 = mul_28 + x_179 + mul_28 = x_179 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_30 = x_180 * view_12 + view_12 = None + layer_norm_27 = torch.nn.functional.layer_norm( + x_180, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_180 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) + linear_24 = torch._C._nn.linear( + layer_norm_27, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_27 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_13 = linear_24.reshape(1, pow_25, 3, 10, 32) + linear_24 = None + qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) + reshape_13 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_181 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) + q_6 = k_6 = v_6 = None + transpose_7 = x_181.transpose(1, 2) + x_181 = None + x_182 = transpose_7.reshape(1, pow_25, 320) + transpose_7 = None + x_183 = torch._C._nn.linear( + x_182, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_182 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_184 = torch.nn.functional.dropout(x_183, 0.0, False, False) + x_183 = None + x_185 = mul_30 + x_184 + mul_30 = x_184 = None + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_31 = x_185 * view_13 + view_13 = None + layer_norm_28 = torch.nn.functional.layer_norm( + x_185, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_185 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) + x_186 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_28 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_18 = torch.nn.functional.relu(x_186, inplace=False) + x_186 = None + pow_19 = relu_18**2 + relu_18 = None + mul_32 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + * pow_19 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( + pow_19 + ) = None + x_187 = ( + mul_32 + + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + ) + mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) + x_188 = torch.nn.functional.dropout(x_187, 0.0, False, False) + x_187 = None + x_189 = torch._C._nn.linear( + x_188, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_188 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = mul_31 + x_190 + mul_31 = x_190 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_33 = x_191 * view_14 + view_14 = None + layer_norm_29 = torch.nn.functional.layer_norm( + x_191, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_191 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) + linear_28 = torch._C._nn.linear( + layer_norm_29, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_29 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_15 = linear_28.reshape(1, pow_25, 3, 10, 32) + linear_28 = None + qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) + reshape_15 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_192 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) + q_7 = k_7 = v_7 = None + transpose_8 = x_192.transpose(1, 2) + x_192 = None + x_193 = transpose_8.reshape(1, pow_25, 320) + transpose_8 = None + x_194 = torch._C._nn.linear( + x_193, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_193 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_195 = torch.nn.functional.dropout(x_194, 0.0, False, False) + x_194 = None + x_196 = mul_33 + x_195 + mul_33 = x_195 = None + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_34 = x_196 * view_15 + view_15 = None + layer_norm_30 = torch.nn.functional.layer_norm( + x_196, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_196 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) + x_197 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_30 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_19 = torch.nn.functional.relu(x_197, inplace=False) + x_197 = None + pow_20 = relu_19**2 + relu_19 = None + mul_35 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + * pow_20 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( + pow_20 + ) = None + x_198 = ( + mul_35 + + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + ) + mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) + x_199 = torch.nn.functional.dropout(x_198, 0.0, False, False) + x_198 = None + x_200 = torch._C._nn.linear( + x_199, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_199 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_202 = mul_34 + x_201 + mul_34 = x_201 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_36 = x_202 * view_16 + view_16 = None + layer_norm_31 = torch.nn.functional.layer_norm( + x_202, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_202 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) + linear_32 = torch._C._nn.linear( + layer_norm_31, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_31 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_17 = linear_32.reshape(1, pow_25, 3, 10, 32) + linear_32 = None + qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) + reshape_17 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_203 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) + q_8 = k_8 = v_8 = None + transpose_9 = x_203.transpose(1, 2) + x_203 = None + x_204 = transpose_9.reshape(1, pow_25, 320) + transpose_9 = pow_25 = None + x_205 = torch._C._nn.linear( + x_204, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_204 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_206 = torch.nn.functional.dropout(x_205, 0.0, False, False) + x_205 = None + x_207 = mul_36 + x_206 + mul_36 = x_206 = None + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_37 = x_207 * view_17 + view_17 = None + layer_norm_32 = torch.nn.functional.layer_norm( + x_207, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_207 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) + x_208 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_32 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_20 = torch.nn.functional.relu(x_208, inplace=False) + x_208 = None + pow_21 = relu_20**2 + relu_20 = None + mul_38 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + * pow_21 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( + pow_21 + ) = None + x_209 = ( + mul_38 + + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + ) + mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) + x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) + x_209 = None + x_211 = torch._C._nn.linear( + x_210, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_210 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_213 = mul_37 + x_212 + mul_37 = x_212 = None + transpose_10 = x_213.transpose(1, 2) + x_213 = None + x_214 = transpose_10.reshape(1, 320, sym_sum_3, sym_sum_3) + transpose_10 = sym_sum_3 = None + x_215 = x_214.permute(0, 2, 3, 1) + x_214 = None + x_216 = torch.nn.functional.layer_norm( + x_215, + (320,), + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_215 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) + x_217 = x_216.permute(0, 3, 1, 2) + x_216 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_217 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_3 = sym_sum // 32 + sym_sum = None + sym_sum_4 = torch.sym_sum([1, floordiv_3]) + floordiv_3 = None + reshape_20 = x_218.reshape(1, 512, -1) + x_218 = None + x_219 = reshape_20.transpose(1, 2) + reshape_20 = None + view_18 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_39 = x_219 * view_18 + view_18 = None + layer_norm_34 = torch.nn.functional.layer_norm( + x_219, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_219 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_26 = sym_sum_4**2 + linear_36 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_34 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_21 = linear_36.reshape(1, pow_26, 3, 16, 32) + linear_36 = None + qkv_9 = reshape_21.permute(2, 0, 3, 1, 4) + reshape_21 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_220 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) + q_9 = k_9 = v_9 = None + transpose_12 = x_220.transpose(1, 2) + x_220 = None + x_221 = transpose_12.reshape(1, pow_26, 512) + transpose_12 = None + x_222 = torch._C._nn.linear( + x_221, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_221 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_224 = mul_39 + x_223 + mul_39 = x_223 = None + view_19 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_40 = x_224 * view_19 + view_19 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_224, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_224 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_225 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_35 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_21 = torch.nn.functional.relu(x_225, inplace=False) + x_225 = None + pow_22 = relu_21**2 + relu_21 = None + mul_41 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_22 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_22 + ) = None + x_226 = ( + mul_41 + + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_41 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) + x_226 = None + x_228 = torch._C._nn.linear( + x_227, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_227 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_229 = torch.nn.functional.dropout(x_228, 0.0, False, False) + x_228 = None + x_230 = mul_40 + x_229 + mul_40 = x_229 = None + view_20 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_42 = x_230 * view_20 + view_20 = None + layer_norm_36 = torch.nn.functional.layer_norm( + x_230, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_230 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_36 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_23 = linear_40.reshape(1, pow_26, 3, 16, 32) + linear_40 = None + qkv_10 = reshape_23.permute(2, 0, 3, 1, 4) + reshape_23 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_231 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_13 = x_231.transpose(1, 2) + x_231 = None + x_232 = transpose_13.reshape(1, pow_26, 512) + transpose_13 = None + x_233 = torch._C._nn.linear( + x_232, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_232 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_235 = mul_42 + x_234 + mul_42 = x_234 = None + view_21 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_43 = x_235 * view_21 + view_21 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_235, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_235 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_236 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_37 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_22 = torch.nn.functional.relu(x_236, inplace=False) + x_236 = None + pow_23 = relu_22**2 + relu_22 = None + mul_44 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_23 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_23 + ) = None + x_237 = ( + mul_44 + + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_44 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_238 = torch.nn.functional.dropout(x_237, 0.0, False, False) + x_237 = None + x_239 = torch._C._nn.linear( + x_238, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_238 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_240 = torch.nn.functional.dropout(x_239, 0.0, False, False) + x_239 = None + x_241 = mul_43 + x_240 + mul_43 = x_240 = None + view_22 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_45 = x_241 * view_22 + view_22 = None + layer_norm_38 = torch.nn.functional.layer_norm( + x_241, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_241 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_38 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_25 = linear_44.reshape(1, pow_26, 3, 16, 32) + linear_44 = None + qkv_11 = reshape_25.permute(2, 0, 3, 1, 4) + reshape_25 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_242 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_14 = x_242.transpose(1, 2) + x_242 = None + x_243 = transpose_14.reshape(1, pow_26, 512) + transpose_14 = pow_26 = None + x_244 = torch._C._nn.linear( + x_243, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_243 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_246 = mul_45 + x_245 + mul_45 = x_245 = None + view_23 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_46 = x_246 * view_23 + view_23 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_246, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_246 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_247 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_39 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_23 = torch.nn.functional.relu(x_247, inplace=False) + x_247 = None + pow_24 = relu_23**2 + relu_23 = None + mul_47 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_24 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_24 + ) = None + x_248 = ( + mul_47 + + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_47 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_249 = torch.nn.functional.dropout(x_248, 0.0, False, False) + x_248 = None + x_250 = torch._C._nn.linear( + x_249, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_249 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_251 = torch.nn.functional.dropout(x_250, 0.0, False, False) + x_250 = None + x_252 = mul_46 + x_251 + mul_46 = x_251 = None + transpose_15 = x_252.transpose(1, 2) + x_252 = None + x_253 = transpose_15.reshape(1, 512, sym_sum_4, sym_sum_4) + transpose_15 = sym_sum_4 = None + x_254 = torch.nn.functional.adaptive_avg_pool2d(x_253, 1) + x_253 = None + x_255 = x_254.permute(0, 2, 3, 1) + x_254 = None + x_256 = torch.nn.functional.layer_norm( + x_255, + (512,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_255 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_257 = x_256.permute(0, 3, 1, 2) + x_256 = None + x_258 = x_257.flatten(1, -1) + x_257 = None + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = torch._C._nn.linear( + x_259, + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, + ) + x_259 = ( + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None + relu_24 = torch.nn.functional.relu(x_260, inplace=False) + x_260 = None + x_261 = torch.square(relu_24) + relu_24 = None + x_262 = torch.nn.functional.layer_norm( + x_261, + (2048,), + l_self_modules_head_modules_fc_modules_norm_parameters_weight_, + l_self_modules_head_modules_fc_modules_norm_parameters_bias_, + 1e-06, + ) + x_261 = ( + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None + x_263 = torch.nn.functional.dropout(x_262, 0.0, False, False) + x_262 = None + x_264 = torch._C._nn.linear( + x_263, + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, + ) + x_263 = ( + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None + return (x_264,) diff --git a/samples/timm/caformer_s18.sail_in1k/weight_meta.py b/samples/timm/caformer_s18.sail_in1k/weight_meta.py new file mode 100644 index 0000000000..9c99be36a5 --- /dev/null +++ b/samples/timm/caformer_s18.sail_in1k/weight_meta.py @@ -0,0 +1,2078 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv_parameters_weight_" + shape = [64, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: + name = "L_self_modules_stem_modules_conv_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.230 + std = 1.288 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: + name = "L_self_modules_stem_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" + shape = [128, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" + shape = [320, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" + shape = [512, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" + shape = [1000, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/caformer_s18.sail_in1k_384/graph_net.json b/samples/timm/caformer_s18.sail_in1k_384/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/caformer_s18.sail_in1k_384/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/caformer_s18.sail_in1k_384/input_meta.py b/samples/timm/caformer_s18.sail_in1k_384/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_s18.sail_in1k_384/input_tensor_constraints.py b/samples/timm/caformer_s18.sail_in1k_384/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_s18.sail_in1k_384/model.py b/samples/timm/caformer_s18.sail_in1k_384/model.py new file mode 100644 index 0000000000..ba7a8ecdba --- /dev/null +++ b/samples/timm/caformer_s18.sail_in1k_384/model.py @@ -0,0 +1,2468 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv_parameters_bias_ = ( + L_self_modules_stem_modules_conv_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_norm_parameters_weight_ = ( + L_self_modules_stem_modules_norm_parameters_weight_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv_parameters_weight_, + l_self_modules_stem_modules_conv_parameters_bias_, + (4, 4), + (2, 2), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_conv_parameters_weight_ + ) = l_self_modules_stem_modules_conv_parameters_bias_ = None + x_1 = x.permute(0, 2, 3, 1) + x = None + x_2 = torch.nn.functional.layer_norm( + x_1, (64,), l_self_modules_stem_modules_norm_parameters_weight_, None, 1e-06 + ) + x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None + x_3 = x_2.permute(0, 3, 1, 2) + x_2 = None + sym_sum = torch.sym_sum([-3, s1]) + s1 = None + floordiv = sym_sum // 4 + sym_sum_1 = torch.sym_sum([1, floordiv]) + floordiv = sym_sum_1 = None + x_4 = x_3.permute(0, 2, 3, 1) + x_5 = torch.nn.functional.layer_norm( + x_4, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu = torch.nn.functional.relu(x_7, inplace=False) + x_7 = None + pow_1 = relu**2 + relu = None + mul = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_1 + ) = None + x_8 = ( + mul + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_11 = x_3 + x_10 + x_3 = x_10 = None + x_12 = x_11.permute(0, 2, 3, 1) + x_13 = torch.nn.functional.layer_norm( + x_12, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_14 = x_13.permute(0, 3, 1, 2) + x_13 = None + x_15 = torch.conv2d( + x_14, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_1 = torch.nn.functional.relu(x_15, inplace=False) + x_15 = None + pow_2 = relu_1**2 + relu_1 = None + mul_1 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_2 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_2 + ) = None + x_16 = ( + mul_1 + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = x_11 + x_19 + x_11 = x_19 = None + x_21 = x_20.permute(0, 2, 3, 1) + x_22 = torch.nn.functional.layer_norm( + x_21, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = torch.conv2d( + x_23, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_2 = torch.nn.functional.relu(x_24, inplace=False) + x_24 = None + pow_3 = relu_2**2 + relu_2 = None + mul_2 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_3 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_3 + ) = None + x_25 = ( + mul_2 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_27 = torch.conv2d( + x_26, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_28 = x_20 + x_27 + x_20 = x_27 = None + x_29 = x_28.permute(0, 2, 3, 1) + x_30 = torch.nn.functional.layer_norm( + x_29, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_3 = torch.nn.functional.relu(x_32, inplace=False) + x_32 = None + pow_4 = relu_3**2 + relu_3 = None + mul_3 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_4 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_4 + ) = None + x_33 = ( + mul_3 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + x_37 = x_28 + x_36 + x_28 = x_36 = None + x_38 = x_37.permute(0, 2, 3, 1) + x_39 = torch.nn.functional.layer_norm( + x_38, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_40 = x_39.permute(0, 3, 1, 2) + x_39 = None + x_41 = torch.conv2d( + x_40, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_4 = torch.nn.functional.relu(x_41, inplace=False) + x_41 = None + pow_5 = relu_4**2 + relu_4 = None + mul_4 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_5 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_5 + ) = None + x_42 = ( + mul_4 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_44 = torch.conv2d( + x_43, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_45 = x_37 + x_44 + x_37 = x_44 = None + x_46 = x_45.permute(0, 2, 3, 1) + x_47 = torch.nn.functional.layer_norm( + x_46, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_5 = torch.nn.functional.relu(x_49, inplace=False) + x_49 = None + pow_6 = relu_5**2 + relu_5 = None + mul_5 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_6 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_6 + ) = None + x_50 = ( + mul_5 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = x_45 + x_53 + x_45 = x_53 = None + x_55 = x_54.permute(0, 2, 3, 1) + x_54 = None + x_56 = torch.nn.functional.layer_norm( + x_55, + (64,), + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) + x_57 = x_56.permute(0, 3, 1, 2) + x_56 = None + x_58 = torch.conv2d( + x_57, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_1 = sym_sum // 8 + sym_sum_2 = torch.sym_sum([1, floordiv_1]) + floordiv_1 = sym_sum_2 = None + x_59 = x_58.permute(0, 2, 3, 1) + x_60 = torch.nn.functional.layer_norm( + x_59, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_61 = x_60.permute(0, 3, 1, 2) + x_60 = None + x_62 = torch.conv2d( + x_61, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_6 = torch.nn.functional.relu(x_62, inplace=False) + x_62 = None + pow_7 = relu_6**2 + relu_6 = None + mul_6 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_7 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_7 + ) = None + x_63 = ( + mul_6 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_66 = x_58 + x_65 + x_58 = x_65 = None + x_67 = x_66.permute(0, 2, 3, 1) + x_68 = torch.nn.functional.layer_norm( + x_67, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + x_70 = torch.conv2d( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_7 = torch.nn.functional.relu(x_70, inplace=False) + x_70 = None + pow_8 = relu_7**2 + relu_7 = None + mul_7 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_8 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_8 + ) = None + x_71 = ( + mul_7 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + x_73 = torch.conv2d( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + x_75 = x_66 + x_74 + x_66 = x_74 = None + x_76 = x_75.permute(0, 2, 3, 1) + x_77 = torch.nn.functional.layer_norm( + x_76, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_8 = torch.nn.functional.relu(x_79, inplace=False) + x_79 = None + pow_9 = relu_8**2 + relu_8 = None + mul_8 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_9 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_9 + ) = None + x_80 = ( + mul_8 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_81 = torch.conv2d( + x_80, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_83 = x_75 + x_82 + x_75 = x_82 = None + x_84 = x_83.permute(0, 2, 3, 1) + x_85 = torch.nn.functional.layer_norm( + x_84, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_86 = x_85.permute(0, 3, 1, 2) + x_85 = None + x_87 = torch.conv2d( + x_86, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_9 = torch.nn.functional.relu(x_87, inplace=False) + x_87 = None + pow_10 = relu_9**2 + relu_9 = None + mul_9 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_10 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_10 + ) = None + x_88 = ( + mul_9 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = x_83 + x_91 + x_83 = x_91 = None + x_93 = x_92.permute(0, 2, 3, 1) + x_94 = torch.nn.functional.layer_norm( + x_93, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = torch.conv2d( + x_95, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_10 = torch.nn.functional.relu(x_96, inplace=False) + x_96 = None + pow_11 = relu_10**2 + relu_10 = None + mul_10 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_11 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_11 + ) = None + x_97 = ( + mul_10 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_98 = torch.conv2d( + x_97, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_99 = torch.conv2d( + x_98, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_100 = x_92 + x_99 + x_92 = x_99 = None + x_101 = x_100.permute(0, 2, 3, 1) + x_102 = torch.nn.functional.layer_norm( + x_101, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_103 = x_102.permute(0, 3, 1, 2) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_11 = torch.nn.functional.relu(x_104, inplace=False) + x_104 = None + pow_12 = relu_11**2 + relu_11 = None + mul_11 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_12 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_12 + ) = None + x_105 = ( + mul_11 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_100 + x_108 + x_100 = x_108 = None + x_110 = x_109.permute(0, 2, 3, 1) + x_109 = None + x_111 = torch.nn.functional.layer_norm( + x_110, + (128,), + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_110 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) + x_112 = x_111.permute(0, 3, 1, 2) + x_111 = None + x_113 = torch.conv2d( + x_112, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_112 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_2 = sym_sum // 16 + sym_sum_3 = torch.sym_sum([1, floordiv_2]) + floordiv_2 = None + reshape = x_113.reshape(1, 320, -1) + x_113 = None + x_114 = reshape.transpose(1, 2) + reshape = None + view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_12 = x_114 * view + view = None + layer_norm_15 = torch.nn.functional.layer_norm( + x_114, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_25 = sym_sum_3**2 + linear = torch._C._nn.linear( + layer_norm_15, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_15 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_1 = linear.reshape(1, pow_25, 3, 10, 32) + linear = None + qkv = reshape_1.permute(2, 0, 3, 1, 4) + reshape_1 = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_115 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) + q = k = v = None + transpose_1 = x_115.transpose(1, 2) + x_115 = None + x_116 = transpose_1.reshape(1, pow_25, 320) + transpose_1 = None + x_117 = torch._C._nn.linear( + x_116, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_116 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = mul_12 + x_118 + mul_12 = x_118 = None + view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_13 = x_119 * view_1 + view_1 = None + layer_norm_16 = torch.nn.functional.layer_norm( + x_119, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_119 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_120 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_16 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_12 = torch.nn.functional.relu(x_120, inplace=False) + x_120 = None + pow_13 = relu_12**2 + relu_12 = None + mul_14 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_13 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_13 + ) = None + x_121 = ( + mul_14 + + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_14 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_122 = torch.nn.functional.dropout(x_121, 0.0, False, False) + x_121 = None + x_123 = torch._C._nn.linear( + x_122, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = mul_13 + x_124 + mul_13 = x_124 = None + view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_15 = x_125 * view_2 + view_2 = None + layer_norm_17 = torch.nn.functional.layer_norm( + x_125, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_17, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_17 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_3 = linear_4.reshape(1, pow_25, 3, 10, 32) + linear_4 = None + qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) + reshape_3 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_126 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) + q_1 = k_1 = v_1 = None + transpose_2 = x_126.transpose(1, 2) + x_126 = None + x_127 = transpose_2.reshape(1, pow_25, 320) + transpose_2 = None + x_128 = torch._C._nn.linear( + x_127, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_127 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_129 = torch.nn.functional.dropout(x_128, 0.0, False, False) + x_128 = None + x_130 = mul_15 + x_129 + mul_15 = x_129 = None + view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_16 = x_130 * view_3 + view_3 = None + layer_norm_18 = torch.nn.functional.layer_norm( + x_130, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_130 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_131 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_18 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_13 = torch.nn.functional.relu(x_131, inplace=False) + x_131 = None + pow_14 = relu_13**2 + relu_13 = None + mul_17 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_14 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_14 + ) = None + x_132 = ( + mul_17 + + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_17 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + x_134 = torch._C._nn.linear( + x_133, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_133 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = mul_16 + x_135 + mul_16 = x_135 = None + view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_18 = x_136 * view_4 + view_4 = None + layer_norm_19 = torch.nn.functional.layer_norm( + x_136, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_19, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_19 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_5 = linear_8.reshape(1, pow_25, 3, 10, 32) + linear_8 = None + qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) + reshape_5 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_137 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) + q_2 = k_2 = v_2 = None + transpose_3 = x_137.transpose(1, 2) + x_137 = None + x_138 = transpose_3.reshape(1, pow_25, 320) + transpose_3 = None + x_139 = torch._C._nn.linear( + x_138, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_138 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) + x_139 = None + x_141 = mul_18 + x_140 + mul_18 = x_140 = None + view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_19 = x_141 * view_5 + view_5 = None + layer_norm_20 = torch.nn.functional.layer_norm( + x_141, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_141 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_142 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_20 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_14 = torch.nn.functional.relu(x_142, inplace=False) + x_142 = None + pow_15 = relu_14**2 + relu_14 = None + mul_20 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_15 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_15 + ) = None + x_143 = ( + mul_20 + + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_20 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) + x_143 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_144 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = mul_19 + x_146 + mul_19 = x_146 = None + view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_21 = x_147 * view_6 + view_6 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_147, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_147 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_21 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_7 = linear_12.reshape(1, pow_25, 3, 10, 32) + linear_12 = None + qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) + reshape_7 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_148 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) + q_3 = k_3 = v_3 = None + transpose_4 = x_148.transpose(1, 2) + x_148 = None + x_149 = transpose_4.reshape(1, pow_25, 320) + transpose_4 = None + x_150 = torch._C._nn.linear( + x_149, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_149 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_151 = torch.nn.functional.dropout(x_150, 0.0, False, False) + x_150 = None + x_152 = mul_21 + x_151 + mul_21 = x_151 = None + view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_22 = x_152 * view_7 + view_7 = None + layer_norm_22 = torch.nn.functional.layer_norm( + x_152, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_152 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) + x_153 = torch._C._nn.linear( + layer_norm_22, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_22 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_15 = torch.nn.functional.relu(x_153, inplace=False) + x_153 = None + pow_16 = relu_15**2 + relu_15 = None + mul_23 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + * pow_16 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( + pow_16 + ) = None + x_154 = ( + mul_23 + + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + ) + mul_23 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) + x_155 = torch.nn.functional.dropout(x_154, 0.0, False, False) + x_154 = None + x_156 = torch._C._nn.linear( + x_155, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_155 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = mul_22 + x_157 + mul_22 = x_157 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_24 = x_158 * view_8 + view_8 = None + layer_norm_23 = torch.nn.functional.layer_norm( + x_158, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_23, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_23 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_9 = linear_16.reshape(1, pow_25, 3, 10, 32) + linear_16 = None + qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) + reshape_9 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_159 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) + q_4 = k_4 = v_4 = None + transpose_5 = x_159.transpose(1, 2) + x_159 = None + x_160 = transpose_5.reshape(1, pow_25, 320) + transpose_5 = None + x_161 = torch._C._nn.linear( + x_160, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_160 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_162 = torch.nn.functional.dropout(x_161, 0.0, False, False) + x_161 = None + x_163 = mul_24 + x_162 + mul_24 = x_162 = None + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_25 = x_163 * view_9 + view_9 = None + layer_norm_24 = torch.nn.functional.layer_norm( + x_163, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_163 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) + x_164 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_24 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_16 = torch.nn.functional.relu(x_164, inplace=False) + x_164 = None + pow_17 = relu_16**2 + relu_16 = None + mul_26 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + * pow_17 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( + pow_17 + ) = None + x_165 = ( + mul_26 + + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + ) + mul_26 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) + x_166 = torch.nn.functional.dropout(x_165, 0.0, False, False) + x_165 = None + x_167 = torch._C._nn.linear( + x_166, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_166 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_169 = mul_25 + x_168 + mul_25 = x_168 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_27 = x_169 * view_10 + view_10 = None + layer_norm_25 = torch.nn.functional.layer_norm( + x_169, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) + linear_20 = torch._C._nn.linear( + layer_norm_25, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_25 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_11 = linear_20.reshape(1, pow_25, 3, 10, 32) + linear_20 = None + qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) + reshape_11 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_170 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) + q_5 = k_5 = v_5 = None + transpose_6 = x_170.transpose(1, 2) + x_170 = None + x_171 = transpose_6.reshape(1, pow_25, 320) + transpose_6 = None + x_172 = torch._C._nn.linear( + x_171, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_171 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_173 = torch.nn.functional.dropout(x_172, 0.0, False, False) + x_172 = None + x_174 = mul_27 + x_173 + mul_27 = x_173 = None + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_28 = x_174 * view_11 + view_11 = None + layer_norm_26 = torch.nn.functional.layer_norm( + x_174, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_174 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) + x_175 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_26 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_17 = torch.nn.functional.relu(x_175, inplace=False) + x_175 = None + pow_18 = relu_17**2 + relu_17 = None + mul_29 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + * pow_18 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( + pow_18 + ) = None + x_176 = ( + mul_29 + + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + ) + mul_29 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) + x_177 = torch.nn.functional.dropout(x_176, 0.0, False, False) + x_176 = None + x_178 = torch._C._nn.linear( + x_177, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_177 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_180 = mul_28 + x_179 + mul_28 = x_179 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_30 = x_180 * view_12 + view_12 = None + layer_norm_27 = torch.nn.functional.layer_norm( + x_180, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_180 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) + linear_24 = torch._C._nn.linear( + layer_norm_27, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_27 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_13 = linear_24.reshape(1, pow_25, 3, 10, 32) + linear_24 = None + qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) + reshape_13 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_181 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) + q_6 = k_6 = v_6 = None + transpose_7 = x_181.transpose(1, 2) + x_181 = None + x_182 = transpose_7.reshape(1, pow_25, 320) + transpose_7 = None + x_183 = torch._C._nn.linear( + x_182, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_182 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_184 = torch.nn.functional.dropout(x_183, 0.0, False, False) + x_183 = None + x_185 = mul_30 + x_184 + mul_30 = x_184 = None + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_31 = x_185 * view_13 + view_13 = None + layer_norm_28 = torch.nn.functional.layer_norm( + x_185, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_185 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) + x_186 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_28 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_18 = torch.nn.functional.relu(x_186, inplace=False) + x_186 = None + pow_19 = relu_18**2 + relu_18 = None + mul_32 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + * pow_19 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( + pow_19 + ) = None + x_187 = ( + mul_32 + + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + ) + mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) + x_188 = torch.nn.functional.dropout(x_187, 0.0, False, False) + x_187 = None + x_189 = torch._C._nn.linear( + x_188, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_188 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = mul_31 + x_190 + mul_31 = x_190 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_33 = x_191 * view_14 + view_14 = None + layer_norm_29 = torch.nn.functional.layer_norm( + x_191, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_191 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) + linear_28 = torch._C._nn.linear( + layer_norm_29, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_29 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_15 = linear_28.reshape(1, pow_25, 3, 10, 32) + linear_28 = None + qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) + reshape_15 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_192 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) + q_7 = k_7 = v_7 = None + transpose_8 = x_192.transpose(1, 2) + x_192 = None + x_193 = transpose_8.reshape(1, pow_25, 320) + transpose_8 = None + x_194 = torch._C._nn.linear( + x_193, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_193 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_195 = torch.nn.functional.dropout(x_194, 0.0, False, False) + x_194 = None + x_196 = mul_33 + x_195 + mul_33 = x_195 = None + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_34 = x_196 * view_15 + view_15 = None + layer_norm_30 = torch.nn.functional.layer_norm( + x_196, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_196 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) + x_197 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_30 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_19 = torch.nn.functional.relu(x_197, inplace=False) + x_197 = None + pow_20 = relu_19**2 + relu_19 = None + mul_35 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + * pow_20 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( + pow_20 + ) = None + x_198 = ( + mul_35 + + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + ) + mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) + x_199 = torch.nn.functional.dropout(x_198, 0.0, False, False) + x_198 = None + x_200 = torch._C._nn.linear( + x_199, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_199 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_202 = mul_34 + x_201 + mul_34 = x_201 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_36 = x_202 * view_16 + view_16 = None + layer_norm_31 = torch.nn.functional.layer_norm( + x_202, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_202 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) + linear_32 = torch._C._nn.linear( + layer_norm_31, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_31 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_17 = linear_32.reshape(1, pow_25, 3, 10, 32) + linear_32 = None + qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) + reshape_17 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_203 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) + q_8 = k_8 = v_8 = None + transpose_9 = x_203.transpose(1, 2) + x_203 = None + x_204 = transpose_9.reshape(1, pow_25, 320) + transpose_9 = pow_25 = None + x_205 = torch._C._nn.linear( + x_204, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_204 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_206 = torch.nn.functional.dropout(x_205, 0.0, False, False) + x_205 = None + x_207 = mul_36 + x_206 + mul_36 = x_206 = None + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_37 = x_207 * view_17 + view_17 = None + layer_norm_32 = torch.nn.functional.layer_norm( + x_207, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_207 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) + x_208 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_32 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_20 = torch.nn.functional.relu(x_208, inplace=False) + x_208 = None + pow_21 = relu_20**2 + relu_20 = None + mul_38 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + * pow_21 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( + pow_21 + ) = None + x_209 = ( + mul_38 + + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + ) + mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) + x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) + x_209 = None + x_211 = torch._C._nn.linear( + x_210, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_210 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_213 = mul_37 + x_212 + mul_37 = x_212 = None + transpose_10 = x_213.transpose(1, 2) + x_213 = None + x_214 = transpose_10.reshape(1, 320, sym_sum_3, sym_sum_3) + transpose_10 = sym_sum_3 = None + x_215 = x_214.permute(0, 2, 3, 1) + x_214 = None + x_216 = torch.nn.functional.layer_norm( + x_215, + (320,), + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_215 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) + x_217 = x_216.permute(0, 3, 1, 2) + x_216 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_217 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_3 = sym_sum // 32 + sym_sum = None + sym_sum_4 = torch.sym_sum([1, floordiv_3]) + floordiv_3 = None + reshape_20 = x_218.reshape(1, 512, -1) + x_218 = None + x_219 = reshape_20.transpose(1, 2) + reshape_20 = None + view_18 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_39 = x_219 * view_18 + view_18 = None + layer_norm_34 = torch.nn.functional.layer_norm( + x_219, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_219 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_26 = sym_sum_4**2 + linear_36 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_34 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_21 = linear_36.reshape(1, pow_26, 3, 16, 32) + linear_36 = None + qkv_9 = reshape_21.permute(2, 0, 3, 1, 4) + reshape_21 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_220 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) + q_9 = k_9 = v_9 = None + transpose_12 = x_220.transpose(1, 2) + x_220 = None + x_221 = transpose_12.reshape(1, pow_26, 512) + transpose_12 = None + x_222 = torch._C._nn.linear( + x_221, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_221 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_224 = mul_39 + x_223 + mul_39 = x_223 = None + view_19 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_40 = x_224 * view_19 + view_19 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_224, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_224 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_225 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_35 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_21 = torch.nn.functional.relu(x_225, inplace=False) + x_225 = None + pow_22 = relu_21**2 + relu_21 = None + mul_41 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_22 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_22 + ) = None + x_226 = ( + mul_41 + + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_41 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) + x_226 = None + x_228 = torch._C._nn.linear( + x_227, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_227 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_229 = torch.nn.functional.dropout(x_228, 0.0, False, False) + x_228 = None + x_230 = mul_40 + x_229 + mul_40 = x_229 = None + view_20 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_42 = x_230 * view_20 + view_20 = None + layer_norm_36 = torch.nn.functional.layer_norm( + x_230, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_230 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_36 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_23 = linear_40.reshape(1, pow_26, 3, 16, 32) + linear_40 = None + qkv_10 = reshape_23.permute(2, 0, 3, 1, 4) + reshape_23 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_231 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_13 = x_231.transpose(1, 2) + x_231 = None + x_232 = transpose_13.reshape(1, pow_26, 512) + transpose_13 = None + x_233 = torch._C._nn.linear( + x_232, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_232 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_235 = mul_42 + x_234 + mul_42 = x_234 = None + view_21 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_43 = x_235 * view_21 + view_21 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_235, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_235 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_236 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_37 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_22 = torch.nn.functional.relu(x_236, inplace=False) + x_236 = None + pow_23 = relu_22**2 + relu_22 = None + mul_44 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_23 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_23 + ) = None + x_237 = ( + mul_44 + + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_44 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_238 = torch.nn.functional.dropout(x_237, 0.0, False, False) + x_237 = None + x_239 = torch._C._nn.linear( + x_238, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_238 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_240 = torch.nn.functional.dropout(x_239, 0.0, False, False) + x_239 = None + x_241 = mul_43 + x_240 + mul_43 = x_240 = None + view_22 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_45 = x_241 * view_22 + view_22 = None + layer_norm_38 = torch.nn.functional.layer_norm( + x_241, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_241 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_38 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_25 = linear_44.reshape(1, pow_26, 3, 16, 32) + linear_44 = None + qkv_11 = reshape_25.permute(2, 0, 3, 1, 4) + reshape_25 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_242 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_14 = x_242.transpose(1, 2) + x_242 = None + x_243 = transpose_14.reshape(1, pow_26, 512) + transpose_14 = pow_26 = None + x_244 = torch._C._nn.linear( + x_243, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_243 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_246 = mul_45 + x_245 + mul_45 = x_245 = None + view_23 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_46 = x_246 * view_23 + view_23 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_246, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_246 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_247 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_39 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_23 = torch.nn.functional.relu(x_247, inplace=False) + x_247 = None + pow_24 = relu_23**2 + relu_23 = None + mul_47 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_24 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_24 + ) = None + x_248 = ( + mul_47 + + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_47 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_249 = torch.nn.functional.dropout(x_248, 0.0, False, False) + x_248 = None + x_250 = torch._C._nn.linear( + x_249, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_249 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_251 = torch.nn.functional.dropout(x_250, 0.0, False, False) + x_250 = None + x_252 = mul_46 + x_251 + mul_46 = x_251 = None + transpose_15 = x_252.transpose(1, 2) + x_252 = None + x_253 = transpose_15.reshape(1, 512, sym_sum_4, sym_sum_4) + transpose_15 = sym_sum_4 = None + x_254 = torch.nn.functional.adaptive_avg_pool2d(x_253, 1) + x_253 = None + x_255 = x_254.permute(0, 2, 3, 1) + x_254 = None + x_256 = torch.nn.functional.layer_norm( + x_255, + (512,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_255 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_257 = x_256.permute(0, 3, 1, 2) + x_256 = None + x_258 = x_257.flatten(1, -1) + x_257 = None + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = torch._C._nn.linear( + x_259, + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, + ) + x_259 = ( + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None + relu_24 = torch.nn.functional.relu(x_260, inplace=False) + x_260 = None + x_261 = torch.square(relu_24) + relu_24 = None + x_262 = torch.nn.functional.layer_norm( + x_261, + (2048,), + l_self_modules_head_modules_fc_modules_norm_parameters_weight_, + l_self_modules_head_modules_fc_modules_norm_parameters_bias_, + 1e-06, + ) + x_261 = ( + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None + x_263 = torch.nn.functional.dropout(x_262, 0.0, False, False) + x_262 = None + x_264 = torch._C._nn.linear( + x_263, + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, + ) + x_263 = ( + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None + return (x_264,) diff --git a/samples/timm/caformer_s18.sail_in1k_384/weight_meta.py b/samples/timm/caformer_s18.sail_in1k_384/weight_meta.py new file mode 100644 index 0000000000..51484b5182 --- /dev/null +++ b/samples/timm/caformer_s18.sail_in1k_384/weight_meta.py @@ -0,0 +1,2078 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv_parameters_weight_" + shape = [64, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: + name = "L_self_modules_stem_modules_conv_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.227 + std = 1.288 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: + name = "L_self_modules_stem_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" + shape = [128, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" + shape = [320, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" + shape = [512, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" + shape = [1000, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/caformer_s18.sail_in22k/graph_net.json b/samples/timm/caformer_s18.sail_in22k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/caformer_s18.sail_in22k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/caformer_s18.sail_in22k/input_meta.py b/samples/timm/caformer_s18.sail_in22k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_s18.sail_in22k/input_tensor_constraints.py b/samples/timm/caformer_s18.sail_in22k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/caformer_s18.sail_in22k/model.py b/samples/timm/caformer_s18.sail_in22k/model.py new file mode 100644 index 0000000000..ba7a8ecdba --- /dev/null +++ b/samples/timm/caformer_s18.sail_in22k/model.py @@ -0,0 +1,2468 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv_parameters_bias_ = ( + L_self_modules_stem_modules_conv_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_norm_parameters_weight_ = ( + L_self_modules_stem_modules_norm_parameters_weight_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( + L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv_parameters_weight_, + l_self_modules_stem_modules_conv_parameters_bias_, + (4, 4), + (2, 2), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_conv_parameters_weight_ + ) = l_self_modules_stem_modules_conv_parameters_bias_ = None + x_1 = x.permute(0, 2, 3, 1) + x = None + x_2 = torch.nn.functional.layer_norm( + x_1, (64,), l_self_modules_stem_modules_norm_parameters_weight_, None, 1e-06 + ) + x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None + x_3 = x_2.permute(0, 3, 1, 2) + x_2 = None + sym_sum = torch.sym_sum([-3, s1]) + s1 = None + floordiv = sym_sum // 4 + sym_sum_1 = torch.sym_sum([1, floordiv]) + floordiv = sym_sum_1 = None + x_4 = x_3.permute(0, 2, 3, 1) + x_5 = torch.nn.functional.layer_norm( + x_4, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu = torch.nn.functional.relu(x_7, inplace=False) + x_7 = None + pow_1 = relu**2 + relu = None + mul = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_1 + ) = None + x_8 = ( + mul + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_11 = x_3 + x_10 + x_3 = x_10 = None + x_12 = x_11.permute(0, 2, 3, 1) + x_13 = torch.nn.functional.layer_norm( + x_12, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_14 = x_13.permute(0, 3, 1, 2) + x_13 = None + x_15 = torch.conv2d( + x_14, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_1 = torch.nn.functional.relu(x_15, inplace=False) + x_15 = None + pow_2 = relu_1**2 + relu_1 = None + mul_1 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_2 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_2 + ) = None + x_16 = ( + mul_1 + + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = x_11 + x_19 + x_11 = x_19 = None + x_21 = x_20.permute(0, 2, 3, 1) + x_22 = torch.nn.functional.layer_norm( + x_21, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = torch.conv2d( + x_23, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_2 = torch.nn.functional.relu(x_24, inplace=False) + x_24 = None + pow_3 = relu_2**2 + relu_2 = None + mul_2 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_3 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_3 + ) = None + x_25 = ( + mul_2 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_27 = torch.conv2d( + x_26, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_28 = x_20 + x_27 + x_20 = x_27 = None + x_29 = x_28.permute(0, 2, 3, 1) + x_30 = torch.nn.functional.layer_norm( + x_29, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_3 = torch.nn.functional.relu(x_32, inplace=False) + x_32 = None + pow_4 = relu_3**2 + relu_3 = None + mul_3 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_4 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_4 + ) = None + x_33 = ( + mul_3 + + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + x_37 = x_28 + x_36 + x_28 = x_36 = None + x_38 = x_37.permute(0, 2, 3, 1) + x_39 = torch.nn.functional.layer_norm( + x_38, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_40 = x_39.permute(0, 3, 1, 2) + x_39 = None + x_41 = torch.conv2d( + x_40, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_4 = torch.nn.functional.relu(x_41, inplace=False) + x_41 = None + pow_5 = relu_4**2 + relu_4 = None + mul_4 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_5 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_5 + ) = None + x_42 = ( + mul_4 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_44 = torch.conv2d( + x_43, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_45 = x_37 + x_44 + x_37 = x_44 = None + x_46 = x_45.permute(0, 2, 3, 1) + x_47 = torch.nn.functional.layer_norm( + x_46, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_5 = torch.nn.functional.relu(x_49, inplace=False) + x_49 = None + pow_6 = relu_5**2 + relu_5 = None + mul_5 = ( + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_6 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_6 + ) = None + x_50 = ( + mul_5 + + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = x_45 + x_53 + x_45 = x_53 = None + x_55 = x_54.permute(0, 2, 3, 1) + x_54 = None + x_56 = torch.nn.functional.layer_norm( + x_55, + (64,), + l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) + x_57 = x_56.permute(0, 3, 1, 2) + x_56 = None + x_58 = torch.conv2d( + x_57, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_1 = sym_sum // 8 + sym_sum_2 = torch.sym_sum([1, floordiv_1]) + floordiv_1 = sym_sum_2 = None + x_59 = x_58.permute(0, 2, 3, 1) + x_60 = torch.nn.functional.layer_norm( + x_59, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + x_61 = x_60.permute(0, 3, 1, 2) + x_60 = None + x_62 = torch.conv2d( + x_61, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_6 = torch.nn.functional.relu(x_62, inplace=False) + x_62 = None + pow_7 = relu_6**2 + relu_6 = None + mul_6 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ + * pow_7 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_7 + ) = None + x_63 = ( + mul_6 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_66 = x_58 + x_65 + x_58 = x_65 = None + x_67 = x_66.permute(0, 2, 3, 1) + x_68 = torch.nn.functional.layer_norm( + x_67, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + x_70 = torch.conv2d( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_7 = torch.nn.functional.relu(x_70, inplace=False) + x_70 = None + pow_8 = relu_7**2 + relu_7 = None + mul_7 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_8 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_8 + ) = None + x_71 = ( + mul_7 + + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + x_73 = torch.conv2d( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + x_75 = x_66 + x_74 + x_66 = x_74 = None + x_76 = x_75.permute(0, 2, 3, 1) + x_77 = torch.nn.functional.layer_norm( + x_76, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_8 = torch.nn.functional.relu(x_79, inplace=False) + x_79 = None + pow_9 = relu_8**2 + relu_8 = None + mul_8 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ + * pow_9 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_9 + ) = None + x_80 = ( + mul_8 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_81 = torch.conv2d( + x_80, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_83 = x_75 + x_82 + x_75 = x_82 = None + x_84 = x_83.permute(0, 2, 3, 1) + x_85 = torch.nn.functional.layer_norm( + x_84, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_86 = x_85.permute(0, 3, 1, 2) + x_85 = None + x_87 = torch.conv2d( + x_86, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_9 = torch.nn.functional.relu(x_87, inplace=False) + x_87 = None + pow_10 = relu_9**2 + relu_9 = None + mul_9 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_10 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_10 + ) = None + x_88 = ( + mul_9 + + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = x_83 + x_91 + x_83 = x_91 = None + x_93 = x_92.permute(0, 2, 3, 1) + x_94 = torch.nn.functional.layer_norm( + x_93, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = torch.conv2d( + x_95, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) + relu_10 = torch.nn.functional.relu(x_96, inplace=False) + x_96 = None + pow_11 = relu_10**2 + relu_10 = None + mul_10 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ + * pow_11 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( + pow_11 + ) = None + x_97 = ( + mul_10 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ + ) + mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) + x_98 = torch.conv2d( + x_97, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) + x_99 = torch.conv2d( + x_98, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) + x_100 = x_92 + x_99 + x_92 = x_99 = None + x_101 = x_100.permute(0, 2, 3, 1) + x_102 = torch.nn.functional.layer_norm( + x_101, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_103 = x_102.permute(0, 3, 1, 2) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_11 = torch.nn.functional.relu(x_104, inplace=False) + x_104 = None + pow_12 = relu_11**2 + relu_11 = None + mul_11 = ( + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_12 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_12 + ) = None + x_105 = ( + mul_11 + + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_100 + x_108 + x_100 = x_108 = None + x_110 = x_109.permute(0, 2, 3, 1) + x_109 = None + x_111 = torch.nn.functional.layer_norm( + x_110, + (128,), + l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_110 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) + x_112 = x_111.permute(0, 3, 1, 2) + x_111 = None + x_113 = torch.conv2d( + x_112, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_112 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_2 = sym_sum // 16 + sym_sum_3 = torch.sym_sum([1, floordiv_2]) + floordiv_2 = None + reshape = x_113.reshape(1, 320, -1) + x_113 = None + x_114 = reshape.transpose(1, 2) + reshape = None + view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_12 = x_114 * view + view = None + layer_norm_15 = torch.nn.functional.layer_norm( + x_114, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_25 = sym_sum_3**2 + linear = torch._C._nn.linear( + layer_norm_15, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_15 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_1 = linear.reshape(1, pow_25, 3, 10, 32) + linear = None + qkv = reshape_1.permute(2, 0, 3, 1, 4) + reshape_1 = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_115 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) + q = k = v = None + transpose_1 = x_115.transpose(1, 2) + x_115 = None + x_116 = transpose_1.reshape(1, pow_25, 320) + transpose_1 = None + x_117 = torch._C._nn.linear( + x_116, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_116 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = mul_12 + x_118 + mul_12 = x_118 = None + view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_13 = x_119 * view_1 + view_1 = None + layer_norm_16 = torch.nn.functional.layer_norm( + x_119, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_119 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_120 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_16 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_12 = torch.nn.functional.relu(x_120, inplace=False) + x_120 = None + pow_13 = relu_12**2 + relu_12 = None + mul_14 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_13 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_13 + ) = None + x_121 = ( + mul_14 + + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_14 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_122 = torch.nn.functional.dropout(x_121, 0.0, False, False) + x_121 = None + x_123 = torch._C._nn.linear( + x_122, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = mul_13 + x_124 + mul_13 = x_124 = None + view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_15 = x_125 * view_2 + view_2 = None + layer_norm_17 = torch.nn.functional.layer_norm( + x_125, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_17, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_17 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_3 = linear_4.reshape(1, pow_25, 3, 10, 32) + linear_4 = None + qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) + reshape_3 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_126 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) + q_1 = k_1 = v_1 = None + transpose_2 = x_126.transpose(1, 2) + x_126 = None + x_127 = transpose_2.reshape(1, pow_25, 320) + transpose_2 = None + x_128 = torch._C._nn.linear( + x_127, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_127 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_129 = torch.nn.functional.dropout(x_128, 0.0, False, False) + x_128 = None + x_130 = mul_15 + x_129 + mul_15 = x_129 = None + view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_16 = x_130 * view_3 + view_3 = None + layer_norm_18 = torch.nn.functional.layer_norm( + x_130, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_130 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_131 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_18 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_13 = torch.nn.functional.relu(x_131, inplace=False) + x_131 = None + pow_14 = relu_13**2 + relu_13 = None + mul_17 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_14 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_14 + ) = None + x_132 = ( + mul_17 + + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_17 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + x_134 = torch._C._nn.linear( + x_133, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_133 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = mul_16 + x_135 + mul_16 = x_135 = None + view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_18 = x_136 * view_4 + view_4 = None + layer_norm_19 = torch.nn.functional.layer_norm( + x_136, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_19, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_19 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_5 = linear_8.reshape(1, pow_25, 3, 10, 32) + linear_8 = None + qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) + reshape_5 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_137 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) + q_2 = k_2 = v_2 = None + transpose_3 = x_137.transpose(1, 2) + x_137 = None + x_138 = transpose_3.reshape(1, pow_25, 320) + transpose_3 = None + x_139 = torch._C._nn.linear( + x_138, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_138 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) + x_139 = None + x_141 = mul_18 + x_140 + mul_18 = x_140 = None + view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_19 = x_141 * view_5 + view_5 = None + layer_norm_20 = torch.nn.functional.layer_norm( + x_141, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_141 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_142 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_20 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_14 = torch.nn.functional.relu(x_142, inplace=False) + x_142 = None + pow_15 = relu_14**2 + relu_14 = None + mul_20 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_15 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_15 + ) = None + x_143 = ( + mul_20 + + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_20 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) + x_143 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_144 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = mul_19 + x_146 + mul_19 = x_146 = None + view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_21 = x_147 * view_6 + view_6 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_147, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_147 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_21 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_7 = linear_12.reshape(1, pow_25, 3, 10, 32) + linear_12 = None + qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) + reshape_7 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_148 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) + q_3 = k_3 = v_3 = None + transpose_4 = x_148.transpose(1, 2) + x_148 = None + x_149 = transpose_4.reshape(1, pow_25, 320) + transpose_4 = None + x_150 = torch._C._nn.linear( + x_149, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_149 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_151 = torch.nn.functional.dropout(x_150, 0.0, False, False) + x_150 = None + x_152 = mul_21 + x_151 + mul_21 = x_151 = None + view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_22 = x_152 * view_7 + view_7 = None + layer_norm_22 = torch.nn.functional.layer_norm( + x_152, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_152 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) + x_153 = torch._C._nn.linear( + layer_norm_22, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_22 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_15 = torch.nn.functional.relu(x_153, inplace=False) + x_153 = None + pow_16 = relu_15**2 + relu_15 = None + mul_23 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ + * pow_16 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( + pow_16 + ) = None + x_154 = ( + mul_23 + + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ + ) + mul_23 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) + x_155 = torch.nn.functional.dropout(x_154, 0.0, False, False) + x_154 = None + x_156 = torch._C._nn.linear( + x_155, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_155 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = mul_22 + x_157 + mul_22 = x_157 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_24 = x_158 * view_8 + view_8 = None + layer_norm_23 = torch.nn.functional.layer_norm( + x_158, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_23, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_23 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_9 = linear_16.reshape(1, pow_25, 3, 10, 32) + linear_16 = None + qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) + reshape_9 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_159 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) + q_4 = k_4 = v_4 = None + transpose_5 = x_159.transpose(1, 2) + x_159 = None + x_160 = transpose_5.reshape(1, pow_25, 320) + transpose_5 = None + x_161 = torch._C._nn.linear( + x_160, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_160 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_162 = torch.nn.functional.dropout(x_161, 0.0, False, False) + x_161 = None + x_163 = mul_24 + x_162 + mul_24 = x_162 = None + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_25 = x_163 * view_9 + view_9 = None + layer_norm_24 = torch.nn.functional.layer_norm( + x_163, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_163 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) + x_164 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_24 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_16 = torch.nn.functional.relu(x_164, inplace=False) + x_164 = None + pow_17 = relu_16**2 + relu_16 = None + mul_26 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ + * pow_17 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( + pow_17 + ) = None + x_165 = ( + mul_26 + + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ + ) + mul_26 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) + x_166 = torch.nn.functional.dropout(x_165, 0.0, False, False) + x_165 = None + x_167 = torch._C._nn.linear( + x_166, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_166 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_169 = mul_25 + x_168 + mul_25 = x_168 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_27 = x_169 * view_10 + view_10 = None + layer_norm_25 = torch.nn.functional.layer_norm( + x_169, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) + linear_20 = torch._C._nn.linear( + layer_norm_25, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_25 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_11 = linear_20.reshape(1, pow_25, 3, 10, 32) + linear_20 = None + qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) + reshape_11 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_170 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) + q_5 = k_5 = v_5 = None + transpose_6 = x_170.transpose(1, 2) + x_170 = None + x_171 = transpose_6.reshape(1, pow_25, 320) + transpose_6 = None + x_172 = torch._C._nn.linear( + x_171, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_171 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_173 = torch.nn.functional.dropout(x_172, 0.0, False, False) + x_172 = None + x_174 = mul_27 + x_173 + mul_27 = x_173 = None + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_28 = x_174 * view_11 + view_11 = None + layer_norm_26 = torch.nn.functional.layer_norm( + x_174, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_174 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) + x_175 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_26 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_17 = torch.nn.functional.relu(x_175, inplace=False) + x_175 = None + pow_18 = relu_17**2 + relu_17 = None + mul_29 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ + * pow_18 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( + pow_18 + ) = None + x_176 = ( + mul_29 + + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ + ) + mul_29 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) + x_177 = torch.nn.functional.dropout(x_176, 0.0, False, False) + x_176 = None + x_178 = torch._C._nn.linear( + x_177, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_177 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_180 = mul_28 + x_179 + mul_28 = x_179 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_30 = x_180 * view_12 + view_12 = None + layer_norm_27 = torch.nn.functional.layer_norm( + x_180, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_180 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) + linear_24 = torch._C._nn.linear( + layer_norm_27, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_27 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_13 = linear_24.reshape(1, pow_25, 3, 10, 32) + linear_24 = None + qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) + reshape_13 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_181 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) + q_6 = k_6 = v_6 = None + transpose_7 = x_181.transpose(1, 2) + x_181 = None + x_182 = transpose_7.reshape(1, pow_25, 320) + transpose_7 = None + x_183 = torch._C._nn.linear( + x_182, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_182 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_184 = torch.nn.functional.dropout(x_183, 0.0, False, False) + x_183 = None + x_185 = mul_30 + x_184 + mul_30 = x_184 = None + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_31 = x_185 * view_13 + view_13 = None + layer_norm_28 = torch.nn.functional.layer_norm( + x_185, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_185 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) + x_186 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_28 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_18 = torch.nn.functional.relu(x_186, inplace=False) + x_186 = None + pow_19 = relu_18**2 + relu_18 = None + mul_32 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ + * pow_19 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( + pow_19 + ) = None + x_187 = ( + mul_32 + + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ + ) + mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) + x_188 = torch.nn.functional.dropout(x_187, 0.0, False, False) + x_187 = None + x_189 = torch._C._nn.linear( + x_188, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_188 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = mul_31 + x_190 + mul_31 = x_190 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_33 = x_191 * view_14 + view_14 = None + layer_norm_29 = torch.nn.functional.layer_norm( + x_191, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_191 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) + linear_28 = torch._C._nn.linear( + layer_norm_29, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_29 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_15 = linear_28.reshape(1, pow_25, 3, 10, 32) + linear_28 = None + qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) + reshape_15 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_192 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) + q_7 = k_7 = v_7 = None + transpose_8 = x_192.transpose(1, 2) + x_192 = None + x_193 = transpose_8.reshape(1, pow_25, 320) + transpose_8 = None + x_194 = torch._C._nn.linear( + x_193, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_193 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_195 = torch.nn.functional.dropout(x_194, 0.0, False, False) + x_194 = None + x_196 = mul_33 + x_195 + mul_33 = x_195 = None + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_34 = x_196 * view_15 + view_15 = None + layer_norm_30 = torch.nn.functional.layer_norm( + x_196, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_196 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) + x_197 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_30 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_19 = torch.nn.functional.relu(x_197, inplace=False) + x_197 = None + pow_20 = relu_19**2 + relu_19 = None + mul_35 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ + * pow_20 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( + pow_20 + ) = None + x_198 = ( + mul_35 + + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ + ) + mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) + x_199 = torch.nn.functional.dropout(x_198, 0.0, False, False) + x_198 = None + x_200 = torch._C._nn.linear( + x_199, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_199 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_202 = mul_34 + x_201 + mul_34 = x_201 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_36 = x_202 * view_16 + view_16 = None + layer_norm_31 = torch.nn.functional.layer_norm( + x_202, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_202 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) + linear_32 = torch._C._nn.linear( + layer_norm_31, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_31 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_17 = linear_32.reshape(1, pow_25, 3, 10, 32) + linear_32 = None + qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) + reshape_17 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_203 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) + q_8 = k_8 = v_8 = None + transpose_9 = x_203.transpose(1, 2) + x_203 = None + x_204 = transpose_9.reshape(1, pow_25, 320) + transpose_9 = pow_25 = None + x_205 = torch._C._nn.linear( + x_204, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_204 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_206 = torch.nn.functional.dropout(x_205, 0.0, False, False) + x_205 = None + x_207 = mul_36 + x_206 + mul_36 = x_206 = None + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( + (320,) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_37 = x_207 * view_17 + view_17 = None + layer_norm_32 = torch.nn.functional.layer_norm( + x_207, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_207 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) + x_208 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_32 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_20 = torch.nn.functional.relu(x_208, inplace=False) + x_208 = None + pow_21 = relu_20**2 + relu_20 = None + mul_38 = ( + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ + * pow_21 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( + pow_21 + ) = None + x_209 = ( + mul_38 + + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ + ) + mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) + x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) + x_209 = None + x_211 = torch._C._nn.linear( + x_210, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_210 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_213 = mul_37 + x_212 + mul_37 = x_212 = None + transpose_10 = x_213.transpose(1, 2) + x_213 = None + x_214 = transpose_10.reshape(1, 320, sym_sum_3, sym_sum_3) + transpose_10 = sym_sum_3 = None + x_215 = x_214.permute(0, 2, 3, 1) + x_214 = None + x_216 = torch.nn.functional.layer_norm( + x_215, + (320,), + l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, + None, + 1e-06, + ) + x_215 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) + x_217 = x_216.permute(0, 3, 1, 2) + x_216 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_217 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) + floordiv_3 = sym_sum // 32 + sym_sum = None + sym_sum_4 = torch.sym_sum([1, floordiv_3]) + floordiv_3 = None + reshape_20 = x_218.reshape(1, 512, -1) + x_218 = None + x_219 = reshape_20.transpose(1, 2) + reshape_20 = None + view_18 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_39 = x_219 * view_18 + view_18 = None + layer_norm_34 = torch.nn.functional.layer_norm( + x_219, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_219 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) + pow_26 = sym_sum_4**2 + linear_36 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_34 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_21 = linear_36.reshape(1, pow_26, 3, 16, 32) + linear_36 = None + qkv_9 = reshape_21.permute(2, 0, 3, 1, 4) + reshape_21 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_220 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) + q_9 = k_9 = v_9 = None + transpose_12 = x_220.transpose(1, 2) + x_220 = None + x_221 = transpose_12.reshape(1, pow_26, 512) + transpose_12 = None + x_222 = torch._C._nn.linear( + x_221, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_221 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_224 = mul_39 + x_223 + mul_39 = x_223 = None + view_19 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_40 = x_224 * view_19 + view_19 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_224, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_224 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) + x_225 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_35 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_21 = torch.nn.functional.relu(x_225, inplace=False) + x_225 = None + pow_22 = relu_21**2 + relu_21 = None + mul_41 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ + * pow_22 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( + pow_22 + ) = None + x_226 = ( + mul_41 + + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ + ) + mul_41 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) + x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) + x_226 = None + x_228 = torch._C._nn.linear( + x_227, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_227 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_229 = torch.nn.functional.dropout(x_228, 0.0, False, False) + x_228 = None + x_230 = mul_40 + x_229 + mul_40 = x_229 = None + view_20 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_42 = x_230 * view_20 + view_20 = None + layer_norm_36 = torch.nn.functional.layer_norm( + x_230, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_230 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_36 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_23 = linear_40.reshape(1, pow_26, 3, 16, 32) + linear_40 = None + qkv_10 = reshape_23.permute(2, 0, 3, 1, 4) + reshape_23 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_231 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_13 = x_231.transpose(1, 2) + x_231 = None + x_232 = transpose_13.reshape(1, pow_26, 512) + transpose_13 = None + x_233 = torch._C._nn.linear( + x_232, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_232 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_235 = mul_42 + x_234 + mul_42 = x_234 = None + view_21 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_43 = x_235 * view_21 + view_21 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_235, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_235 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) + x_236 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_37 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_22 = torch.nn.functional.relu(x_236, inplace=False) + x_236 = None + pow_23 = relu_22**2 + relu_22 = None + mul_44 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ + * pow_23 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( + pow_23 + ) = None + x_237 = ( + mul_44 + + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ + ) + mul_44 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) + x_238 = torch.nn.functional.dropout(x_237, 0.0, False, False) + x_237 = None + x_239 = torch._C._nn.linear( + x_238, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_238 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_240 = torch.nn.functional.dropout(x_239, 0.0, False, False) + x_239 = None + x_241 = mul_43 + x_240 + mul_43 = x_240 = None + view_22 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( + None + ) + mul_45 = x_241 * view_22 + view_22 = None + layer_norm_38 = torch.nn.functional.layer_norm( + x_241, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, + None, + 1e-06, + ) + x_241 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, + None, + ) + layer_norm_38 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) + reshape_25 = linear_44.reshape(1, pow_26, 3, 16, 32) + linear_44 = None + qkv_11 = reshape_25.permute(2, 0, 3, 1, 4) + reshape_25 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_242 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_14 = x_242.transpose(1, 2) + x_242 = None + x_243 = transpose_14.reshape(1, pow_26, 512) + transpose_14 = pow_26 = None + x_244 = torch._C._nn.linear( + x_243, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, + None, + ) + x_243 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_246 = mul_45 + x_245 + mul_45 = x_245 = None + view_23 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( + (512,) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( + None + ) + mul_46 = x_246 * view_23 + view_23 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_246, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, + None, + 1e-06, + ) + x_246 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) + x_247 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + None, + ) + layer_norm_39 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) + relu_23 = torch.nn.functional.relu(x_247, inplace=False) + x_247 = None + pow_24 = relu_23**2 + relu_23 = None + mul_47 = ( + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ + * pow_24 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( + pow_24 + ) = None + x_248 = ( + mul_47 + + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ + ) + mul_47 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) + x_249 = torch.nn.functional.dropout(x_248, 0.0, False, False) + x_248 = None + x_250 = torch._C._nn.linear( + x_249, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + None, + ) + x_249 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) + x_251 = torch.nn.functional.dropout(x_250, 0.0, False, False) + x_250 = None + x_252 = mul_46 + x_251 + mul_46 = x_251 = None + transpose_15 = x_252.transpose(1, 2) + x_252 = None + x_253 = transpose_15.reshape(1, 512, sym_sum_4, sym_sum_4) + transpose_15 = sym_sum_4 = None + x_254 = torch.nn.functional.adaptive_avg_pool2d(x_253, 1) + x_253 = None + x_255 = x_254.permute(0, 2, 3, 1) + x_254 = None + x_256 = torch.nn.functional.layer_norm( + x_255, + (512,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_255 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_257 = x_256.permute(0, 3, 1, 2) + x_256 = None + x_258 = x_257.flatten(1, -1) + x_257 = None + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = torch._C._nn.linear( + x_259, + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, + ) + x_259 = ( + l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None + relu_24 = torch.nn.functional.relu(x_260, inplace=False) + x_260 = None + x_261 = torch.square(relu_24) + relu_24 = None + x_262 = torch.nn.functional.layer_norm( + x_261, + (2048,), + l_self_modules_head_modules_fc_modules_norm_parameters_weight_, + l_self_modules_head_modules_fc_modules_norm_parameters_bias_, + 1e-06, + ) + x_261 = ( + l_self_modules_head_modules_fc_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None + x_263 = torch.nn.functional.dropout(x_262, 0.0, False, False) + x_262 = None + x_264 = torch._C._nn.linear( + x_263, + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, + l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, + ) + x_263 = ( + l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ + ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None + return (x_264,) diff --git a/samples/timm/caformer_s18.sail_in22k/weight_meta.py b/samples/timm/caformer_s18.sail_in22k/weight_meta.py new file mode 100644 index 0000000000..c663bbb64a --- /dev/null +++ b/samples/timm/caformer_s18.sail_in22k/weight_meta.py @@ -0,0 +1,2078 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv_parameters_weight_" + shape = [64, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: + name = "L_self_modules_stem_modules_conv_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.226 + std = 1.288 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: + name = "L_self_modules_stem_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" + shape = [128, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" + shape = [320, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" + shape = [512, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" + shape = [1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = nan + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" + shape = [21841, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: + name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" + shape = [21841] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/timm_success_list.txt b/timm_success_list.txt new file mode 100644 index 0000000000..e69de29bb2 From 28853ef03dff1170a157fc65b4b2f57248484e3d Mon Sep 17 00:00:00 2001 From: Liyulingyue <852433440@qq.com> Date: Mon, 4 Aug 2025 22:56:00 +0800 Subject: [PATCH 3/7] timm group --- .../caformer_b36.sail_in1k/graph_net.json | 5 - .../timm/caformer_b36.sail_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/caformer_b36.sail_in1k/model.py | 4758 ----------------- .../caformer_b36.sail_in1k/weight_meta.py | 3968 -------------- .../caformer_b36.sail_in1k_384/graph_net.json | 5 - .../caformer_b36.sail_in1k_384/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/caformer_b36.sail_in1k_384/model.py | 4758 ----------------- .../caformer_b36.sail_in1k_384/weight_meta.py | 3968 -------------- .../caformer_b36.sail_in22k/graph_net.json | 5 - .../caformer_b36.sail_in22k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/caformer_b36.sail_in22k/model.py | 4758 ----------------- .../caformer_b36.sail_in22k/weight_meta.py | 3968 -------------- .../graph_net.json | 5 - .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../caformer_b36.sail_in22k_ft_in1k/model.py | 4758 ----------------- .../weight_meta.py | 3968 -------------- .../graph_net.json | 5 - .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../model.py | 4758 ----------------- .../weight_meta.py | 3968 -------------- .../caformer_m36.sail_in1k/graph_net.json | 5 - .../timm/caformer_m36.sail_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/caformer_m36.sail_in1k/model.py | 4754 ---------------- .../caformer_m36.sail_in1k/weight_meta.py | 3968 -------------- .../caformer_m36.sail_in1k_384/graph_net.json | 5 - .../caformer_m36.sail_in1k_384/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/caformer_m36.sail_in1k_384/model.py | 4754 ---------------- .../caformer_m36.sail_in1k_384/weight_meta.py | 3968 -------------- .../caformer_m36.sail_in22k/graph_net.json | 5 - .../caformer_m36.sail_in22k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/caformer_m36.sail_in22k/model.py | 4754 ---------------- .../caformer_m36.sail_in22k/weight_meta.py | 3968 -------------- .../graph_net.json | 5 - .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../caformer_m36.sail_in22k_ft_in1k/model.py | 4754 ---------------- .../weight_meta.py | 3968 -------------- .../graph_net.json | 5 - .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../model.py | 4754 ---------------- .../weight_meta.py | 3968 -------------- .../caformer_s18.sail_in1k/graph_net.json | 5 - .../timm/caformer_s18.sail_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/caformer_s18.sail_in1k/model.py | 2468 --------- .../caformer_s18.sail_in1k/weight_meta.py | 2078 ------- .../caformer_s18.sail_in1k_384/graph_net.json | 5 - .../caformer_s18.sail_in1k_384/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/caformer_s18.sail_in1k_384/model.py | 2468 --------- .../caformer_s18.sail_in1k_384/weight_meta.py | 2078 ------- .../caformer_s18.sail_in22k/graph_net.json | 5 - .../caformer_s18.sail_in22k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/caformer_s18.sail_in22k/model.py | 2468 --------- .../caformer_s18.sail_in22k/weight_meta.py | 2078 ------- timm_success_list.txt | 0 66 files changed, 100943 deletions(-) delete mode 100644 samples/timm/caformer_b36.sail_in1k/graph_net.json delete mode 100644 samples/timm/caformer_b36.sail_in1k/input_meta.py delete mode 100644 samples/timm/caformer_b36.sail_in1k/input_tensor_constraints.py delete mode 100644 samples/timm/caformer_b36.sail_in1k/model.py delete mode 100644 samples/timm/caformer_b36.sail_in1k/weight_meta.py delete mode 100644 samples/timm/caformer_b36.sail_in1k_384/graph_net.json delete mode 100644 samples/timm/caformer_b36.sail_in1k_384/input_meta.py delete mode 100644 samples/timm/caformer_b36.sail_in1k_384/input_tensor_constraints.py delete mode 100644 samples/timm/caformer_b36.sail_in1k_384/model.py delete mode 100644 samples/timm/caformer_b36.sail_in1k_384/weight_meta.py delete mode 100644 samples/timm/caformer_b36.sail_in22k/graph_net.json delete mode 100644 samples/timm/caformer_b36.sail_in22k/input_meta.py delete mode 100644 samples/timm/caformer_b36.sail_in22k/input_tensor_constraints.py delete mode 100644 samples/timm/caformer_b36.sail_in22k/model.py delete mode 100644 samples/timm/caformer_b36.sail_in22k/weight_meta.py delete mode 100644 samples/timm/caformer_b36.sail_in22k_ft_in1k/graph_net.json delete mode 100644 samples/timm/caformer_b36.sail_in22k_ft_in1k/input_meta.py delete mode 100644 samples/timm/caformer_b36.sail_in22k_ft_in1k/input_tensor_constraints.py delete mode 100644 samples/timm/caformer_b36.sail_in22k_ft_in1k/model.py delete mode 100644 samples/timm/caformer_b36.sail_in22k_ft_in1k/weight_meta.py delete mode 100644 samples/timm/caformer_b36.sail_in22k_ft_in1k_384/graph_net.json delete mode 100644 samples/timm/caformer_b36.sail_in22k_ft_in1k_384/input_meta.py delete mode 100644 samples/timm/caformer_b36.sail_in22k_ft_in1k_384/input_tensor_constraints.py delete mode 100644 samples/timm/caformer_b36.sail_in22k_ft_in1k_384/model.py delete mode 100644 samples/timm/caformer_b36.sail_in22k_ft_in1k_384/weight_meta.py delete mode 100644 samples/timm/caformer_m36.sail_in1k/graph_net.json delete mode 100644 samples/timm/caformer_m36.sail_in1k/input_meta.py delete mode 100644 samples/timm/caformer_m36.sail_in1k/input_tensor_constraints.py delete mode 100644 samples/timm/caformer_m36.sail_in1k/model.py delete mode 100644 samples/timm/caformer_m36.sail_in1k/weight_meta.py delete mode 100644 samples/timm/caformer_m36.sail_in1k_384/graph_net.json delete mode 100644 samples/timm/caformer_m36.sail_in1k_384/input_meta.py delete mode 100644 samples/timm/caformer_m36.sail_in1k_384/input_tensor_constraints.py delete mode 100644 samples/timm/caformer_m36.sail_in1k_384/model.py delete mode 100644 samples/timm/caformer_m36.sail_in1k_384/weight_meta.py delete mode 100644 samples/timm/caformer_m36.sail_in22k/graph_net.json delete mode 100644 samples/timm/caformer_m36.sail_in22k/input_meta.py delete mode 100644 samples/timm/caformer_m36.sail_in22k/input_tensor_constraints.py delete mode 100644 samples/timm/caformer_m36.sail_in22k/model.py delete mode 100644 samples/timm/caformer_m36.sail_in22k/weight_meta.py delete mode 100644 samples/timm/caformer_m36.sail_in22k_ft_in1k/graph_net.json delete mode 100644 samples/timm/caformer_m36.sail_in22k_ft_in1k/input_meta.py delete mode 100644 samples/timm/caformer_m36.sail_in22k_ft_in1k/input_tensor_constraints.py delete mode 100644 samples/timm/caformer_m36.sail_in22k_ft_in1k/model.py delete mode 100644 samples/timm/caformer_m36.sail_in22k_ft_in1k/weight_meta.py delete mode 100644 samples/timm/caformer_m36.sail_in22k_ft_in1k_384/graph_net.json delete mode 100644 samples/timm/caformer_m36.sail_in22k_ft_in1k_384/input_meta.py delete mode 100644 samples/timm/caformer_m36.sail_in22k_ft_in1k_384/input_tensor_constraints.py delete mode 100644 samples/timm/caformer_m36.sail_in22k_ft_in1k_384/model.py delete mode 100644 samples/timm/caformer_m36.sail_in22k_ft_in1k_384/weight_meta.py delete mode 100644 samples/timm/caformer_s18.sail_in1k/graph_net.json delete mode 100644 samples/timm/caformer_s18.sail_in1k/input_meta.py delete mode 100644 samples/timm/caformer_s18.sail_in1k/input_tensor_constraints.py delete mode 100644 samples/timm/caformer_s18.sail_in1k/model.py delete mode 100644 samples/timm/caformer_s18.sail_in1k/weight_meta.py delete mode 100644 samples/timm/caformer_s18.sail_in1k_384/graph_net.json delete mode 100644 samples/timm/caformer_s18.sail_in1k_384/input_meta.py delete mode 100644 samples/timm/caformer_s18.sail_in1k_384/input_tensor_constraints.py delete mode 100644 samples/timm/caformer_s18.sail_in1k_384/model.py delete mode 100644 samples/timm/caformer_s18.sail_in1k_384/weight_meta.py delete mode 100644 samples/timm/caformer_s18.sail_in22k/graph_net.json delete mode 100644 samples/timm/caformer_s18.sail_in22k/input_meta.py delete mode 100644 samples/timm/caformer_s18.sail_in22k/input_tensor_constraints.py delete mode 100644 samples/timm/caformer_s18.sail_in22k/model.py delete mode 100644 samples/timm/caformer_s18.sail_in22k/weight_meta.py delete mode 100644 timm_success_list.txt diff --git a/samples/timm/caformer_b36.sail_in1k/graph_net.json b/samples/timm/caformer_b36.sail_in1k/graph_net.json deleted file mode 100644 index 1373fe3b58..0000000000 --- a/samples/timm/caformer_b36.sail_in1k/graph_net.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "framework": "torch", - "num_devices_required": 1, - "num_nodes_required": 1 -} \ No newline at end of file diff --git a/samples/timm/caformer_b36.sail_in1k/input_meta.py b/samples/timm/caformer_b36.sail_in1k/input_meta.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_b36.sail_in1k/input_tensor_constraints.py b/samples/timm/caformer_b36.sail_in1k/input_tensor_constraints.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_b36.sail_in1k/model.py b/samples/timm/caformer_b36.sail_in1k/model.py deleted file mode 100644 index 08d3de35cc..0000000000 --- a/samples/timm/caformer_b36.sail_in1k/model.py +++ /dev/null @@ -1,4758 +0,0 @@ -import torch - - -class GraphModule(torch.nn.Module): - def forward( - self, - L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - s1: torch.SymInt, - L_x_: torch.Tensor, - L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - ): - l_self_modules_stem_modules_conv_parameters_weight_ = ( - L_self_modules_stem_modules_conv_parameters_weight_ - ) - l_self_modules_stem_modules_conv_parameters_bias_ = ( - L_self_modules_stem_modules_conv_parameters_bias_ - ) - l_x_ = L_x_ - l_self_modules_stem_modules_norm_parameters_weight_ = ( - L_self_modules_stem_modules_norm_parameters_weight_ - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_head_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ - ) - x = torch.conv2d( - l_x_, - l_self_modules_stem_modules_conv_parameters_weight_, - l_self_modules_stem_modules_conv_parameters_bias_, - (4, 4), - (2, 2), - (1, 1), - 1, - ) - l_x_ = ( - l_self_modules_stem_modules_conv_parameters_weight_ - ) = l_self_modules_stem_modules_conv_parameters_bias_ = None - x_1 = x.permute(0, 2, 3, 1) - x = None - x_2 = torch.nn.functional.layer_norm( - x_1, - (128,), - l_self_modules_stem_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None - x_3 = x_2.permute(0, 3, 1, 2) - x_2 = None - sym_sum = torch.sym_sum([-3, s1]) - s1 = None - floordiv = sym_sum // 4 - sym_sum_1 = torch.sym_sum([1, floordiv]) - floordiv = sym_sum_1 = None - x_4 = x_3.permute(0, 2, 3, 1) - x_5 = torch.nn.functional.layer_norm( - x_4, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_6 = x_5.permute(0, 3, 1, 2) - x_5 = None - x_7 = torch.conv2d( - x_6, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu = torch.nn.functional.relu(x_7, inplace=False) - x_7 = None - pow_1 = relu**2 - relu = None - mul = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_1 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_1 - ) = None - x_8 = ( - mul - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_9 = torch.conv2d( - x_8, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 256, - ) - x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_10 = torch.conv2d( - x_9, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_11 = x_3 + x_10 - x_3 = x_10 = None - x_12 = x_11.permute(0, 2, 3, 1) - x_13 = torch.nn.functional.layer_norm( - x_12, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_14 = x_13.permute(0, 3, 1, 2) - x_13 = None - x_15 = torch.conv2d( - x_14, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_1 = torch.nn.functional.relu(x_15, inplace=False) - x_15 = None - pow_2 = relu_1**2 - relu_1 = None - mul_1 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_2 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_2 - ) = None - x_16 = ( - mul_1 - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) - x_16 = None - x_18 = torch.conv2d( - x_17, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) - x_18 = None - x_20 = x_11 + x_19 - x_11 = x_19 = None - x_21 = x_20.permute(0, 2, 3, 1) - x_22 = torch.nn.functional.layer_norm( - x_21, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_23 = x_22.permute(0, 3, 1, 2) - x_22 = None - x_24 = torch.conv2d( - x_23, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_2 = torch.nn.functional.relu(x_24, inplace=False) - x_24 = None - pow_3 = relu_2**2 - relu_2 = None - mul_2 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_3 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_3 - ) = None - x_25 = ( - mul_2 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_26 = torch.conv2d( - x_25, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 256, - ) - x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_27 = torch.conv2d( - x_26, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_28 = x_20 + x_27 - x_20 = x_27 = None - x_29 = x_28.permute(0, 2, 3, 1) - x_30 = torch.nn.functional.layer_norm( - x_29, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_31 = x_30.permute(0, 3, 1, 2) - x_30 = None - x_32 = torch.conv2d( - x_31, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_3 = torch.nn.functional.relu(x_32, inplace=False) - x_32 = None - pow_4 = relu_3**2 - relu_3 = None - mul_3 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_4 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_4 - ) = None - x_33 = ( - mul_3 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) - x_33 = None - x_35 = torch.conv2d( - x_34, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) - x_35 = None - x_37 = x_28 + x_36 - x_28 = x_36 = None - x_38 = x_37.permute(0, 2, 3, 1) - x_39 = torch.nn.functional.layer_norm( - x_38, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_40 = x_39.permute(0, 3, 1, 2) - x_39 = None - x_41 = torch.conv2d( - x_40, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_4 = torch.nn.functional.relu(x_41, inplace=False) - x_41 = None - pow_5 = relu_4**2 - relu_4 = None - mul_4 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_5 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_5 - ) = None - x_42 = ( - mul_4 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_43 = torch.conv2d( - x_42, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 256, - ) - x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_44 = torch.conv2d( - x_43, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_45 = x_37 + x_44 - x_37 = x_44 = None - x_46 = x_45.permute(0, 2, 3, 1) - x_47 = torch.nn.functional.layer_norm( - x_46, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_48 = x_47.permute(0, 3, 1, 2) - x_47 = None - x_49 = torch.conv2d( - x_48, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_5 = torch.nn.functional.relu(x_49, inplace=False) - x_49 = None - pow_6 = relu_5**2 - relu_5 = None - mul_5 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_6 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_6 - ) = None - x_50 = ( - mul_5 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) - x_50 = None - x_52 = torch.conv2d( - x_51, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) - x_52 = None - x_54 = x_45 + x_53 - x_45 = x_53 = None - x_55 = x_54.permute(0, 2, 3, 1) - x_54 = None - x_56 = torch.nn.functional.layer_norm( - x_55, - (128,), - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) - x_57 = x_56.permute(0, 3, 1, 2) - x_56 = None - x_58 = torch.conv2d( - x_57, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_1 = sym_sum // 8 - sym_sum_2 = torch.sym_sum([1, floordiv_1]) - floordiv_1 = sym_sum_2 = None - x_59 = x_58.permute(0, 2, 3, 1) - x_60 = torch.nn.functional.layer_norm( - x_59, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_61 = x_60.permute(0, 3, 1, 2) - x_60 = None - x_62 = torch.conv2d( - x_61, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_6 = torch.nn.functional.relu(x_62, inplace=False) - x_62 = None - pow_7 = relu_6**2 - relu_6 = None - mul_6 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_7 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_7 - ) = None - x_63 = ( - mul_6 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_64 = torch.conv2d( - x_63, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_65 = torch.conv2d( - x_64, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_66 = x_58 + x_65 - x_58 = x_65 = None - x_67 = x_66.permute(0, 2, 3, 1) - x_68 = torch.nn.functional.layer_norm( - x_67, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_69 = x_68.permute(0, 3, 1, 2) - x_68 = None - x_70 = torch.conv2d( - x_69, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_7 = torch.nn.functional.relu(x_70, inplace=False) - x_70 = None - pow_8 = relu_7**2 - relu_7 = None - mul_7 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_8 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_8 - ) = None - x_71 = ( - mul_7 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) - x_71 = None - x_73 = torch.conv2d( - x_72, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) - x_73 = None - x_75 = x_66 + x_74 - x_66 = x_74 = None - x_76 = x_75.permute(0, 2, 3, 1) - x_77 = torch.nn.functional.layer_norm( - x_76, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_78 = x_77.permute(0, 3, 1, 2) - x_77 = None - x_79 = torch.conv2d( - x_78, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_8 = torch.nn.functional.relu(x_79, inplace=False) - x_79 = None - pow_9 = relu_8**2 - relu_8 = None - mul_8 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_9 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_9 - ) = None - x_80 = ( - mul_8 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_81 = torch.conv2d( - x_80, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_82 = torch.conv2d( - x_81, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_83 = x_75 + x_82 - x_75 = x_82 = None - x_84 = x_83.permute(0, 2, 3, 1) - x_85 = torch.nn.functional.layer_norm( - x_84, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_86 = x_85.permute(0, 3, 1, 2) - x_85 = None - x_87 = torch.conv2d( - x_86, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_9 = torch.nn.functional.relu(x_87, inplace=False) - x_87 = None - pow_10 = relu_9**2 - relu_9 = None - mul_9 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_10 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_10 - ) = None - x_88 = ( - mul_9 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) - x_88 = None - x_90 = torch.conv2d( - x_89, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) - x_90 = None - x_92 = x_83 + x_91 - x_83 = x_91 = None - x_93 = x_92.permute(0, 2, 3, 1) - x_94 = torch.nn.functional.layer_norm( - x_93, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_95 = x_94.permute(0, 3, 1, 2) - x_94 = None - x_96 = torch.conv2d( - x_95, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_10 = torch.nn.functional.relu(x_96, inplace=False) - x_96 = None - pow_11 = relu_10**2 - relu_10 = None - mul_10 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_11 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_11 - ) = None - x_97 = ( - mul_10 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_98 = torch.conv2d( - x_97, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_99 = torch.conv2d( - x_98, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_100 = x_92 + x_99 - x_92 = x_99 = None - x_101 = x_100.permute(0, 2, 3, 1) - x_102 = torch.nn.functional.layer_norm( - x_101, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_103 = x_102.permute(0, 3, 1, 2) - x_102 = None - x_104 = torch.conv2d( - x_103, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_11 = torch.nn.functional.relu(x_104, inplace=False) - x_104 = None - pow_12 = relu_11**2 - relu_11 = None - mul_11 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_12 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_12 - ) = None - x_105 = ( - mul_11 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) - x_105 = None - x_107 = torch.conv2d( - x_106, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) - x_107 = None - x_109 = x_100 + x_108 - x_100 = x_108 = None - x_110 = x_109.permute(0, 2, 3, 1) - x_111 = torch.nn.functional.layer_norm( - x_110, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_110 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) - x_112 = x_111.permute(0, 3, 1, 2) - x_111 = None - x_113 = torch.conv2d( - x_112, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_112 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_12 = torch.nn.functional.relu(x_113, inplace=False) - x_113 = None - pow_13 = relu_12**2 - relu_12 = None - mul_12 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ - * pow_13 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_13 - ) = None - x_114 = ( - mul_12 - + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_12 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_115 = torch.conv2d( - x_114, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_114 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_116 = torch.conv2d( - x_115, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_115 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_117 = x_109 + x_116 - x_109 = x_116 = None - x_118 = x_117.permute(0, 2, 3, 1) - x_119 = torch.nn.functional.layer_norm( - x_118, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_118 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) - x_120 = x_119.permute(0, 3, 1, 2) - x_119 = None - x_121 = torch.conv2d( - x_120, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_120 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_13 = torch.nn.functional.relu(x_121, inplace=False) - x_121 = None - pow_14 = relu_13**2 - relu_13 = None - mul_13 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - * pow_14 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( - pow_14 - ) = None - x_122 = ( - mul_13 - + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - ) - mul_13 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) - x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) - x_122 = None - x_124 = torch.conv2d( - x_123, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_123 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) - x_124 = None - x_126 = x_117 + x_125 - x_117 = x_125 = None - x_127 = x_126.permute(0, 2, 3, 1) - x_128 = torch.nn.functional.layer_norm( - x_127, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_127 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) - x_129 = x_128.permute(0, 3, 1, 2) - x_128 = None - x_130 = torch.conv2d( - x_129, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_129 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_14 = torch.nn.functional.relu(x_130, inplace=False) - x_130 = None - pow_15 = relu_14**2 - relu_14 = None - mul_14 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ - * pow_15 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_15 - ) = None - x_131 = ( - mul_14 - + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_14 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_132 = torch.conv2d( - x_131, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_131 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_133 = torch.conv2d( - x_132, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_132 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_134 = x_126 + x_133 - x_126 = x_133 = None - x_135 = x_134.permute(0, 2, 3, 1) - x_136 = torch.nn.functional.layer_norm( - x_135, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_135 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) - x_137 = x_136.permute(0, 3, 1, 2) - x_136 = None - x_138 = torch.conv2d( - x_137, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_137 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_15 = torch.nn.functional.relu(x_138, inplace=False) - x_138 = None - pow_16 = relu_15**2 - relu_15 = None - mul_15 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - * pow_16 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( - pow_16 - ) = None - x_139 = ( - mul_15 - + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - ) - mul_15 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) - x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) - x_139 = None - x_141 = torch.conv2d( - x_140, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_140 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) - x_141 = None - x_143 = x_134 + x_142 - x_134 = x_142 = None - x_144 = x_143.permute(0, 2, 3, 1) - x_145 = torch.nn.functional.layer_norm( - x_144, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_144 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) - x_146 = x_145.permute(0, 3, 1, 2) - x_145 = None - x_147 = torch.conv2d( - x_146, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_146 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_16 = torch.nn.functional.relu(x_147, inplace=False) - x_147 = None - pow_17 = relu_16**2 - relu_16 = None - mul_16 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ - * pow_17 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_17 - ) = None - x_148 = ( - mul_16 - + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_16 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_149 = torch.conv2d( - x_148, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_148 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_150 = torch.conv2d( - x_149, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_149 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_151 = x_143 + x_150 - x_143 = x_150 = None - x_152 = x_151.permute(0, 2, 3, 1) - x_153 = torch.nn.functional.layer_norm( - x_152, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_152 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) - x_154 = x_153.permute(0, 3, 1, 2) - x_153 = None - x_155 = torch.conv2d( - x_154, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_154 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_17 = torch.nn.functional.relu(x_155, inplace=False) - x_155 = None - pow_18 = relu_17**2 - relu_17 = None - mul_17 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - * pow_18 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( - pow_18 - ) = None - x_156 = ( - mul_17 - + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - ) - mul_17 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) - x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) - x_156 = None - x_158 = torch.conv2d( - x_157, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_157 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) - x_158 = None - x_160 = x_151 + x_159 - x_151 = x_159 = None - x_161 = x_160.permute(0, 2, 3, 1) - x_162 = torch.nn.functional.layer_norm( - x_161, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_161 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) - x_163 = x_162.permute(0, 3, 1, 2) - x_162 = None - x_164 = torch.conv2d( - x_163, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_163 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_18 = torch.nn.functional.relu(x_164, inplace=False) - x_164 = None - pow_19 = relu_18**2 - relu_18 = None - mul_18 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ - * pow_19 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_19 - ) = None - x_165 = ( - mul_18 - + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_18 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_166 = torch.conv2d( - x_165, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_165 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_167 = torch.conv2d( - x_166, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_166 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_168 = x_160 + x_167 - x_160 = x_167 = None - x_169 = x_168.permute(0, 2, 3, 1) - x_170 = torch.nn.functional.layer_norm( - x_169, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_169 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) - x_171 = x_170.permute(0, 3, 1, 2) - x_170 = None - x_172 = torch.conv2d( - x_171, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_171 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_19 = torch.nn.functional.relu(x_172, inplace=False) - x_172 = None - pow_20 = relu_19**2 - relu_19 = None - mul_19 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - * pow_20 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( - pow_20 - ) = None - x_173 = ( - mul_19 - + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - ) - mul_19 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) - x_174 = torch.nn.functional.dropout(x_173, 0.0, False, False) - x_173 = None - x_175 = torch.conv2d( - x_174, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_174 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) - x_175 = None - x_177 = x_168 + x_176 - x_168 = x_176 = None - x_178 = x_177.permute(0, 2, 3, 1) - x_179 = torch.nn.functional.layer_norm( - x_178, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_178 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) - x_180 = x_179.permute(0, 3, 1, 2) - x_179 = None - x_181 = torch.conv2d( - x_180, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_180 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_20 = torch.nn.functional.relu(x_181, inplace=False) - x_181 = None - pow_21 = relu_20**2 - relu_20 = None - mul_20 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ - * pow_21 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_21 - ) = None - x_182 = ( - mul_20 - + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_20 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_183 = torch.conv2d( - x_182, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_182 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_184 = torch.conv2d( - x_183, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_183 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_185 = x_177 + x_184 - x_177 = x_184 = None - x_186 = x_185.permute(0, 2, 3, 1) - x_187 = torch.nn.functional.layer_norm( - x_186, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_186 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) - x_188 = x_187.permute(0, 3, 1, 2) - x_187 = None - x_189 = torch.conv2d( - x_188, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_188 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_21 = torch.nn.functional.relu(x_189, inplace=False) - x_189 = None - pow_22 = relu_21**2 - relu_21 = None - mul_21 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - * pow_22 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( - pow_22 - ) = None - x_190 = ( - mul_21 - + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - ) - mul_21 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) - x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) - x_190 = None - x_192 = torch.conv2d( - x_191, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_191 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) - x_192 = None - x_194 = x_185 + x_193 - x_185 = x_193 = None - x_195 = x_194.permute(0, 2, 3, 1) - x_196 = torch.nn.functional.layer_norm( - x_195, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_195 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) - x_197 = x_196.permute(0, 3, 1, 2) - x_196 = None - x_198 = torch.conv2d( - x_197, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_197 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_22 = torch.nn.functional.relu(x_198, inplace=False) - x_198 = None - pow_23 = relu_22**2 - relu_22 = None - mul_22 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ - * pow_23 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_23 - ) = None - x_199 = ( - mul_22 - + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_22 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_200 = torch.conv2d( - x_199, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_199 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_201 = torch.conv2d( - x_200, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_200 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_202 = x_194 + x_201 - x_194 = x_201 = None - x_203 = x_202.permute(0, 2, 3, 1) - x_204 = torch.nn.functional.layer_norm( - x_203, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_203 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) - x_205 = x_204.permute(0, 3, 1, 2) - x_204 = None - x_206 = torch.conv2d( - x_205, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_205 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_23 = torch.nn.functional.relu(x_206, inplace=False) - x_206 = None - pow_24 = relu_23**2 - relu_23 = None - mul_23 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - * pow_24 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( - pow_24 - ) = None - x_207 = ( - mul_23 - + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - ) - mul_23 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) - x_208 = torch.nn.functional.dropout(x_207, 0.0, False, False) - x_207 = None - x_209 = torch.conv2d( - x_208, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_208 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) - x_209 = None - x_211 = x_202 + x_210 - x_202 = x_210 = None - x_212 = x_211.permute(0, 2, 3, 1) - x_213 = torch.nn.functional.layer_norm( - x_212, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_212 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) - x_214 = x_213.permute(0, 3, 1, 2) - x_213 = None - x_215 = torch.conv2d( - x_214, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_214 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_24 = torch.nn.functional.relu(x_215, inplace=False) - x_215 = None - pow_25 = relu_24**2 - relu_24 = None - mul_24 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ - * pow_25 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_25 - ) = None - x_216 = ( - mul_24 - + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_24 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_217 = torch.conv2d( - x_216, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_216 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_218 = torch.conv2d( - x_217, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_217 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_219 = x_211 + x_218 - x_211 = x_218 = None - x_220 = x_219.permute(0, 2, 3, 1) - x_221 = torch.nn.functional.layer_norm( - x_220, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_220 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) - x_222 = x_221.permute(0, 3, 1, 2) - x_221 = None - x_223 = torch.conv2d( - x_222, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_222 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_25 = torch.nn.functional.relu(x_223, inplace=False) - x_223 = None - pow_26 = relu_25**2 - relu_25 = None - mul_25 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - * pow_26 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( - pow_26 - ) = None - x_224 = ( - mul_25 - + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - ) - mul_25 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) - x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) - x_224 = None - x_226 = torch.conv2d( - x_225, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_225 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) - x_226 = None - x_228 = x_219 + x_227 - x_219 = x_227 = None - x_229 = x_228.permute(0, 2, 3, 1) - x_230 = torch.nn.functional.layer_norm( - x_229, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_229 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) - x_231 = x_230.permute(0, 3, 1, 2) - x_230 = None - x_232 = torch.conv2d( - x_231, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_231 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_26 = torch.nn.functional.relu(x_232, inplace=False) - x_232 = None - pow_27 = relu_26**2 - relu_26 = None - mul_26 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ - * pow_27 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_27 - ) = None - x_233 = ( - mul_26 - + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_26 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_234 = torch.conv2d( - x_233, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_233 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_235 = torch.conv2d( - x_234, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_234 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_236 = x_228 + x_235 - x_228 = x_235 = None - x_237 = x_236.permute(0, 2, 3, 1) - x_238 = torch.nn.functional.layer_norm( - x_237, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_237 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) - x_239 = x_238.permute(0, 3, 1, 2) - x_238 = None - x_240 = torch.conv2d( - x_239, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_239 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_27 = torch.nn.functional.relu(x_240, inplace=False) - x_240 = None - pow_28 = relu_27**2 - relu_27 = None - mul_27 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - * pow_28 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( - pow_28 - ) = None - x_241 = ( - mul_27 - + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - ) - mul_27 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) - x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) - x_241 = None - x_243 = torch.conv2d( - x_242, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_242 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_244 = torch.nn.functional.dropout(x_243, 0.0, False, False) - x_243 = None - x_245 = x_236 + x_244 - x_236 = x_244 = None - x_246 = x_245.permute(0, 2, 3, 1) - x_247 = torch.nn.functional.layer_norm( - x_246, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_246 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) - x_248 = x_247.permute(0, 3, 1, 2) - x_247 = None - x_249 = torch.conv2d( - x_248, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_248 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_28 = torch.nn.functional.relu(x_249, inplace=False) - x_249 = None - pow_29 = relu_28**2 - relu_28 = None - mul_28 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ - * pow_29 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_29 - ) = None - x_250 = ( - mul_28 - + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_28 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_251 = torch.conv2d( - x_250, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_250 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_252 = torch.conv2d( - x_251, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_251 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_253 = x_245 + x_252 - x_245 = x_252 = None - x_254 = x_253.permute(0, 2, 3, 1) - x_255 = torch.nn.functional.layer_norm( - x_254, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_254 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) - x_256 = x_255.permute(0, 3, 1, 2) - x_255 = None - x_257 = torch.conv2d( - x_256, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_256 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_29 = torch.nn.functional.relu(x_257, inplace=False) - x_257 = None - pow_30 = relu_29**2 - relu_29 = None - mul_29 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - * pow_30 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( - pow_30 - ) = None - x_258 = ( - mul_29 - + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - ) - mul_29 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) - x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) - x_258 = None - x_260 = torch.conv2d( - x_259, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_259 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_261 = torch.nn.functional.dropout(x_260, 0.0, False, False) - x_260 = None - x_262 = x_253 + x_261 - x_253 = x_261 = None - x_263 = x_262.permute(0, 2, 3, 1) - x_262 = None - x_264 = torch.nn.functional.layer_norm( - x_263, - (256,), - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_263 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) - x_265 = x_264.permute(0, 3, 1, 2) - x_264 = None - x_266 = torch.conv2d( - x_265, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_265 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_2 = sym_sum // 16 - sym_sum_3 = torch.sym_sum([1, floordiv_2]) - floordiv_2 = None - reshape = x_266.reshape(1, 512, -1) - x_266 = None - x_267 = reshape.transpose(1, 2) - reshape = None - view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_30 = x_267 * view - view = None - layer_norm_33 = torch.nn.functional.layer_norm( - x_267, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_52 = sym_sum_3**2 - linear = torch._C._nn.linear( - layer_norm_33, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_33 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_1 = linear.reshape(1, pow_52, 3, 16, 32) - linear = None - qkv = reshape_1.permute(2, 0, 3, 1, 4) - reshape_1 = None - unbind = qkv.unbind(0) - qkv = None - q = unbind[0] - k = unbind[1] - v = unbind[2] - unbind = None - x_268 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) - q = k = v = None - transpose_1 = x_268.transpose(1, 2) - x_268 = None - x_269 = transpose_1.reshape(1, pow_52, 512) - transpose_1 = None - x_270 = torch._C._nn.linear( - x_269, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_269 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_271 = torch.nn.functional.dropout(x_270, 0.0, False, False) - x_270 = None - x_272 = mul_30 + x_271 - mul_30 = x_271 = None - view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_31 = x_272 * view_1 - view_1 = None - layer_norm_34 = torch.nn.functional.layer_norm( - x_272, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_272 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_273 = torch._C._nn.linear( - layer_norm_34, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_34 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_30 = torch.nn.functional.relu(x_273, inplace=False) - x_273 = None - pow_31 = relu_30**2 - relu_30 = None - mul_32 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_31 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_31 - ) = None - x_274 = ( - mul_32 - + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_275 = torch.nn.functional.dropout(x_274, 0.0, False, False) - x_274 = None - x_276 = torch._C._nn.linear( - x_275, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_277 = torch.nn.functional.dropout(x_276, 0.0, False, False) - x_276 = None - x_278 = mul_31 + x_277 - mul_31 = x_277 = None - view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_33 = x_278 * view_2 - view_2 = None - layer_norm_35 = torch.nn.functional.layer_norm( - x_278, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_4 = torch._C._nn.linear( - layer_norm_35, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_3 = linear_4.reshape(1, pow_52, 3, 16, 32) - linear_4 = None - qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) - reshape_3 = None - unbind_1 = qkv_1.unbind(0) - qkv_1 = None - q_1 = unbind_1[0] - k_1 = unbind_1[1] - v_1 = unbind_1[2] - unbind_1 = None - x_279 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) - q_1 = k_1 = v_1 = None - transpose_2 = x_279.transpose(1, 2) - x_279 = None - x_280 = transpose_2.reshape(1, pow_52, 512) - transpose_2 = None - x_281 = torch._C._nn.linear( - x_280, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_280 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) - x_281 = None - x_283 = mul_33 + x_282 - mul_33 = x_282 = None - view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_34 = x_283 * view_3 - view_3 = None - layer_norm_36 = torch.nn.functional.layer_norm( - x_283, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_283 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_284 = torch._C._nn.linear( - layer_norm_36, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_36 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_31 = torch.nn.functional.relu(x_284, inplace=False) - x_284 = None - pow_32 = relu_31**2 - relu_31 = None - mul_35 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_32 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_32 - ) = None - x_285 = ( - mul_35 - + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_286 = torch.nn.functional.dropout(x_285, 0.0, False, False) - x_285 = None - x_287 = torch._C._nn.linear( - x_286, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) - x_287 = None - x_289 = mul_34 + x_288 - mul_34 = x_288 = None - view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_36 = x_289 * view_4 - view_4 = None - layer_norm_37 = torch.nn.functional.layer_norm( - x_289, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_8 = torch._C._nn.linear( - layer_norm_37, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_37 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_5 = linear_8.reshape(1, pow_52, 3, 16, 32) - linear_8 = None - qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) - reshape_5 = None - unbind_2 = qkv_2.unbind(0) - qkv_2 = None - q_2 = unbind_2[0] - k_2 = unbind_2[1] - v_2 = unbind_2[2] - unbind_2 = None - x_290 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) - q_2 = k_2 = v_2 = None - transpose_3 = x_290.transpose(1, 2) - x_290 = None - x_291 = transpose_3.reshape(1, pow_52, 512) - transpose_3 = None - x_292 = torch._C._nn.linear( - x_291, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_291 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_293 = torch.nn.functional.dropout(x_292, 0.0, False, False) - x_292 = None - x_294 = mul_36 + x_293 - mul_36 = x_293 = None - view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_37 = x_294 * view_5 - view_5 = None - layer_norm_38 = torch.nn.functional.layer_norm( - x_294, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_294 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_295 = torch._C._nn.linear( - layer_norm_38, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_32 = torch.nn.functional.relu(x_295, inplace=False) - x_295 = None - pow_33 = relu_32**2 - relu_32 = None - mul_38 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_33 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_33 - ) = None - x_296 = ( - mul_38 - + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_297 = torch.nn.functional.dropout(x_296, 0.0, False, False) - x_296 = None - x_298 = torch._C._nn.linear( - x_297, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_299 = torch.nn.functional.dropout(x_298, 0.0, False, False) - x_298 = None - x_300 = mul_37 + x_299 - mul_37 = x_299 = None - view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_39 = x_300 * view_6 - view_6 = None - layer_norm_39 = torch.nn.functional.layer_norm( - x_300, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) - linear_12 = torch._C._nn.linear( - layer_norm_39, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_39 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_7 = linear_12.reshape(1, pow_52, 3, 16, 32) - linear_12 = None - qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) - reshape_7 = None - unbind_3 = qkv_3.unbind(0) - qkv_3 = None - q_3 = unbind_3[0] - k_3 = unbind_3[1] - v_3 = unbind_3[2] - unbind_3 = None - x_301 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) - q_3 = k_3 = v_3 = None - transpose_4 = x_301.transpose(1, 2) - x_301 = None - x_302 = transpose_4.reshape(1, pow_52, 512) - transpose_4 = None - x_303 = torch._C._nn.linear( - x_302, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_302 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_304 = torch.nn.functional.dropout(x_303, 0.0, False, False) - x_303 = None - x_305 = mul_39 + x_304 - mul_39 = x_304 = None - view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_40 = x_305 * view_7 - view_7 = None - layer_norm_40 = torch.nn.functional.layer_norm( - x_305, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_305 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) - x_306 = torch._C._nn.linear( - layer_norm_40, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_40 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_33 = torch.nn.functional.relu(x_306, inplace=False) - x_306 = None - pow_34 = relu_33**2 - relu_33 = None - mul_41 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - * pow_34 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( - pow_34 - ) = None - x_307 = ( - mul_41 - + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - ) - mul_41 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) - x_308 = torch.nn.functional.dropout(x_307, 0.0, False, False) - x_307 = None - x_309 = torch._C._nn.linear( - x_308, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) - x_309 = None - x_311 = mul_40 + x_310 - mul_40 = x_310 = None - view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_42 = x_311 * view_8 - view_8 = None - layer_norm_41 = torch.nn.functional.layer_norm( - x_311, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) - linear_16 = torch._C._nn.linear( - layer_norm_41, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_41 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_9 = linear_16.reshape(1, pow_52, 3, 16, 32) - linear_16 = None - qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) - reshape_9 = None - unbind_4 = qkv_4.unbind(0) - qkv_4 = None - q_4 = unbind_4[0] - k_4 = unbind_4[1] - v_4 = unbind_4[2] - unbind_4 = None - x_312 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) - q_4 = k_4 = v_4 = None - transpose_5 = x_312.transpose(1, 2) - x_312 = None - x_313 = transpose_5.reshape(1, pow_52, 512) - transpose_5 = None - x_314 = torch._C._nn.linear( - x_313, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_313 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) - x_314 = None - x_316 = mul_42 + x_315 - mul_42 = x_315 = None - view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_43 = x_316 * view_9 - view_9 = None - layer_norm_42 = torch.nn.functional.layer_norm( - x_316, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_316 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) - x_317 = torch._C._nn.linear( - layer_norm_42, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_42 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_34 = torch.nn.functional.relu(x_317, inplace=False) - x_317 = None - pow_35 = relu_34**2 - relu_34 = None - mul_44 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - * pow_35 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( - pow_35 - ) = None - x_318 = ( - mul_44 - + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - ) - mul_44 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) - x_319 = torch.nn.functional.dropout(x_318, 0.0, False, False) - x_318 = None - x_320 = torch._C._nn.linear( - x_319, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_321 = torch.nn.functional.dropout(x_320, 0.0, False, False) - x_320 = None - x_322 = mul_43 + x_321 - mul_43 = x_321 = None - view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_45 = x_322 * view_10 - view_10 = None - layer_norm_43 = torch.nn.functional.layer_norm( - x_322, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) - linear_20 = torch._C._nn.linear( - layer_norm_43, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_43 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_11 = linear_20.reshape(1, pow_52, 3, 16, 32) - linear_20 = None - qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) - reshape_11 = None - unbind_5 = qkv_5.unbind(0) - qkv_5 = None - q_5 = unbind_5[0] - k_5 = unbind_5[1] - v_5 = unbind_5[2] - unbind_5 = None - x_323 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) - q_5 = k_5 = v_5 = None - transpose_6 = x_323.transpose(1, 2) - x_323 = None - x_324 = transpose_6.reshape(1, pow_52, 512) - transpose_6 = None - x_325 = torch._C._nn.linear( - x_324, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_324 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_326 = torch.nn.functional.dropout(x_325, 0.0, False, False) - x_325 = None - x_327 = mul_45 + x_326 - mul_45 = x_326 = None - view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_46 = x_327 * view_11 - view_11 = None - layer_norm_44 = torch.nn.functional.layer_norm( - x_327, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_327 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) - x_328 = torch._C._nn.linear( - layer_norm_44, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_44 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_35 = torch.nn.functional.relu(x_328, inplace=False) - x_328 = None - pow_36 = relu_35**2 - relu_35 = None - mul_47 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - * pow_36 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( - pow_36 - ) = None - x_329 = ( - mul_47 - + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - ) - mul_47 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) - x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) - x_329 = None - x_331 = torch._C._nn.linear( - x_330, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_332 = torch.nn.functional.dropout(x_331, 0.0, False, False) - x_331 = None - x_333 = mul_46 + x_332 - mul_46 = x_332 = None - view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_48 = x_333 * view_12 - view_12 = None - layer_norm_45 = torch.nn.functional.layer_norm( - x_333, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) - linear_24 = torch._C._nn.linear( - layer_norm_45, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_45 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_13 = linear_24.reshape(1, pow_52, 3, 16, 32) - linear_24 = None - qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) - reshape_13 = None - unbind_6 = qkv_6.unbind(0) - qkv_6 = None - q_6 = unbind_6[0] - k_6 = unbind_6[1] - v_6 = unbind_6[2] - unbind_6 = None - x_334 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) - q_6 = k_6 = v_6 = None - transpose_7 = x_334.transpose(1, 2) - x_334 = None - x_335 = transpose_7.reshape(1, pow_52, 512) - transpose_7 = None - x_336 = torch._C._nn.linear( - x_335, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_335 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_337 = torch.nn.functional.dropout(x_336, 0.0, False, False) - x_336 = None - x_338 = mul_48 + x_337 - mul_48 = x_337 = None - view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_49 = x_338 * view_13 - view_13 = None - layer_norm_46 = torch.nn.functional.layer_norm( - x_338, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_338 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) - x_339 = torch._C._nn.linear( - layer_norm_46, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_46 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_36 = torch.nn.functional.relu(x_339, inplace=False) - x_339 = None - pow_37 = relu_36**2 - relu_36 = None - mul_50 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - * pow_37 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( - pow_37 - ) = None - x_340 = ( - mul_50 - + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - ) - mul_50 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) - x_341 = torch.nn.functional.dropout(x_340, 0.0, False, False) - x_340 = None - x_342 = torch._C._nn.linear( - x_341, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_343 = torch.nn.functional.dropout(x_342, 0.0, False, False) - x_342 = None - x_344 = mul_49 + x_343 - mul_49 = x_343 = None - view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_51 = x_344 * view_14 - view_14 = None - layer_norm_47 = torch.nn.functional.layer_norm( - x_344, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) - linear_28 = torch._C._nn.linear( - layer_norm_47, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_47 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_15 = linear_28.reshape(1, pow_52, 3, 16, 32) - linear_28 = None - qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) - reshape_15 = None - unbind_7 = qkv_7.unbind(0) - qkv_7 = None - q_7 = unbind_7[0] - k_7 = unbind_7[1] - v_7 = unbind_7[2] - unbind_7 = None - x_345 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) - q_7 = k_7 = v_7 = None - transpose_8 = x_345.transpose(1, 2) - x_345 = None - x_346 = transpose_8.reshape(1, pow_52, 512) - transpose_8 = None - x_347 = torch._C._nn.linear( - x_346, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_346 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) - x_347 = None - x_349 = mul_51 + x_348 - mul_51 = x_348 = None - view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_52 = x_349 * view_15 - view_15 = None - layer_norm_48 = torch.nn.functional.layer_norm( - x_349, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_349 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) - x_350 = torch._C._nn.linear( - layer_norm_48, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_48 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_37 = torch.nn.functional.relu(x_350, inplace=False) - x_350 = None - pow_38 = relu_37**2 - relu_37 = None - mul_53 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - * pow_38 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( - pow_38 - ) = None - x_351 = ( - mul_53 - + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - ) - mul_53 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) - x_352 = torch.nn.functional.dropout(x_351, 0.0, False, False) - x_351 = None - x_353 = torch._C._nn.linear( - x_352, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_354 = torch.nn.functional.dropout(x_353, 0.0, False, False) - x_353 = None - x_355 = mul_52 + x_354 - mul_52 = x_354 = None - view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_54 = x_355 * view_16 - view_16 = None - layer_norm_49 = torch.nn.functional.layer_norm( - x_355, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) - linear_32 = torch._C._nn.linear( - layer_norm_49, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_49 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_17 = linear_32.reshape(1, pow_52, 3, 16, 32) - linear_32 = None - qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) - reshape_17 = None - unbind_8 = qkv_8.unbind(0) - qkv_8 = None - q_8 = unbind_8[0] - k_8 = unbind_8[1] - v_8 = unbind_8[2] - unbind_8 = None - x_356 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) - q_8 = k_8 = v_8 = None - transpose_9 = x_356.transpose(1, 2) - x_356 = None - x_357 = transpose_9.reshape(1, pow_52, 512) - transpose_9 = None - x_358 = torch._C._nn.linear( - x_357, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_357 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_359 = torch.nn.functional.dropout(x_358, 0.0, False, False) - x_358 = None - x_360 = mul_54 + x_359 - mul_54 = x_359 = None - view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_55 = x_360 * view_17 - view_17 = None - layer_norm_50 = torch.nn.functional.layer_norm( - x_360, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_360 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) - x_361 = torch._C._nn.linear( - layer_norm_50, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_50 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_38 = torch.nn.functional.relu(x_361, inplace=False) - x_361 = None - pow_39 = relu_38**2 - relu_38 = None - mul_56 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - * pow_39 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( - pow_39 - ) = None - x_362 = ( - mul_56 - + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - ) - mul_56 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) - x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) - x_362 = None - x_364 = torch._C._nn.linear( - x_363, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_365 = torch.nn.functional.dropout(x_364, 0.0, False, False) - x_364 = None - x_366 = mul_55 + x_365 - mul_55 = x_365 = None - view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_57 = x_366 * view_18 - view_18 = None - layer_norm_51 = torch.nn.functional.layer_norm( - x_366, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) - linear_36 = torch._C._nn.linear( - layer_norm_51, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_51 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_19 = linear_36.reshape(1, pow_52, 3, 16, 32) - linear_36 = None - qkv_9 = reshape_19.permute(2, 0, 3, 1, 4) - reshape_19 = None - unbind_9 = qkv_9.unbind(0) - qkv_9 = None - q_9 = unbind_9[0] - k_9 = unbind_9[1] - v_9 = unbind_9[2] - unbind_9 = None - x_367 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) - q_9 = k_9 = v_9 = None - transpose_10 = x_367.transpose(1, 2) - x_367 = None - x_368 = transpose_10.reshape(1, pow_52, 512) - transpose_10 = None - x_369 = torch._C._nn.linear( - x_368, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_368 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_370 = torch.nn.functional.dropout(x_369, 0.0, False, False) - x_369 = None - x_371 = mul_57 + x_370 - mul_57 = x_370 = None - view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_58 = x_371 * view_19 - view_19 = None - layer_norm_52 = torch.nn.functional.layer_norm( - x_371, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_371 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) - x_372 = torch._C._nn.linear( - layer_norm_52, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_52 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_39 = torch.nn.functional.relu(x_372, inplace=False) - x_372 = None - pow_40 = relu_39**2 - relu_39 = None - mul_59 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - * pow_40 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( - pow_40 - ) = None - x_373 = ( - mul_59 - + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - ) - mul_59 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) - x_374 = torch.nn.functional.dropout(x_373, 0.0, False, False) - x_373 = None - x_375 = torch._C._nn.linear( - x_374, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_374 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_376 = torch.nn.functional.dropout(x_375, 0.0, False, False) - x_375 = None - x_377 = mul_58 + x_376 - mul_58 = x_376 = None - view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_60 = x_377 * view_20 - view_20 = None - layer_norm_53 = torch.nn.functional.layer_norm( - x_377, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_377 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) - linear_40 = torch._C._nn.linear( - layer_norm_53, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_53 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_21 = linear_40.reshape(1, pow_52, 3, 16, 32) - linear_40 = None - qkv_10 = reshape_21.permute(2, 0, 3, 1, 4) - reshape_21 = None - unbind_10 = qkv_10.unbind(0) - qkv_10 = None - q_10 = unbind_10[0] - k_10 = unbind_10[1] - v_10 = unbind_10[2] - unbind_10 = None - x_378 = torch._C._nn.scaled_dot_product_attention( - q_10, k_10, v_10, dropout_p=0.0 - ) - q_10 = k_10 = v_10 = None - transpose_11 = x_378.transpose(1, 2) - x_378 = None - x_379 = transpose_11.reshape(1, pow_52, 512) - transpose_11 = None - x_380 = torch._C._nn.linear( - x_379, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_379 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_381 = torch.nn.functional.dropout(x_380, 0.0, False, False) - x_380 = None - x_382 = mul_60 + x_381 - mul_60 = x_381 = None - view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_61 = x_382 * view_21 - view_21 = None - layer_norm_54 = torch.nn.functional.layer_norm( - x_382, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_382 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) - x_383 = torch._C._nn.linear( - layer_norm_54, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_54 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_40 = torch.nn.functional.relu(x_383, inplace=False) - x_383 = None - pow_41 = relu_40**2 - relu_40 = None - mul_62 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - * pow_41 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( - pow_41 - ) = None - x_384 = ( - mul_62 - + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - ) - mul_62 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) - x_385 = torch.nn.functional.dropout(x_384, 0.0, False, False) - x_384 = None - x_386 = torch._C._nn.linear( - x_385, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_385 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_387 = torch.nn.functional.dropout(x_386, 0.0, False, False) - x_386 = None - x_388 = mul_61 + x_387 - mul_61 = x_387 = None - view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_63 = x_388 * view_22 - view_22 = None - layer_norm_55 = torch.nn.functional.layer_norm( - x_388, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_388 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) - linear_44 = torch._C._nn.linear( - layer_norm_55, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_55 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_23 = linear_44.reshape(1, pow_52, 3, 16, 32) - linear_44 = None - qkv_11 = reshape_23.permute(2, 0, 3, 1, 4) - reshape_23 = None - unbind_11 = qkv_11.unbind(0) - qkv_11 = None - q_11 = unbind_11[0] - k_11 = unbind_11[1] - v_11 = unbind_11[2] - unbind_11 = None - x_389 = torch._C._nn.scaled_dot_product_attention( - q_11, k_11, v_11, dropout_p=0.0 - ) - q_11 = k_11 = v_11 = None - transpose_12 = x_389.transpose(1, 2) - x_389 = None - x_390 = transpose_12.reshape(1, pow_52, 512) - transpose_12 = None - x_391 = torch._C._nn.linear( - x_390, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_390 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_392 = torch.nn.functional.dropout(x_391, 0.0, False, False) - x_391 = None - x_393 = mul_63 + x_392 - mul_63 = x_392 = None - view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_64 = x_393 * view_23 - view_23 = None - layer_norm_56 = torch.nn.functional.layer_norm( - x_393, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_393 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) - x_394 = torch._C._nn.linear( - layer_norm_56, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_56 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_41 = torch.nn.functional.relu(x_394, inplace=False) - x_394 = None - pow_42 = relu_41**2 - relu_41 = None - mul_65 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - * pow_42 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( - pow_42 - ) = None - x_395 = ( - mul_65 - + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - ) - mul_65 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) - x_396 = torch.nn.functional.dropout(x_395, 0.0, False, False) - x_395 = None - x_397 = torch._C._nn.linear( - x_396, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_396 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_398 = torch.nn.functional.dropout(x_397, 0.0, False, False) - x_397 = None - x_399 = mul_64 + x_398 - mul_64 = x_398 = None - view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_66 = x_399 * view_24 - view_24 = None - layer_norm_57 = torch.nn.functional.layer_norm( - x_399, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_399 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = (None) - linear_48 = torch._C._nn.linear( - layer_norm_57, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_57 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_25 = linear_48.reshape(1, pow_52, 3, 16, 32) - linear_48 = None - qkv_12 = reshape_25.permute(2, 0, 3, 1, 4) - reshape_25 = None - unbind_12 = qkv_12.unbind(0) - qkv_12 = None - q_12 = unbind_12[0] - k_12 = unbind_12[1] - v_12 = unbind_12[2] - unbind_12 = None - x_400 = torch._C._nn.scaled_dot_product_attention( - q_12, k_12, v_12, dropout_p=0.0 - ) - q_12 = k_12 = v_12 = None - transpose_13 = x_400.transpose(1, 2) - x_400 = None - x_401 = transpose_13.reshape(1, pow_52, 512) - transpose_13 = None - x_402 = torch._C._nn.linear( - x_401, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_401 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_403 = torch.nn.functional.dropout(x_402, 0.0, False, False) - x_402 = None - x_404 = mul_66 + x_403 - mul_66 = x_403 = None - view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_67 = x_404 * view_25 - view_25 = None - layer_norm_58 = torch.nn.functional.layer_norm( - x_404, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_404 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = (None) - x_405 = torch._C._nn.linear( - layer_norm_58, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_58 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_42 = torch.nn.functional.relu(x_405, inplace=False) - x_405 = None - pow_43 = relu_42**2 - relu_42 = None - mul_68 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ - * pow_43 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = ( - pow_43 - ) = None - x_406 = ( - mul_68 - + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ - ) - mul_68 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = (None) - x_407 = torch.nn.functional.dropout(x_406, 0.0, False, False) - x_406 = None - x_408 = torch._C._nn.linear( - x_407, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_407 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_409 = torch.nn.functional.dropout(x_408, 0.0, False, False) - x_408 = None - x_410 = mul_67 + x_409 - mul_67 = x_409 = None - view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_69 = x_410 * view_26 - view_26 = None - layer_norm_59 = torch.nn.functional.layer_norm( - x_410, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_410 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = (None) - linear_52 = torch._C._nn.linear( - layer_norm_59, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_59 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_27 = linear_52.reshape(1, pow_52, 3, 16, 32) - linear_52 = None - qkv_13 = reshape_27.permute(2, 0, 3, 1, 4) - reshape_27 = None - unbind_13 = qkv_13.unbind(0) - qkv_13 = None - q_13 = unbind_13[0] - k_13 = unbind_13[1] - v_13 = unbind_13[2] - unbind_13 = None - x_411 = torch._C._nn.scaled_dot_product_attention( - q_13, k_13, v_13, dropout_p=0.0 - ) - q_13 = k_13 = v_13 = None - transpose_14 = x_411.transpose(1, 2) - x_411 = None - x_412 = transpose_14.reshape(1, pow_52, 512) - transpose_14 = None - x_413 = torch._C._nn.linear( - x_412, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_412 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_414 = torch.nn.functional.dropout(x_413, 0.0, False, False) - x_413 = None - x_415 = mul_69 + x_414 - mul_69 = x_414 = None - view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_70 = x_415 * view_27 - view_27 = None - layer_norm_60 = torch.nn.functional.layer_norm( - x_415, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_415 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = (None) - x_416 = torch._C._nn.linear( - layer_norm_60, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_60 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_43 = torch.nn.functional.relu(x_416, inplace=False) - x_416 = None - pow_44 = relu_43**2 - relu_43 = None - mul_71 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ - * pow_44 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = ( - pow_44 - ) = None - x_417 = ( - mul_71 - + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ - ) - mul_71 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = (None) - x_418 = torch.nn.functional.dropout(x_417, 0.0, False, False) - x_417 = None - x_419 = torch._C._nn.linear( - x_418, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_418 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_420 = torch.nn.functional.dropout(x_419, 0.0, False, False) - x_419 = None - x_421 = mul_70 + x_420 - mul_70 = x_420 = None - view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_72 = x_421 * view_28 - view_28 = None - layer_norm_61 = torch.nn.functional.layer_norm( - x_421, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_421 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = (None) - linear_56 = torch._C._nn.linear( - layer_norm_61, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_61 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_29 = linear_56.reshape(1, pow_52, 3, 16, 32) - linear_56 = None - qkv_14 = reshape_29.permute(2, 0, 3, 1, 4) - reshape_29 = None - unbind_14 = qkv_14.unbind(0) - qkv_14 = None - q_14 = unbind_14[0] - k_14 = unbind_14[1] - v_14 = unbind_14[2] - unbind_14 = None - x_422 = torch._C._nn.scaled_dot_product_attention( - q_14, k_14, v_14, dropout_p=0.0 - ) - q_14 = k_14 = v_14 = None - transpose_15 = x_422.transpose(1, 2) - x_422 = None - x_423 = transpose_15.reshape(1, pow_52, 512) - transpose_15 = None - x_424 = torch._C._nn.linear( - x_423, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_423 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_425 = torch.nn.functional.dropout(x_424, 0.0, False, False) - x_424 = None - x_426 = mul_72 + x_425 - mul_72 = x_425 = None - view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_73 = x_426 * view_29 - view_29 = None - layer_norm_62 = torch.nn.functional.layer_norm( - x_426, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_426 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = (None) - x_427 = torch._C._nn.linear( - layer_norm_62, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_62 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_44 = torch.nn.functional.relu(x_427, inplace=False) - x_427 = None - pow_45 = relu_44**2 - relu_44 = None - mul_74 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ - * pow_45 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = ( - pow_45 - ) = None - x_428 = ( - mul_74 - + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ - ) - mul_74 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = (None) - x_429 = torch.nn.functional.dropout(x_428, 0.0, False, False) - x_428 = None - x_430 = torch._C._nn.linear( - x_429, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_429 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_431 = torch.nn.functional.dropout(x_430, 0.0, False, False) - x_430 = None - x_432 = mul_73 + x_431 - mul_73 = x_431 = None - view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_75 = x_432 * view_30 - view_30 = None - layer_norm_63 = torch.nn.functional.layer_norm( - x_432, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_432 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = (None) - linear_60 = torch._C._nn.linear( - layer_norm_63, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_63 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_31 = linear_60.reshape(1, pow_52, 3, 16, 32) - linear_60 = None - qkv_15 = reshape_31.permute(2, 0, 3, 1, 4) - reshape_31 = None - unbind_15 = qkv_15.unbind(0) - qkv_15 = None - q_15 = unbind_15[0] - k_15 = unbind_15[1] - v_15 = unbind_15[2] - unbind_15 = None - x_433 = torch._C._nn.scaled_dot_product_attention( - q_15, k_15, v_15, dropout_p=0.0 - ) - q_15 = k_15 = v_15 = None - transpose_16 = x_433.transpose(1, 2) - x_433 = None - x_434 = transpose_16.reshape(1, pow_52, 512) - transpose_16 = None - x_435 = torch._C._nn.linear( - x_434, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_434 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_436 = torch.nn.functional.dropout(x_435, 0.0, False, False) - x_435 = None - x_437 = mul_75 + x_436 - mul_75 = x_436 = None - view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_76 = x_437 * view_31 - view_31 = None - layer_norm_64 = torch.nn.functional.layer_norm( - x_437, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_437 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = (None) - x_438 = torch._C._nn.linear( - layer_norm_64, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_64 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_45 = torch.nn.functional.relu(x_438, inplace=False) - x_438 = None - pow_46 = relu_45**2 - relu_45 = None - mul_77 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ - * pow_46 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = ( - pow_46 - ) = None - x_439 = ( - mul_77 - + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ - ) - mul_77 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = (None) - x_440 = torch.nn.functional.dropout(x_439, 0.0, False, False) - x_439 = None - x_441 = torch._C._nn.linear( - x_440, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_440 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_442 = torch.nn.functional.dropout(x_441, 0.0, False, False) - x_441 = None - x_443 = mul_76 + x_442 - mul_76 = x_442 = None - view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_78 = x_443 * view_32 - view_32 = None - layer_norm_65 = torch.nn.functional.layer_norm( - x_443, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_443 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = (None) - linear_64 = torch._C._nn.linear( - layer_norm_65, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_65 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_33 = linear_64.reshape(1, pow_52, 3, 16, 32) - linear_64 = None - qkv_16 = reshape_33.permute(2, 0, 3, 1, 4) - reshape_33 = None - unbind_16 = qkv_16.unbind(0) - qkv_16 = None - q_16 = unbind_16[0] - k_16 = unbind_16[1] - v_16 = unbind_16[2] - unbind_16 = None - x_444 = torch._C._nn.scaled_dot_product_attention( - q_16, k_16, v_16, dropout_p=0.0 - ) - q_16 = k_16 = v_16 = None - transpose_17 = x_444.transpose(1, 2) - x_444 = None - x_445 = transpose_17.reshape(1, pow_52, 512) - transpose_17 = None - x_446 = torch._C._nn.linear( - x_445, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_445 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_447 = torch.nn.functional.dropout(x_446, 0.0, False, False) - x_446 = None - x_448 = mul_78 + x_447 - mul_78 = x_447 = None - view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_79 = x_448 * view_33 - view_33 = None - layer_norm_66 = torch.nn.functional.layer_norm( - x_448, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_448 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = (None) - x_449 = torch._C._nn.linear( - layer_norm_66, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_66 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_46 = torch.nn.functional.relu(x_449, inplace=False) - x_449 = None - pow_47 = relu_46**2 - relu_46 = None - mul_80 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ - * pow_47 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = ( - pow_47 - ) = None - x_450 = ( - mul_80 - + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ - ) - mul_80 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = (None) - x_451 = torch.nn.functional.dropout(x_450, 0.0, False, False) - x_450 = None - x_452 = torch._C._nn.linear( - x_451, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_451 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_453 = torch.nn.functional.dropout(x_452, 0.0, False, False) - x_452 = None - x_454 = mul_79 + x_453 - mul_79 = x_453 = None - view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_81 = x_454 * view_34 - view_34 = None - layer_norm_67 = torch.nn.functional.layer_norm( - x_454, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_454 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = (None) - linear_68 = torch._C._nn.linear( - layer_norm_67, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_67 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_35 = linear_68.reshape(1, pow_52, 3, 16, 32) - linear_68 = None - qkv_17 = reshape_35.permute(2, 0, 3, 1, 4) - reshape_35 = None - unbind_17 = qkv_17.unbind(0) - qkv_17 = None - q_17 = unbind_17[0] - k_17 = unbind_17[1] - v_17 = unbind_17[2] - unbind_17 = None - x_455 = torch._C._nn.scaled_dot_product_attention( - q_17, k_17, v_17, dropout_p=0.0 - ) - q_17 = k_17 = v_17 = None - transpose_18 = x_455.transpose(1, 2) - x_455 = None - x_456 = transpose_18.reshape(1, pow_52, 512) - transpose_18 = pow_52 = None - x_457 = torch._C._nn.linear( - x_456, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_456 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_458 = torch.nn.functional.dropout(x_457, 0.0, False, False) - x_457 = None - x_459 = mul_81 + x_458 - mul_81 = x_458 = None - view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_82 = x_459 * view_35 - view_35 = None - layer_norm_68 = torch.nn.functional.layer_norm( - x_459, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_459 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = (None) - x_460 = torch._C._nn.linear( - layer_norm_68, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_68 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_47 = torch.nn.functional.relu(x_460, inplace=False) - x_460 = None - pow_48 = relu_47**2 - relu_47 = None - mul_83 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ - * pow_48 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = ( - pow_48 - ) = None - x_461 = ( - mul_83 - + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ - ) - mul_83 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = (None) - x_462 = torch.nn.functional.dropout(x_461, 0.0, False, False) - x_461 = None - x_463 = torch._C._nn.linear( - x_462, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_462 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_464 = torch.nn.functional.dropout(x_463, 0.0, False, False) - x_463 = None - x_465 = mul_82 + x_464 - mul_82 = x_464 = None - transpose_19 = x_465.transpose(1, 2) - x_465 = None - x_466 = transpose_19.reshape(1, 512, sym_sum_3, sym_sum_3) - transpose_19 = sym_sum_3 = None - x_467 = x_466.permute(0, 2, 3, 1) - x_466 = None - x_468 = torch.nn.functional.layer_norm( - x_467, - (512,), - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_467 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) - x_469 = x_468.permute(0, 3, 1, 2) - x_468 = None - x_470 = torch.conv2d( - x_469, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_469 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_3 = sym_sum // 32 - sym_sum = None - sym_sum_4 = torch.sym_sum([1, floordiv_3]) - floordiv_3 = None - reshape_38 = x_470.reshape(1, 768, -1) - x_470 = None - x_471 = reshape_38.transpose(1, 2) - reshape_38 = None - view_36 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_84 = x_471 * view_36 - view_36 = None - layer_norm_70 = torch.nn.functional.layer_norm( - x_471, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_471 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_53 = sym_sum_4**2 - linear_72 = torch._C._nn.linear( - layer_norm_70, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_70 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_39 = linear_72.reshape(1, pow_53, 3, 24, 32) - linear_72 = None - qkv_18 = reshape_39.permute(2, 0, 3, 1, 4) - reshape_39 = None - unbind_18 = qkv_18.unbind(0) - qkv_18 = None - q_18 = unbind_18[0] - k_18 = unbind_18[1] - v_18 = unbind_18[2] - unbind_18 = None - x_472 = torch._C._nn.scaled_dot_product_attention( - q_18, k_18, v_18, dropout_p=0.0 - ) - q_18 = k_18 = v_18 = None - transpose_21 = x_472.transpose(1, 2) - x_472 = None - x_473 = transpose_21.reshape(1, pow_53, 768) - transpose_21 = None - x_474 = torch._C._nn.linear( - x_473, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_473 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_475 = torch.nn.functional.dropout(x_474, 0.0, False, False) - x_474 = None - x_476 = mul_84 + x_475 - mul_84 = x_475 = None - view_37 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_85 = x_476 * view_37 - view_37 = None - layer_norm_71 = torch.nn.functional.layer_norm( - x_476, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_476 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_477 = torch._C._nn.linear( - layer_norm_71, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_71 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_48 = torch.nn.functional.relu(x_477, inplace=False) - x_477 = None - pow_49 = relu_48**2 - relu_48 = None - mul_86 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_49 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_49 - ) = None - x_478 = ( - mul_86 - + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_86 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_479 = torch.nn.functional.dropout(x_478, 0.0, False, False) - x_478 = None - x_480 = torch._C._nn.linear( - x_479, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_479 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_481 = torch.nn.functional.dropout(x_480, 0.0, False, False) - x_480 = None - x_482 = mul_85 + x_481 - mul_85 = x_481 = None - view_38 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_87 = x_482 * view_38 - view_38 = None - layer_norm_72 = torch.nn.functional.layer_norm( - x_482, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_482 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_76 = torch._C._nn.linear( - layer_norm_72, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_72 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_41 = linear_76.reshape(1, pow_53, 3, 24, 32) - linear_76 = None - qkv_19 = reshape_41.permute(2, 0, 3, 1, 4) - reshape_41 = None - unbind_19 = qkv_19.unbind(0) - qkv_19 = None - q_19 = unbind_19[0] - k_19 = unbind_19[1] - v_19 = unbind_19[2] - unbind_19 = None - x_483 = torch._C._nn.scaled_dot_product_attention( - q_19, k_19, v_19, dropout_p=0.0 - ) - q_19 = k_19 = v_19 = None - transpose_22 = x_483.transpose(1, 2) - x_483 = None - x_484 = transpose_22.reshape(1, pow_53, 768) - transpose_22 = None - x_485 = torch._C._nn.linear( - x_484, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_484 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_486 = torch.nn.functional.dropout(x_485, 0.0, False, False) - x_485 = None - x_487 = mul_87 + x_486 - mul_87 = x_486 = None - view_39 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_88 = x_487 * view_39 - view_39 = None - layer_norm_73 = torch.nn.functional.layer_norm( - x_487, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_487 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_488 = torch._C._nn.linear( - layer_norm_73, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_73 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_49 = torch.nn.functional.relu(x_488, inplace=False) - x_488 = None - pow_50 = relu_49**2 - relu_49 = None - mul_89 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_50 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_50 - ) = None - x_489 = ( - mul_89 - + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_89 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_490 = torch.nn.functional.dropout(x_489, 0.0, False, False) - x_489 = None - x_491 = torch._C._nn.linear( - x_490, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_490 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_492 = torch.nn.functional.dropout(x_491, 0.0, False, False) - x_491 = None - x_493 = mul_88 + x_492 - mul_88 = x_492 = None - view_40 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_90 = x_493 * view_40 - view_40 = None - layer_norm_74 = torch.nn.functional.layer_norm( - x_493, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_493 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_80 = torch._C._nn.linear( - layer_norm_74, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_74 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_43 = linear_80.reshape(1, pow_53, 3, 24, 32) - linear_80 = None - qkv_20 = reshape_43.permute(2, 0, 3, 1, 4) - reshape_43 = None - unbind_20 = qkv_20.unbind(0) - qkv_20 = None - q_20 = unbind_20[0] - k_20 = unbind_20[1] - v_20 = unbind_20[2] - unbind_20 = None - x_494 = torch._C._nn.scaled_dot_product_attention( - q_20, k_20, v_20, dropout_p=0.0 - ) - q_20 = k_20 = v_20 = None - transpose_23 = x_494.transpose(1, 2) - x_494 = None - x_495 = transpose_23.reshape(1, pow_53, 768) - transpose_23 = pow_53 = None - x_496 = torch._C._nn.linear( - x_495, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_495 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_497 = torch.nn.functional.dropout(x_496, 0.0, False, False) - x_496 = None - x_498 = mul_90 + x_497 - mul_90 = x_497 = None - view_41 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_91 = x_498 * view_41 - view_41 = None - layer_norm_75 = torch.nn.functional.layer_norm( - x_498, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_498 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_499 = torch._C._nn.linear( - layer_norm_75, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_75 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_50 = torch.nn.functional.relu(x_499, inplace=False) - x_499 = None - pow_51 = relu_50**2 - relu_50 = None - mul_92 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_51 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_51 - ) = None - x_500 = ( - mul_92 - + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_92 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_501 = torch.nn.functional.dropout(x_500, 0.0, False, False) - x_500 = None - x_502 = torch._C._nn.linear( - x_501, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_501 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_503 = torch.nn.functional.dropout(x_502, 0.0, False, False) - x_502 = None - x_504 = mul_91 + x_503 - mul_91 = x_503 = None - transpose_24 = x_504.transpose(1, 2) - x_504 = None - x_505 = transpose_24.reshape(1, 768, sym_sum_4, sym_sum_4) - transpose_24 = sym_sum_4 = None - x_506 = torch.nn.functional.adaptive_avg_pool2d(x_505, 1) - x_505 = None - x_507 = x_506.permute(0, 2, 3, 1) - x_506 = None - x_508 = torch.nn.functional.layer_norm( - x_507, - (768,), - l_self_modules_head_modules_norm_parameters_weight_, - l_self_modules_head_modules_norm_parameters_bias_, - 1e-06, - ) - x_507 = ( - l_self_modules_head_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_norm_parameters_bias_ = None - x_509 = x_508.permute(0, 3, 1, 2) - x_508 = None - x_510 = x_509.flatten(1, -1) - x_509 = None - x_511 = torch.nn.functional.dropout(x_510, 0.0, False, False) - x_510 = None - x_512 = torch._C._nn.linear( - x_511, - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, - ) - x_511 = ( - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None - relu_51 = torch.nn.functional.relu(x_512, inplace=False) - x_512 = None - x_513 = torch.square(relu_51) - relu_51 = None - x_514 = torch.nn.functional.layer_norm( - x_513, - (3072,), - l_self_modules_head_modules_fc_modules_norm_parameters_weight_, - l_self_modules_head_modules_fc_modules_norm_parameters_bias_, - 1e-06, - ) - x_513 = ( - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None - x_515 = torch.nn.functional.dropout(x_514, 0.0, False, False) - x_514 = None - x_516 = torch._C._nn.linear( - x_515, - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, - ) - x_515 = ( - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None - return (x_516,) diff --git a/samples/timm/caformer_b36.sail_in1k/weight_meta.py b/samples/timm/caformer_b36.sail_in1k/weight_meta.py deleted file mode 100644 index 1067172347..0000000000 --- a/samples/timm/caformer_b36.sail_in1k/weight_meta.py +++ /dev/null @@ -1,3968 +0,0 @@ -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: - name = "L_self_modules_stem_modules_conv_parameters_weight_" - shape = [128, 3, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: - name = "L_self_modules_stem_modules_conv_parameters_bias_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_s1: - name = "s1" - shape = [] - dtype = "torch.int64" - device = "cpu" - mean = None - std = None - data = [4] - - -class Program_weight_tensor_meta_L_x_: - name = "L_x_" - shape = [1, 3, 224, 224] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.228 - std = 1.286 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: - name = "L_self_modules_stem_modules_norm_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [256, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [256, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [128, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [512, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [128, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [256, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [256, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [128, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [512, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [128, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [256, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [256, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [128, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [512, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [128, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" - shape = [256, 128, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" - shape = [512, 256, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" - shape = [768, 512, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [2304, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [768, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [3072, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [768, 3072] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [2304, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [768, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [3072, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [768, 3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [2304, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [768, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [3072, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [768, 3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_norm_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_norm_parameters_bias_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" - shape = [3072, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" - shape = [1000, 3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" - shape = [1000] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None diff --git a/samples/timm/caformer_b36.sail_in1k_384/graph_net.json b/samples/timm/caformer_b36.sail_in1k_384/graph_net.json deleted file mode 100644 index 1373fe3b58..0000000000 --- a/samples/timm/caformer_b36.sail_in1k_384/graph_net.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "framework": "torch", - "num_devices_required": 1, - "num_nodes_required": 1 -} \ No newline at end of file diff --git a/samples/timm/caformer_b36.sail_in1k_384/input_meta.py b/samples/timm/caformer_b36.sail_in1k_384/input_meta.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_b36.sail_in1k_384/input_tensor_constraints.py b/samples/timm/caformer_b36.sail_in1k_384/input_tensor_constraints.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_b36.sail_in1k_384/model.py b/samples/timm/caformer_b36.sail_in1k_384/model.py deleted file mode 100644 index 08d3de35cc..0000000000 --- a/samples/timm/caformer_b36.sail_in1k_384/model.py +++ /dev/null @@ -1,4758 +0,0 @@ -import torch - - -class GraphModule(torch.nn.Module): - def forward( - self, - L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - s1: torch.SymInt, - L_x_: torch.Tensor, - L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - ): - l_self_modules_stem_modules_conv_parameters_weight_ = ( - L_self_modules_stem_modules_conv_parameters_weight_ - ) - l_self_modules_stem_modules_conv_parameters_bias_ = ( - L_self_modules_stem_modules_conv_parameters_bias_ - ) - l_x_ = L_x_ - l_self_modules_stem_modules_norm_parameters_weight_ = ( - L_self_modules_stem_modules_norm_parameters_weight_ - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_head_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ - ) - x = torch.conv2d( - l_x_, - l_self_modules_stem_modules_conv_parameters_weight_, - l_self_modules_stem_modules_conv_parameters_bias_, - (4, 4), - (2, 2), - (1, 1), - 1, - ) - l_x_ = ( - l_self_modules_stem_modules_conv_parameters_weight_ - ) = l_self_modules_stem_modules_conv_parameters_bias_ = None - x_1 = x.permute(0, 2, 3, 1) - x = None - x_2 = torch.nn.functional.layer_norm( - x_1, - (128,), - l_self_modules_stem_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None - x_3 = x_2.permute(0, 3, 1, 2) - x_2 = None - sym_sum = torch.sym_sum([-3, s1]) - s1 = None - floordiv = sym_sum // 4 - sym_sum_1 = torch.sym_sum([1, floordiv]) - floordiv = sym_sum_1 = None - x_4 = x_3.permute(0, 2, 3, 1) - x_5 = torch.nn.functional.layer_norm( - x_4, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_6 = x_5.permute(0, 3, 1, 2) - x_5 = None - x_7 = torch.conv2d( - x_6, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu = torch.nn.functional.relu(x_7, inplace=False) - x_7 = None - pow_1 = relu**2 - relu = None - mul = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_1 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_1 - ) = None - x_8 = ( - mul - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_9 = torch.conv2d( - x_8, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 256, - ) - x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_10 = torch.conv2d( - x_9, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_11 = x_3 + x_10 - x_3 = x_10 = None - x_12 = x_11.permute(0, 2, 3, 1) - x_13 = torch.nn.functional.layer_norm( - x_12, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_14 = x_13.permute(0, 3, 1, 2) - x_13 = None - x_15 = torch.conv2d( - x_14, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_1 = torch.nn.functional.relu(x_15, inplace=False) - x_15 = None - pow_2 = relu_1**2 - relu_1 = None - mul_1 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_2 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_2 - ) = None - x_16 = ( - mul_1 - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) - x_16 = None - x_18 = torch.conv2d( - x_17, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) - x_18 = None - x_20 = x_11 + x_19 - x_11 = x_19 = None - x_21 = x_20.permute(0, 2, 3, 1) - x_22 = torch.nn.functional.layer_norm( - x_21, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_23 = x_22.permute(0, 3, 1, 2) - x_22 = None - x_24 = torch.conv2d( - x_23, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_2 = torch.nn.functional.relu(x_24, inplace=False) - x_24 = None - pow_3 = relu_2**2 - relu_2 = None - mul_2 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_3 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_3 - ) = None - x_25 = ( - mul_2 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_26 = torch.conv2d( - x_25, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 256, - ) - x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_27 = torch.conv2d( - x_26, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_28 = x_20 + x_27 - x_20 = x_27 = None - x_29 = x_28.permute(0, 2, 3, 1) - x_30 = torch.nn.functional.layer_norm( - x_29, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_31 = x_30.permute(0, 3, 1, 2) - x_30 = None - x_32 = torch.conv2d( - x_31, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_3 = torch.nn.functional.relu(x_32, inplace=False) - x_32 = None - pow_4 = relu_3**2 - relu_3 = None - mul_3 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_4 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_4 - ) = None - x_33 = ( - mul_3 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) - x_33 = None - x_35 = torch.conv2d( - x_34, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) - x_35 = None - x_37 = x_28 + x_36 - x_28 = x_36 = None - x_38 = x_37.permute(0, 2, 3, 1) - x_39 = torch.nn.functional.layer_norm( - x_38, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_40 = x_39.permute(0, 3, 1, 2) - x_39 = None - x_41 = torch.conv2d( - x_40, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_4 = torch.nn.functional.relu(x_41, inplace=False) - x_41 = None - pow_5 = relu_4**2 - relu_4 = None - mul_4 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_5 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_5 - ) = None - x_42 = ( - mul_4 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_43 = torch.conv2d( - x_42, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 256, - ) - x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_44 = torch.conv2d( - x_43, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_45 = x_37 + x_44 - x_37 = x_44 = None - x_46 = x_45.permute(0, 2, 3, 1) - x_47 = torch.nn.functional.layer_norm( - x_46, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_48 = x_47.permute(0, 3, 1, 2) - x_47 = None - x_49 = torch.conv2d( - x_48, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_5 = torch.nn.functional.relu(x_49, inplace=False) - x_49 = None - pow_6 = relu_5**2 - relu_5 = None - mul_5 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_6 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_6 - ) = None - x_50 = ( - mul_5 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) - x_50 = None - x_52 = torch.conv2d( - x_51, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) - x_52 = None - x_54 = x_45 + x_53 - x_45 = x_53 = None - x_55 = x_54.permute(0, 2, 3, 1) - x_54 = None - x_56 = torch.nn.functional.layer_norm( - x_55, - (128,), - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) - x_57 = x_56.permute(0, 3, 1, 2) - x_56 = None - x_58 = torch.conv2d( - x_57, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_1 = sym_sum // 8 - sym_sum_2 = torch.sym_sum([1, floordiv_1]) - floordiv_1 = sym_sum_2 = None - x_59 = x_58.permute(0, 2, 3, 1) - x_60 = torch.nn.functional.layer_norm( - x_59, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_61 = x_60.permute(0, 3, 1, 2) - x_60 = None - x_62 = torch.conv2d( - x_61, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_6 = torch.nn.functional.relu(x_62, inplace=False) - x_62 = None - pow_7 = relu_6**2 - relu_6 = None - mul_6 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_7 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_7 - ) = None - x_63 = ( - mul_6 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_64 = torch.conv2d( - x_63, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_65 = torch.conv2d( - x_64, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_66 = x_58 + x_65 - x_58 = x_65 = None - x_67 = x_66.permute(0, 2, 3, 1) - x_68 = torch.nn.functional.layer_norm( - x_67, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_69 = x_68.permute(0, 3, 1, 2) - x_68 = None - x_70 = torch.conv2d( - x_69, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_7 = torch.nn.functional.relu(x_70, inplace=False) - x_70 = None - pow_8 = relu_7**2 - relu_7 = None - mul_7 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_8 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_8 - ) = None - x_71 = ( - mul_7 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) - x_71 = None - x_73 = torch.conv2d( - x_72, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) - x_73 = None - x_75 = x_66 + x_74 - x_66 = x_74 = None - x_76 = x_75.permute(0, 2, 3, 1) - x_77 = torch.nn.functional.layer_norm( - x_76, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_78 = x_77.permute(0, 3, 1, 2) - x_77 = None - x_79 = torch.conv2d( - x_78, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_8 = torch.nn.functional.relu(x_79, inplace=False) - x_79 = None - pow_9 = relu_8**2 - relu_8 = None - mul_8 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_9 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_9 - ) = None - x_80 = ( - mul_8 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_81 = torch.conv2d( - x_80, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_82 = torch.conv2d( - x_81, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_83 = x_75 + x_82 - x_75 = x_82 = None - x_84 = x_83.permute(0, 2, 3, 1) - x_85 = torch.nn.functional.layer_norm( - x_84, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_86 = x_85.permute(0, 3, 1, 2) - x_85 = None - x_87 = torch.conv2d( - x_86, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_9 = torch.nn.functional.relu(x_87, inplace=False) - x_87 = None - pow_10 = relu_9**2 - relu_9 = None - mul_9 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_10 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_10 - ) = None - x_88 = ( - mul_9 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) - x_88 = None - x_90 = torch.conv2d( - x_89, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) - x_90 = None - x_92 = x_83 + x_91 - x_83 = x_91 = None - x_93 = x_92.permute(0, 2, 3, 1) - x_94 = torch.nn.functional.layer_norm( - x_93, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_95 = x_94.permute(0, 3, 1, 2) - x_94 = None - x_96 = torch.conv2d( - x_95, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_10 = torch.nn.functional.relu(x_96, inplace=False) - x_96 = None - pow_11 = relu_10**2 - relu_10 = None - mul_10 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_11 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_11 - ) = None - x_97 = ( - mul_10 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_98 = torch.conv2d( - x_97, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_99 = torch.conv2d( - x_98, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_100 = x_92 + x_99 - x_92 = x_99 = None - x_101 = x_100.permute(0, 2, 3, 1) - x_102 = torch.nn.functional.layer_norm( - x_101, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_103 = x_102.permute(0, 3, 1, 2) - x_102 = None - x_104 = torch.conv2d( - x_103, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_11 = torch.nn.functional.relu(x_104, inplace=False) - x_104 = None - pow_12 = relu_11**2 - relu_11 = None - mul_11 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_12 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_12 - ) = None - x_105 = ( - mul_11 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) - x_105 = None - x_107 = torch.conv2d( - x_106, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) - x_107 = None - x_109 = x_100 + x_108 - x_100 = x_108 = None - x_110 = x_109.permute(0, 2, 3, 1) - x_111 = torch.nn.functional.layer_norm( - x_110, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_110 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) - x_112 = x_111.permute(0, 3, 1, 2) - x_111 = None - x_113 = torch.conv2d( - x_112, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_112 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_12 = torch.nn.functional.relu(x_113, inplace=False) - x_113 = None - pow_13 = relu_12**2 - relu_12 = None - mul_12 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ - * pow_13 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_13 - ) = None - x_114 = ( - mul_12 - + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_12 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_115 = torch.conv2d( - x_114, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_114 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_116 = torch.conv2d( - x_115, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_115 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_117 = x_109 + x_116 - x_109 = x_116 = None - x_118 = x_117.permute(0, 2, 3, 1) - x_119 = torch.nn.functional.layer_norm( - x_118, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_118 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) - x_120 = x_119.permute(0, 3, 1, 2) - x_119 = None - x_121 = torch.conv2d( - x_120, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_120 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_13 = torch.nn.functional.relu(x_121, inplace=False) - x_121 = None - pow_14 = relu_13**2 - relu_13 = None - mul_13 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - * pow_14 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( - pow_14 - ) = None - x_122 = ( - mul_13 - + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - ) - mul_13 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) - x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) - x_122 = None - x_124 = torch.conv2d( - x_123, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_123 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) - x_124 = None - x_126 = x_117 + x_125 - x_117 = x_125 = None - x_127 = x_126.permute(0, 2, 3, 1) - x_128 = torch.nn.functional.layer_norm( - x_127, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_127 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) - x_129 = x_128.permute(0, 3, 1, 2) - x_128 = None - x_130 = torch.conv2d( - x_129, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_129 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_14 = torch.nn.functional.relu(x_130, inplace=False) - x_130 = None - pow_15 = relu_14**2 - relu_14 = None - mul_14 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ - * pow_15 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_15 - ) = None - x_131 = ( - mul_14 - + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_14 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_132 = torch.conv2d( - x_131, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_131 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_133 = torch.conv2d( - x_132, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_132 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_134 = x_126 + x_133 - x_126 = x_133 = None - x_135 = x_134.permute(0, 2, 3, 1) - x_136 = torch.nn.functional.layer_norm( - x_135, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_135 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) - x_137 = x_136.permute(0, 3, 1, 2) - x_136 = None - x_138 = torch.conv2d( - x_137, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_137 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_15 = torch.nn.functional.relu(x_138, inplace=False) - x_138 = None - pow_16 = relu_15**2 - relu_15 = None - mul_15 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - * pow_16 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( - pow_16 - ) = None - x_139 = ( - mul_15 - + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - ) - mul_15 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) - x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) - x_139 = None - x_141 = torch.conv2d( - x_140, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_140 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) - x_141 = None - x_143 = x_134 + x_142 - x_134 = x_142 = None - x_144 = x_143.permute(0, 2, 3, 1) - x_145 = torch.nn.functional.layer_norm( - x_144, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_144 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) - x_146 = x_145.permute(0, 3, 1, 2) - x_145 = None - x_147 = torch.conv2d( - x_146, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_146 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_16 = torch.nn.functional.relu(x_147, inplace=False) - x_147 = None - pow_17 = relu_16**2 - relu_16 = None - mul_16 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ - * pow_17 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_17 - ) = None - x_148 = ( - mul_16 - + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_16 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_149 = torch.conv2d( - x_148, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_148 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_150 = torch.conv2d( - x_149, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_149 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_151 = x_143 + x_150 - x_143 = x_150 = None - x_152 = x_151.permute(0, 2, 3, 1) - x_153 = torch.nn.functional.layer_norm( - x_152, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_152 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) - x_154 = x_153.permute(0, 3, 1, 2) - x_153 = None - x_155 = torch.conv2d( - x_154, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_154 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_17 = torch.nn.functional.relu(x_155, inplace=False) - x_155 = None - pow_18 = relu_17**2 - relu_17 = None - mul_17 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - * pow_18 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( - pow_18 - ) = None - x_156 = ( - mul_17 - + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - ) - mul_17 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) - x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) - x_156 = None - x_158 = torch.conv2d( - x_157, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_157 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) - x_158 = None - x_160 = x_151 + x_159 - x_151 = x_159 = None - x_161 = x_160.permute(0, 2, 3, 1) - x_162 = torch.nn.functional.layer_norm( - x_161, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_161 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) - x_163 = x_162.permute(0, 3, 1, 2) - x_162 = None - x_164 = torch.conv2d( - x_163, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_163 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_18 = torch.nn.functional.relu(x_164, inplace=False) - x_164 = None - pow_19 = relu_18**2 - relu_18 = None - mul_18 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ - * pow_19 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_19 - ) = None - x_165 = ( - mul_18 - + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_18 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_166 = torch.conv2d( - x_165, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_165 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_167 = torch.conv2d( - x_166, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_166 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_168 = x_160 + x_167 - x_160 = x_167 = None - x_169 = x_168.permute(0, 2, 3, 1) - x_170 = torch.nn.functional.layer_norm( - x_169, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_169 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) - x_171 = x_170.permute(0, 3, 1, 2) - x_170 = None - x_172 = torch.conv2d( - x_171, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_171 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_19 = torch.nn.functional.relu(x_172, inplace=False) - x_172 = None - pow_20 = relu_19**2 - relu_19 = None - mul_19 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - * pow_20 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( - pow_20 - ) = None - x_173 = ( - mul_19 - + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - ) - mul_19 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) - x_174 = torch.nn.functional.dropout(x_173, 0.0, False, False) - x_173 = None - x_175 = torch.conv2d( - x_174, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_174 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) - x_175 = None - x_177 = x_168 + x_176 - x_168 = x_176 = None - x_178 = x_177.permute(0, 2, 3, 1) - x_179 = torch.nn.functional.layer_norm( - x_178, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_178 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) - x_180 = x_179.permute(0, 3, 1, 2) - x_179 = None - x_181 = torch.conv2d( - x_180, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_180 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_20 = torch.nn.functional.relu(x_181, inplace=False) - x_181 = None - pow_21 = relu_20**2 - relu_20 = None - mul_20 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ - * pow_21 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_21 - ) = None - x_182 = ( - mul_20 - + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_20 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_183 = torch.conv2d( - x_182, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_182 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_184 = torch.conv2d( - x_183, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_183 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_185 = x_177 + x_184 - x_177 = x_184 = None - x_186 = x_185.permute(0, 2, 3, 1) - x_187 = torch.nn.functional.layer_norm( - x_186, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_186 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) - x_188 = x_187.permute(0, 3, 1, 2) - x_187 = None - x_189 = torch.conv2d( - x_188, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_188 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_21 = torch.nn.functional.relu(x_189, inplace=False) - x_189 = None - pow_22 = relu_21**2 - relu_21 = None - mul_21 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - * pow_22 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( - pow_22 - ) = None - x_190 = ( - mul_21 - + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - ) - mul_21 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) - x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) - x_190 = None - x_192 = torch.conv2d( - x_191, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_191 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) - x_192 = None - x_194 = x_185 + x_193 - x_185 = x_193 = None - x_195 = x_194.permute(0, 2, 3, 1) - x_196 = torch.nn.functional.layer_norm( - x_195, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_195 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) - x_197 = x_196.permute(0, 3, 1, 2) - x_196 = None - x_198 = torch.conv2d( - x_197, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_197 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_22 = torch.nn.functional.relu(x_198, inplace=False) - x_198 = None - pow_23 = relu_22**2 - relu_22 = None - mul_22 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ - * pow_23 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_23 - ) = None - x_199 = ( - mul_22 - + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_22 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_200 = torch.conv2d( - x_199, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_199 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_201 = torch.conv2d( - x_200, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_200 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_202 = x_194 + x_201 - x_194 = x_201 = None - x_203 = x_202.permute(0, 2, 3, 1) - x_204 = torch.nn.functional.layer_norm( - x_203, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_203 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) - x_205 = x_204.permute(0, 3, 1, 2) - x_204 = None - x_206 = torch.conv2d( - x_205, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_205 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_23 = torch.nn.functional.relu(x_206, inplace=False) - x_206 = None - pow_24 = relu_23**2 - relu_23 = None - mul_23 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - * pow_24 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( - pow_24 - ) = None - x_207 = ( - mul_23 - + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - ) - mul_23 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) - x_208 = torch.nn.functional.dropout(x_207, 0.0, False, False) - x_207 = None - x_209 = torch.conv2d( - x_208, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_208 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) - x_209 = None - x_211 = x_202 + x_210 - x_202 = x_210 = None - x_212 = x_211.permute(0, 2, 3, 1) - x_213 = torch.nn.functional.layer_norm( - x_212, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_212 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) - x_214 = x_213.permute(0, 3, 1, 2) - x_213 = None - x_215 = torch.conv2d( - x_214, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_214 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_24 = torch.nn.functional.relu(x_215, inplace=False) - x_215 = None - pow_25 = relu_24**2 - relu_24 = None - mul_24 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ - * pow_25 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_25 - ) = None - x_216 = ( - mul_24 - + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_24 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_217 = torch.conv2d( - x_216, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_216 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_218 = torch.conv2d( - x_217, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_217 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_219 = x_211 + x_218 - x_211 = x_218 = None - x_220 = x_219.permute(0, 2, 3, 1) - x_221 = torch.nn.functional.layer_norm( - x_220, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_220 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) - x_222 = x_221.permute(0, 3, 1, 2) - x_221 = None - x_223 = torch.conv2d( - x_222, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_222 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_25 = torch.nn.functional.relu(x_223, inplace=False) - x_223 = None - pow_26 = relu_25**2 - relu_25 = None - mul_25 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - * pow_26 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( - pow_26 - ) = None - x_224 = ( - mul_25 - + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - ) - mul_25 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) - x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) - x_224 = None - x_226 = torch.conv2d( - x_225, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_225 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) - x_226 = None - x_228 = x_219 + x_227 - x_219 = x_227 = None - x_229 = x_228.permute(0, 2, 3, 1) - x_230 = torch.nn.functional.layer_norm( - x_229, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_229 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) - x_231 = x_230.permute(0, 3, 1, 2) - x_230 = None - x_232 = torch.conv2d( - x_231, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_231 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_26 = torch.nn.functional.relu(x_232, inplace=False) - x_232 = None - pow_27 = relu_26**2 - relu_26 = None - mul_26 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ - * pow_27 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_27 - ) = None - x_233 = ( - mul_26 - + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_26 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_234 = torch.conv2d( - x_233, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_233 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_235 = torch.conv2d( - x_234, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_234 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_236 = x_228 + x_235 - x_228 = x_235 = None - x_237 = x_236.permute(0, 2, 3, 1) - x_238 = torch.nn.functional.layer_norm( - x_237, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_237 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) - x_239 = x_238.permute(0, 3, 1, 2) - x_238 = None - x_240 = torch.conv2d( - x_239, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_239 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_27 = torch.nn.functional.relu(x_240, inplace=False) - x_240 = None - pow_28 = relu_27**2 - relu_27 = None - mul_27 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - * pow_28 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( - pow_28 - ) = None - x_241 = ( - mul_27 - + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - ) - mul_27 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) - x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) - x_241 = None - x_243 = torch.conv2d( - x_242, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_242 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_244 = torch.nn.functional.dropout(x_243, 0.0, False, False) - x_243 = None - x_245 = x_236 + x_244 - x_236 = x_244 = None - x_246 = x_245.permute(0, 2, 3, 1) - x_247 = torch.nn.functional.layer_norm( - x_246, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_246 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) - x_248 = x_247.permute(0, 3, 1, 2) - x_247 = None - x_249 = torch.conv2d( - x_248, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_248 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_28 = torch.nn.functional.relu(x_249, inplace=False) - x_249 = None - pow_29 = relu_28**2 - relu_28 = None - mul_28 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ - * pow_29 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_29 - ) = None - x_250 = ( - mul_28 - + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_28 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_251 = torch.conv2d( - x_250, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_250 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_252 = torch.conv2d( - x_251, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_251 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_253 = x_245 + x_252 - x_245 = x_252 = None - x_254 = x_253.permute(0, 2, 3, 1) - x_255 = torch.nn.functional.layer_norm( - x_254, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_254 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) - x_256 = x_255.permute(0, 3, 1, 2) - x_255 = None - x_257 = torch.conv2d( - x_256, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_256 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_29 = torch.nn.functional.relu(x_257, inplace=False) - x_257 = None - pow_30 = relu_29**2 - relu_29 = None - mul_29 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - * pow_30 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( - pow_30 - ) = None - x_258 = ( - mul_29 - + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - ) - mul_29 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) - x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) - x_258 = None - x_260 = torch.conv2d( - x_259, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_259 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_261 = torch.nn.functional.dropout(x_260, 0.0, False, False) - x_260 = None - x_262 = x_253 + x_261 - x_253 = x_261 = None - x_263 = x_262.permute(0, 2, 3, 1) - x_262 = None - x_264 = torch.nn.functional.layer_norm( - x_263, - (256,), - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_263 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) - x_265 = x_264.permute(0, 3, 1, 2) - x_264 = None - x_266 = torch.conv2d( - x_265, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_265 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_2 = sym_sum // 16 - sym_sum_3 = torch.sym_sum([1, floordiv_2]) - floordiv_2 = None - reshape = x_266.reshape(1, 512, -1) - x_266 = None - x_267 = reshape.transpose(1, 2) - reshape = None - view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_30 = x_267 * view - view = None - layer_norm_33 = torch.nn.functional.layer_norm( - x_267, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_52 = sym_sum_3**2 - linear = torch._C._nn.linear( - layer_norm_33, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_33 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_1 = linear.reshape(1, pow_52, 3, 16, 32) - linear = None - qkv = reshape_1.permute(2, 0, 3, 1, 4) - reshape_1 = None - unbind = qkv.unbind(0) - qkv = None - q = unbind[0] - k = unbind[1] - v = unbind[2] - unbind = None - x_268 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) - q = k = v = None - transpose_1 = x_268.transpose(1, 2) - x_268 = None - x_269 = transpose_1.reshape(1, pow_52, 512) - transpose_1 = None - x_270 = torch._C._nn.linear( - x_269, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_269 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_271 = torch.nn.functional.dropout(x_270, 0.0, False, False) - x_270 = None - x_272 = mul_30 + x_271 - mul_30 = x_271 = None - view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_31 = x_272 * view_1 - view_1 = None - layer_norm_34 = torch.nn.functional.layer_norm( - x_272, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_272 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_273 = torch._C._nn.linear( - layer_norm_34, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_34 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_30 = torch.nn.functional.relu(x_273, inplace=False) - x_273 = None - pow_31 = relu_30**2 - relu_30 = None - mul_32 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_31 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_31 - ) = None - x_274 = ( - mul_32 - + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_275 = torch.nn.functional.dropout(x_274, 0.0, False, False) - x_274 = None - x_276 = torch._C._nn.linear( - x_275, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_277 = torch.nn.functional.dropout(x_276, 0.0, False, False) - x_276 = None - x_278 = mul_31 + x_277 - mul_31 = x_277 = None - view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_33 = x_278 * view_2 - view_2 = None - layer_norm_35 = torch.nn.functional.layer_norm( - x_278, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_4 = torch._C._nn.linear( - layer_norm_35, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_3 = linear_4.reshape(1, pow_52, 3, 16, 32) - linear_4 = None - qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) - reshape_3 = None - unbind_1 = qkv_1.unbind(0) - qkv_1 = None - q_1 = unbind_1[0] - k_1 = unbind_1[1] - v_1 = unbind_1[2] - unbind_1 = None - x_279 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) - q_1 = k_1 = v_1 = None - transpose_2 = x_279.transpose(1, 2) - x_279 = None - x_280 = transpose_2.reshape(1, pow_52, 512) - transpose_2 = None - x_281 = torch._C._nn.linear( - x_280, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_280 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) - x_281 = None - x_283 = mul_33 + x_282 - mul_33 = x_282 = None - view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_34 = x_283 * view_3 - view_3 = None - layer_norm_36 = torch.nn.functional.layer_norm( - x_283, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_283 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_284 = torch._C._nn.linear( - layer_norm_36, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_36 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_31 = torch.nn.functional.relu(x_284, inplace=False) - x_284 = None - pow_32 = relu_31**2 - relu_31 = None - mul_35 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_32 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_32 - ) = None - x_285 = ( - mul_35 - + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_286 = torch.nn.functional.dropout(x_285, 0.0, False, False) - x_285 = None - x_287 = torch._C._nn.linear( - x_286, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) - x_287 = None - x_289 = mul_34 + x_288 - mul_34 = x_288 = None - view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_36 = x_289 * view_4 - view_4 = None - layer_norm_37 = torch.nn.functional.layer_norm( - x_289, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_8 = torch._C._nn.linear( - layer_norm_37, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_37 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_5 = linear_8.reshape(1, pow_52, 3, 16, 32) - linear_8 = None - qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) - reshape_5 = None - unbind_2 = qkv_2.unbind(0) - qkv_2 = None - q_2 = unbind_2[0] - k_2 = unbind_2[1] - v_2 = unbind_2[2] - unbind_2 = None - x_290 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) - q_2 = k_2 = v_2 = None - transpose_3 = x_290.transpose(1, 2) - x_290 = None - x_291 = transpose_3.reshape(1, pow_52, 512) - transpose_3 = None - x_292 = torch._C._nn.linear( - x_291, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_291 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_293 = torch.nn.functional.dropout(x_292, 0.0, False, False) - x_292 = None - x_294 = mul_36 + x_293 - mul_36 = x_293 = None - view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_37 = x_294 * view_5 - view_5 = None - layer_norm_38 = torch.nn.functional.layer_norm( - x_294, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_294 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_295 = torch._C._nn.linear( - layer_norm_38, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_32 = torch.nn.functional.relu(x_295, inplace=False) - x_295 = None - pow_33 = relu_32**2 - relu_32 = None - mul_38 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_33 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_33 - ) = None - x_296 = ( - mul_38 - + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_297 = torch.nn.functional.dropout(x_296, 0.0, False, False) - x_296 = None - x_298 = torch._C._nn.linear( - x_297, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_299 = torch.nn.functional.dropout(x_298, 0.0, False, False) - x_298 = None - x_300 = mul_37 + x_299 - mul_37 = x_299 = None - view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_39 = x_300 * view_6 - view_6 = None - layer_norm_39 = torch.nn.functional.layer_norm( - x_300, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) - linear_12 = torch._C._nn.linear( - layer_norm_39, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_39 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_7 = linear_12.reshape(1, pow_52, 3, 16, 32) - linear_12 = None - qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) - reshape_7 = None - unbind_3 = qkv_3.unbind(0) - qkv_3 = None - q_3 = unbind_3[0] - k_3 = unbind_3[1] - v_3 = unbind_3[2] - unbind_3 = None - x_301 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) - q_3 = k_3 = v_3 = None - transpose_4 = x_301.transpose(1, 2) - x_301 = None - x_302 = transpose_4.reshape(1, pow_52, 512) - transpose_4 = None - x_303 = torch._C._nn.linear( - x_302, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_302 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_304 = torch.nn.functional.dropout(x_303, 0.0, False, False) - x_303 = None - x_305 = mul_39 + x_304 - mul_39 = x_304 = None - view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_40 = x_305 * view_7 - view_7 = None - layer_norm_40 = torch.nn.functional.layer_norm( - x_305, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_305 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) - x_306 = torch._C._nn.linear( - layer_norm_40, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_40 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_33 = torch.nn.functional.relu(x_306, inplace=False) - x_306 = None - pow_34 = relu_33**2 - relu_33 = None - mul_41 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - * pow_34 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( - pow_34 - ) = None - x_307 = ( - mul_41 - + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - ) - mul_41 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) - x_308 = torch.nn.functional.dropout(x_307, 0.0, False, False) - x_307 = None - x_309 = torch._C._nn.linear( - x_308, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) - x_309 = None - x_311 = mul_40 + x_310 - mul_40 = x_310 = None - view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_42 = x_311 * view_8 - view_8 = None - layer_norm_41 = torch.nn.functional.layer_norm( - x_311, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) - linear_16 = torch._C._nn.linear( - layer_norm_41, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_41 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_9 = linear_16.reshape(1, pow_52, 3, 16, 32) - linear_16 = None - qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) - reshape_9 = None - unbind_4 = qkv_4.unbind(0) - qkv_4 = None - q_4 = unbind_4[0] - k_4 = unbind_4[1] - v_4 = unbind_4[2] - unbind_4 = None - x_312 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) - q_4 = k_4 = v_4 = None - transpose_5 = x_312.transpose(1, 2) - x_312 = None - x_313 = transpose_5.reshape(1, pow_52, 512) - transpose_5 = None - x_314 = torch._C._nn.linear( - x_313, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_313 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) - x_314 = None - x_316 = mul_42 + x_315 - mul_42 = x_315 = None - view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_43 = x_316 * view_9 - view_9 = None - layer_norm_42 = torch.nn.functional.layer_norm( - x_316, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_316 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) - x_317 = torch._C._nn.linear( - layer_norm_42, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_42 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_34 = torch.nn.functional.relu(x_317, inplace=False) - x_317 = None - pow_35 = relu_34**2 - relu_34 = None - mul_44 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - * pow_35 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( - pow_35 - ) = None - x_318 = ( - mul_44 - + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - ) - mul_44 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) - x_319 = torch.nn.functional.dropout(x_318, 0.0, False, False) - x_318 = None - x_320 = torch._C._nn.linear( - x_319, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_321 = torch.nn.functional.dropout(x_320, 0.0, False, False) - x_320 = None - x_322 = mul_43 + x_321 - mul_43 = x_321 = None - view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_45 = x_322 * view_10 - view_10 = None - layer_norm_43 = torch.nn.functional.layer_norm( - x_322, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) - linear_20 = torch._C._nn.linear( - layer_norm_43, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_43 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_11 = linear_20.reshape(1, pow_52, 3, 16, 32) - linear_20 = None - qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) - reshape_11 = None - unbind_5 = qkv_5.unbind(0) - qkv_5 = None - q_5 = unbind_5[0] - k_5 = unbind_5[1] - v_5 = unbind_5[2] - unbind_5 = None - x_323 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) - q_5 = k_5 = v_5 = None - transpose_6 = x_323.transpose(1, 2) - x_323 = None - x_324 = transpose_6.reshape(1, pow_52, 512) - transpose_6 = None - x_325 = torch._C._nn.linear( - x_324, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_324 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_326 = torch.nn.functional.dropout(x_325, 0.0, False, False) - x_325 = None - x_327 = mul_45 + x_326 - mul_45 = x_326 = None - view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_46 = x_327 * view_11 - view_11 = None - layer_norm_44 = torch.nn.functional.layer_norm( - x_327, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_327 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) - x_328 = torch._C._nn.linear( - layer_norm_44, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_44 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_35 = torch.nn.functional.relu(x_328, inplace=False) - x_328 = None - pow_36 = relu_35**2 - relu_35 = None - mul_47 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - * pow_36 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( - pow_36 - ) = None - x_329 = ( - mul_47 - + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - ) - mul_47 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) - x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) - x_329 = None - x_331 = torch._C._nn.linear( - x_330, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_332 = torch.nn.functional.dropout(x_331, 0.0, False, False) - x_331 = None - x_333 = mul_46 + x_332 - mul_46 = x_332 = None - view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_48 = x_333 * view_12 - view_12 = None - layer_norm_45 = torch.nn.functional.layer_norm( - x_333, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) - linear_24 = torch._C._nn.linear( - layer_norm_45, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_45 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_13 = linear_24.reshape(1, pow_52, 3, 16, 32) - linear_24 = None - qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) - reshape_13 = None - unbind_6 = qkv_6.unbind(0) - qkv_6 = None - q_6 = unbind_6[0] - k_6 = unbind_6[1] - v_6 = unbind_6[2] - unbind_6 = None - x_334 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) - q_6 = k_6 = v_6 = None - transpose_7 = x_334.transpose(1, 2) - x_334 = None - x_335 = transpose_7.reshape(1, pow_52, 512) - transpose_7 = None - x_336 = torch._C._nn.linear( - x_335, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_335 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_337 = torch.nn.functional.dropout(x_336, 0.0, False, False) - x_336 = None - x_338 = mul_48 + x_337 - mul_48 = x_337 = None - view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_49 = x_338 * view_13 - view_13 = None - layer_norm_46 = torch.nn.functional.layer_norm( - x_338, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_338 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) - x_339 = torch._C._nn.linear( - layer_norm_46, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_46 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_36 = torch.nn.functional.relu(x_339, inplace=False) - x_339 = None - pow_37 = relu_36**2 - relu_36 = None - mul_50 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - * pow_37 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( - pow_37 - ) = None - x_340 = ( - mul_50 - + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - ) - mul_50 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) - x_341 = torch.nn.functional.dropout(x_340, 0.0, False, False) - x_340 = None - x_342 = torch._C._nn.linear( - x_341, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_343 = torch.nn.functional.dropout(x_342, 0.0, False, False) - x_342 = None - x_344 = mul_49 + x_343 - mul_49 = x_343 = None - view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_51 = x_344 * view_14 - view_14 = None - layer_norm_47 = torch.nn.functional.layer_norm( - x_344, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) - linear_28 = torch._C._nn.linear( - layer_norm_47, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_47 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_15 = linear_28.reshape(1, pow_52, 3, 16, 32) - linear_28 = None - qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) - reshape_15 = None - unbind_7 = qkv_7.unbind(0) - qkv_7 = None - q_7 = unbind_7[0] - k_7 = unbind_7[1] - v_7 = unbind_7[2] - unbind_7 = None - x_345 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) - q_7 = k_7 = v_7 = None - transpose_8 = x_345.transpose(1, 2) - x_345 = None - x_346 = transpose_8.reshape(1, pow_52, 512) - transpose_8 = None - x_347 = torch._C._nn.linear( - x_346, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_346 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) - x_347 = None - x_349 = mul_51 + x_348 - mul_51 = x_348 = None - view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_52 = x_349 * view_15 - view_15 = None - layer_norm_48 = torch.nn.functional.layer_norm( - x_349, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_349 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) - x_350 = torch._C._nn.linear( - layer_norm_48, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_48 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_37 = torch.nn.functional.relu(x_350, inplace=False) - x_350 = None - pow_38 = relu_37**2 - relu_37 = None - mul_53 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - * pow_38 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( - pow_38 - ) = None - x_351 = ( - mul_53 - + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - ) - mul_53 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) - x_352 = torch.nn.functional.dropout(x_351, 0.0, False, False) - x_351 = None - x_353 = torch._C._nn.linear( - x_352, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_354 = torch.nn.functional.dropout(x_353, 0.0, False, False) - x_353 = None - x_355 = mul_52 + x_354 - mul_52 = x_354 = None - view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_54 = x_355 * view_16 - view_16 = None - layer_norm_49 = torch.nn.functional.layer_norm( - x_355, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) - linear_32 = torch._C._nn.linear( - layer_norm_49, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_49 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_17 = linear_32.reshape(1, pow_52, 3, 16, 32) - linear_32 = None - qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) - reshape_17 = None - unbind_8 = qkv_8.unbind(0) - qkv_8 = None - q_8 = unbind_8[0] - k_8 = unbind_8[1] - v_8 = unbind_8[2] - unbind_8 = None - x_356 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) - q_8 = k_8 = v_8 = None - transpose_9 = x_356.transpose(1, 2) - x_356 = None - x_357 = transpose_9.reshape(1, pow_52, 512) - transpose_9 = None - x_358 = torch._C._nn.linear( - x_357, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_357 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_359 = torch.nn.functional.dropout(x_358, 0.0, False, False) - x_358 = None - x_360 = mul_54 + x_359 - mul_54 = x_359 = None - view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_55 = x_360 * view_17 - view_17 = None - layer_norm_50 = torch.nn.functional.layer_norm( - x_360, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_360 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) - x_361 = torch._C._nn.linear( - layer_norm_50, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_50 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_38 = torch.nn.functional.relu(x_361, inplace=False) - x_361 = None - pow_39 = relu_38**2 - relu_38 = None - mul_56 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - * pow_39 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( - pow_39 - ) = None - x_362 = ( - mul_56 - + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - ) - mul_56 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) - x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) - x_362 = None - x_364 = torch._C._nn.linear( - x_363, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_365 = torch.nn.functional.dropout(x_364, 0.0, False, False) - x_364 = None - x_366 = mul_55 + x_365 - mul_55 = x_365 = None - view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_57 = x_366 * view_18 - view_18 = None - layer_norm_51 = torch.nn.functional.layer_norm( - x_366, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) - linear_36 = torch._C._nn.linear( - layer_norm_51, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_51 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_19 = linear_36.reshape(1, pow_52, 3, 16, 32) - linear_36 = None - qkv_9 = reshape_19.permute(2, 0, 3, 1, 4) - reshape_19 = None - unbind_9 = qkv_9.unbind(0) - qkv_9 = None - q_9 = unbind_9[0] - k_9 = unbind_9[1] - v_9 = unbind_9[2] - unbind_9 = None - x_367 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) - q_9 = k_9 = v_9 = None - transpose_10 = x_367.transpose(1, 2) - x_367 = None - x_368 = transpose_10.reshape(1, pow_52, 512) - transpose_10 = None - x_369 = torch._C._nn.linear( - x_368, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_368 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_370 = torch.nn.functional.dropout(x_369, 0.0, False, False) - x_369 = None - x_371 = mul_57 + x_370 - mul_57 = x_370 = None - view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_58 = x_371 * view_19 - view_19 = None - layer_norm_52 = torch.nn.functional.layer_norm( - x_371, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_371 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) - x_372 = torch._C._nn.linear( - layer_norm_52, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_52 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_39 = torch.nn.functional.relu(x_372, inplace=False) - x_372 = None - pow_40 = relu_39**2 - relu_39 = None - mul_59 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - * pow_40 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( - pow_40 - ) = None - x_373 = ( - mul_59 - + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - ) - mul_59 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) - x_374 = torch.nn.functional.dropout(x_373, 0.0, False, False) - x_373 = None - x_375 = torch._C._nn.linear( - x_374, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_374 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_376 = torch.nn.functional.dropout(x_375, 0.0, False, False) - x_375 = None - x_377 = mul_58 + x_376 - mul_58 = x_376 = None - view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_60 = x_377 * view_20 - view_20 = None - layer_norm_53 = torch.nn.functional.layer_norm( - x_377, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_377 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) - linear_40 = torch._C._nn.linear( - layer_norm_53, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_53 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_21 = linear_40.reshape(1, pow_52, 3, 16, 32) - linear_40 = None - qkv_10 = reshape_21.permute(2, 0, 3, 1, 4) - reshape_21 = None - unbind_10 = qkv_10.unbind(0) - qkv_10 = None - q_10 = unbind_10[0] - k_10 = unbind_10[1] - v_10 = unbind_10[2] - unbind_10 = None - x_378 = torch._C._nn.scaled_dot_product_attention( - q_10, k_10, v_10, dropout_p=0.0 - ) - q_10 = k_10 = v_10 = None - transpose_11 = x_378.transpose(1, 2) - x_378 = None - x_379 = transpose_11.reshape(1, pow_52, 512) - transpose_11 = None - x_380 = torch._C._nn.linear( - x_379, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_379 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_381 = torch.nn.functional.dropout(x_380, 0.0, False, False) - x_380 = None - x_382 = mul_60 + x_381 - mul_60 = x_381 = None - view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_61 = x_382 * view_21 - view_21 = None - layer_norm_54 = torch.nn.functional.layer_norm( - x_382, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_382 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) - x_383 = torch._C._nn.linear( - layer_norm_54, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_54 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_40 = torch.nn.functional.relu(x_383, inplace=False) - x_383 = None - pow_41 = relu_40**2 - relu_40 = None - mul_62 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - * pow_41 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( - pow_41 - ) = None - x_384 = ( - mul_62 - + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - ) - mul_62 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) - x_385 = torch.nn.functional.dropout(x_384, 0.0, False, False) - x_384 = None - x_386 = torch._C._nn.linear( - x_385, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_385 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_387 = torch.nn.functional.dropout(x_386, 0.0, False, False) - x_386 = None - x_388 = mul_61 + x_387 - mul_61 = x_387 = None - view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_63 = x_388 * view_22 - view_22 = None - layer_norm_55 = torch.nn.functional.layer_norm( - x_388, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_388 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) - linear_44 = torch._C._nn.linear( - layer_norm_55, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_55 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_23 = linear_44.reshape(1, pow_52, 3, 16, 32) - linear_44 = None - qkv_11 = reshape_23.permute(2, 0, 3, 1, 4) - reshape_23 = None - unbind_11 = qkv_11.unbind(0) - qkv_11 = None - q_11 = unbind_11[0] - k_11 = unbind_11[1] - v_11 = unbind_11[2] - unbind_11 = None - x_389 = torch._C._nn.scaled_dot_product_attention( - q_11, k_11, v_11, dropout_p=0.0 - ) - q_11 = k_11 = v_11 = None - transpose_12 = x_389.transpose(1, 2) - x_389 = None - x_390 = transpose_12.reshape(1, pow_52, 512) - transpose_12 = None - x_391 = torch._C._nn.linear( - x_390, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_390 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_392 = torch.nn.functional.dropout(x_391, 0.0, False, False) - x_391 = None - x_393 = mul_63 + x_392 - mul_63 = x_392 = None - view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_64 = x_393 * view_23 - view_23 = None - layer_norm_56 = torch.nn.functional.layer_norm( - x_393, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_393 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) - x_394 = torch._C._nn.linear( - layer_norm_56, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_56 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_41 = torch.nn.functional.relu(x_394, inplace=False) - x_394 = None - pow_42 = relu_41**2 - relu_41 = None - mul_65 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - * pow_42 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( - pow_42 - ) = None - x_395 = ( - mul_65 - + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - ) - mul_65 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) - x_396 = torch.nn.functional.dropout(x_395, 0.0, False, False) - x_395 = None - x_397 = torch._C._nn.linear( - x_396, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_396 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_398 = torch.nn.functional.dropout(x_397, 0.0, False, False) - x_397 = None - x_399 = mul_64 + x_398 - mul_64 = x_398 = None - view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_66 = x_399 * view_24 - view_24 = None - layer_norm_57 = torch.nn.functional.layer_norm( - x_399, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_399 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = (None) - linear_48 = torch._C._nn.linear( - layer_norm_57, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_57 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_25 = linear_48.reshape(1, pow_52, 3, 16, 32) - linear_48 = None - qkv_12 = reshape_25.permute(2, 0, 3, 1, 4) - reshape_25 = None - unbind_12 = qkv_12.unbind(0) - qkv_12 = None - q_12 = unbind_12[0] - k_12 = unbind_12[1] - v_12 = unbind_12[2] - unbind_12 = None - x_400 = torch._C._nn.scaled_dot_product_attention( - q_12, k_12, v_12, dropout_p=0.0 - ) - q_12 = k_12 = v_12 = None - transpose_13 = x_400.transpose(1, 2) - x_400 = None - x_401 = transpose_13.reshape(1, pow_52, 512) - transpose_13 = None - x_402 = torch._C._nn.linear( - x_401, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_401 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_403 = torch.nn.functional.dropout(x_402, 0.0, False, False) - x_402 = None - x_404 = mul_66 + x_403 - mul_66 = x_403 = None - view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_67 = x_404 * view_25 - view_25 = None - layer_norm_58 = torch.nn.functional.layer_norm( - x_404, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_404 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = (None) - x_405 = torch._C._nn.linear( - layer_norm_58, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_58 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_42 = torch.nn.functional.relu(x_405, inplace=False) - x_405 = None - pow_43 = relu_42**2 - relu_42 = None - mul_68 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ - * pow_43 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = ( - pow_43 - ) = None - x_406 = ( - mul_68 - + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ - ) - mul_68 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = (None) - x_407 = torch.nn.functional.dropout(x_406, 0.0, False, False) - x_406 = None - x_408 = torch._C._nn.linear( - x_407, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_407 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_409 = torch.nn.functional.dropout(x_408, 0.0, False, False) - x_408 = None - x_410 = mul_67 + x_409 - mul_67 = x_409 = None - view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_69 = x_410 * view_26 - view_26 = None - layer_norm_59 = torch.nn.functional.layer_norm( - x_410, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_410 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = (None) - linear_52 = torch._C._nn.linear( - layer_norm_59, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_59 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_27 = linear_52.reshape(1, pow_52, 3, 16, 32) - linear_52 = None - qkv_13 = reshape_27.permute(2, 0, 3, 1, 4) - reshape_27 = None - unbind_13 = qkv_13.unbind(0) - qkv_13 = None - q_13 = unbind_13[0] - k_13 = unbind_13[1] - v_13 = unbind_13[2] - unbind_13 = None - x_411 = torch._C._nn.scaled_dot_product_attention( - q_13, k_13, v_13, dropout_p=0.0 - ) - q_13 = k_13 = v_13 = None - transpose_14 = x_411.transpose(1, 2) - x_411 = None - x_412 = transpose_14.reshape(1, pow_52, 512) - transpose_14 = None - x_413 = torch._C._nn.linear( - x_412, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_412 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_414 = torch.nn.functional.dropout(x_413, 0.0, False, False) - x_413 = None - x_415 = mul_69 + x_414 - mul_69 = x_414 = None - view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_70 = x_415 * view_27 - view_27 = None - layer_norm_60 = torch.nn.functional.layer_norm( - x_415, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_415 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = (None) - x_416 = torch._C._nn.linear( - layer_norm_60, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_60 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_43 = torch.nn.functional.relu(x_416, inplace=False) - x_416 = None - pow_44 = relu_43**2 - relu_43 = None - mul_71 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ - * pow_44 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = ( - pow_44 - ) = None - x_417 = ( - mul_71 - + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ - ) - mul_71 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = (None) - x_418 = torch.nn.functional.dropout(x_417, 0.0, False, False) - x_417 = None - x_419 = torch._C._nn.linear( - x_418, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_418 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_420 = torch.nn.functional.dropout(x_419, 0.0, False, False) - x_419 = None - x_421 = mul_70 + x_420 - mul_70 = x_420 = None - view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_72 = x_421 * view_28 - view_28 = None - layer_norm_61 = torch.nn.functional.layer_norm( - x_421, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_421 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = (None) - linear_56 = torch._C._nn.linear( - layer_norm_61, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_61 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_29 = linear_56.reshape(1, pow_52, 3, 16, 32) - linear_56 = None - qkv_14 = reshape_29.permute(2, 0, 3, 1, 4) - reshape_29 = None - unbind_14 = qkv_14.unbind(0) - qkv_14 = None - q_14 = unbind_14[0] - k_14 = unbind_14[1] - v_14 = unbind_14[2] - unbind_14 = None - x_422 = torch._C._nn.scaled_dot_product_attention( - q_14, k_14, v_14, dropout_p=0.0 - ) - q_14 = k_14 = v_14 = None - transpose_15 = x_422.transpose(1, 2) - x_422 = None - x_423 = transpose_15.reshape(1, pow_52, 512) - transpose_15 = None - x_424 = torch._C._nn.linear( - x_423, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_423 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_425 = torch.nn.functional.dropout(x_424, 0.0, False, False) - x_424 = None - x_426 = mul_72 + x_425 - mul_72 = x_425 = None - view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_73 = x_426 * view_29 - view_29 = None - layer_norm_62 = torch.nn.functional.layer_norm( - x_426, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_426 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = (None) - x_427 = torch._C._nn.linear( - layer_norm_62, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_62 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_44 = torch.nn.functional.relu(x_427, inplace=False) - x_427 = None - pow_45 = relu_44**2 - relu_44 = None - mul_74 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ - * pow_45 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = ( - pow_45 - ) = None - x_428 = ( - mul_74 - + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ - ) - mul_74 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = (None) - x_429 = torch.nn.functional.dropout(x_428, 0.0, False, False) - x_428 = None - x_430 = torch._C._nn.linear( - x_429, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_429 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_431 = torch.nn.functional.dropout(x_430, 0.0, False, False) - x_430 = None - x_432 = mul_73 + x_431 - mul_73 = x_431 = None - view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_75 = x_432 * view_30 - view_30 = None - layer_norm_63 = torch.nn.functional.layer_norm( - x_432, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_432 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = (None) - linear_60 = torch._C._nn.linear( - layer_norm_63, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_63 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_31 = linear_60.reshape(1, pow_52, 3, 16, 32) - linear_60 = None - qkv_15 = reshape_31.permute(2, 0, 3, 1, 4) - reshape_31 = None - unbind_15 = qkv_15.unbind(0) - qkv_15 = None - q_15 = unbind_15[0] - k_15 = unbind_15[1] - v_15 = unbind_15[2] - unbind_15 = None - x_433 = torch._C._nn.scaled_dot_product_attention( - q_15, k_15, v_15, dropout_p=0.0 - ) - q_15 = k_15 = v_15 = None - transpose_16 = x_433.transpose(1, 2) - x_433 = None - x_434 = transpose_16.reshape(1, pow_52, 512) - transpose_16 = None - x_435 = torch._C._nn.linear( - x_434, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_434 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_436 = torch.nn.functional.dropout(x_435, 0.0, False, False) - x_435 = None - x_437 = mul_75 + x_436 - mul_75 = x_436 = None - view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_76 = x_437 * view_31 - view_31 = None - layer_norm_64 = torch.nn.functional.layer_norm( - x_437, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_437 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = (None) - x_438 = torch._C._nn.linear( - layer_norm_64, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_64 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_45 = torch.nn.functional.relu(x_438, inplace=False) - x_438 = None - pow_46 = relu_45**2 - relu_45 = None - mul_77 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ - * pow_46 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = ( - pow_46 - ) = None - x_439 = ( - mul_77 - + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ - ) - mul_77 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = (None) - x_440 = torch.nn.functional.dropout(x_439, 0.0, False, False) - x_439 = None - x_441 = torch._C._nn.linear( - x_440, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_440 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_442 = torch.nn.functional.dropout(x_441, 0.0, False, False) - x_441 = None - x_443 = mul_76 + x_442 - mul_76 = x_442 = None - view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_78 = x_443 * view_32 - view_32 = None - layer_norm_65 = torch.nn.functional.layer_norm( - x_443, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_443 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = (None) - linear_64 = torch._C._nn.linear( - layer_norm_65, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_65 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_33 = linear_64.reshape(1, pow_52, 3, 16, 32) - linear_64 = None - qkv_16 = reshape_33.permute(2, 0, 3, 1, 4) - reshape_33 = None - unbind_16 = qkv_16.unbind(0) - qkv_16 = None - q_16 = unbind_16[0] - k_16 = unbind_16[1] - v_16 = unbind_16[2] - unbind_16 = None - x_444 = torch._C._nn.scaled_dot_product_attention( - q_16, k_16, v_16, dropout_p=0.0 - ) - q_16 = k_16 = v_16 = None - transpose_17 = x_444.transpose(1, 2) - x_444 = None - x_445 = transpose_17.reshape(1, pow_52, 512) - transpose_17 = None - x_446 = torch._C._nn.linear( - x_445, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_445 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_447 = torch.nn.functional.dropout(x_446, 0.0, False, False) - x_446 = None - x_448 = mul_78 + x_447 - mul_78 = x_447 = None - view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_79 = x_448 * view_33 - view_33 = None - layer_norm_66 = torch.nn.functional.layer_norm( - x_448, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_448 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = (None) - x_449 = torch._C._nn.linear( - layer_norm_66, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_66 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_46 = torch.nn.functional.relu(x_449, inplace=False) - x_449 = None - pow_47 = relu_46**2 - relu_46 = None - mul_80 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ - * pow_47 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = ( - pow_47 - ) = None - x_450 = ( - mul_80 - + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ - ) - mul_80 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = (None) - x_451 = torch.nn.functional.dropout(x_450, 0.0, False, False) - x_450 = None - x_452 = torch._C._nn.linear( - x_451, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_451 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_453 = torch.nn.functional.dropout(x_452, 0.0, False, False) - x_452 = None - x_454 = mul_79 + x_453 - mul_79 = x_453 = None - view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_81 = x_454 * view_34 - view_34 = None - layer_norm_67 = torch.nn.functional.layer_norm( - x_454, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_454 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = (None) - linear_68 = torch._C._nn.linear( - layer_norm_67, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_67 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_35 = linear_68.reshape(1, pow_52, 3, 16, 32) - linear_68 = None - qkv_17 = reshape_35.permute(2, 0, 3, 1, 4) - reshape_35 = None - unbind_17 = qkv_17.unbind(0) - qkv_17 = None - q_17 = unbind_17[0] - k_17 = unbind_17[1] - v_17 = unbind_17[2] - unbind_17 = None - x_455 = torch._C._nn.scaled_dot_product_attention( - q_17, k_17, v_17, dropout_p=0.0 - ) - q_17 = k_17 = v_17 = None - transpose_18 = x_455.transpose(1, 2) - x_455 = None - x_456 = transpose_18.reshape(1, pow_52, 512) - transpose_18 = pow_52 = None - x_457 = torch._C._nn.linear( - x_456, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_456 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_458 = torch.nn.functional.dropout(x_457, 0.0, False, False) - x_457 = None - x_459 = mul_81 + x_458 - mul_81 = x_458 = None - view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_82 = x_459 * view_35 - view_35 = None - layer_norm_68 = torch.nn.functional.layer_norm( - x_459, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_459 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = (None) - x_460 = torch._C._nn.linear( - layer_norm_68, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_68 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_47 = torch.nn.functional.relu(x_460, inplace=False) - x_460 = None - pow_48 = relu_47**2 - relu_47 = None - mul_83 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ - * pow_48 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = ( - pow_48 - ) = None - x_461 = ( - mul_83 - + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ - ) - mul_83 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = (None) - x_462 = torch.nn.functional.dropout(x_461, 0.0, False, False) - x_461 = None - x_463 = torch._C._nn.linear( - x_462, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_462 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_464 = torch.nn.functional.dropout(x_463, 0.0, False, False) - x_463 = None - x_465 = mul_82 + x_464 - mul_82 = x_464 = None - transpose_19 = x_465.transpose(1, 2) - x_465 = None - x_466 = transpose_19.reshape(1, 512, sym_sum_3, sym_sum_3) - transpose_19 = sym_sum_3 = None - x_467 = x_466.permute(0, 2, 3, 1) - x_466 = None - x_468 = torch.nn.functional.layer_norm( - x_467, - (512,), - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_467 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) - x_469 = x_468.permute(0, 3, 1, 2) - x_468 = None - x_470 = torch.conv2d( - x_469, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_469 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_3 = sym_sum // 32 - sym_sum = None - sym_sum_4 = torch.sym_sum([1, floordiv_3]) - floordiv_3 = None - reshape_38 = x_470.reshape(1, 768, -1) - x_470 = None - x_471 = reshape_38.transpose(1, 2) - reshape_38 = None - view_36 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_84 = x_471 * view_36 - view_36 = None - layer_norm_70 = torch.nn.functional.layer_norm( - x_471, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_471 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_53 = sym_sum_4**2 - linear_72 = torch._C._nn.linear( - layer_norm_70, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_70 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_39 = linear_72.reshape(1, pow_53, 3, 24, 32) - linear_72 = None - qkv_18 = reshape_39.permute(2, 0, 3, 1, 4) - reshape_39 = None - unbind_18 = qkv_18.unbind(0) - qkv_18 = None - q_18 = unbind_18[0] - k_18 = unbind_18[1] - v_18 = unbind_18[2] - unbind_18 = None - x_472 = torch._C._nn.scaled_dot_product_attention( - q_18, k_18, v_18, dropout_p=0.0 - ) - q_18 = k_18 = v_18 = None - transpose_21 = x_472.transpose(1, 2) - x_472 = None - x_473 = transpose_21.reshape(1, pow_53, 768) - transpose_21 = None - x_474 = torch._C._nn.linear( - x_473, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_473 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_475 = torch.nn.functional.dropout(x_474, 0.0, False, False) - x_474 = None - x_476 = mul_84 + x_475 - mul_84 = x_475 = None - view_37 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_85 = x_476 * view_37 - view_37 = None - layer_norm_71 = torch.nn.functional.layer_norm( - x_476, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_476 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_477 = torch._C._nn.linear( - layer_norm_71, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_71 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_48 = torch.nn.functional.relu(x_477, inplace=False) - x_477 = None - pow_49 = relu_48**2 - relu_48 = None - mul_86 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_49 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_49 - ) = None - x_478 = ( - mul_86 - + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_86 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_479 = torch.nn.functional.dropout(x_478, 0.0, False, False) - x_478 = None - x_480 = torch._C._nn.linear( - x_479, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_479 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_481 = torch.nn.functional.dropout(x_480, 0.0, False, False) - x_480 = None - x_482 = mul_85 + x_481 - mul_85 = x_481 = None - view_38 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_87 = x_482 * view_38 - view_38 = None - layer_norm_72 = torch.nn.functional.layer_norm( - x_482, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_482 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_76 = torch._C._nn.linear( - layer_norm_72, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_72 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_41 = linear_76.reshape(1, pow_53, 3, 24, 32) - linear_76 = None - qkv_19 = reshape_41.permute(2, 0, 3, 1, 4) - reshape_41 = None - unbind_19 = qkv_19.unbind(0) - qkv_19 = None - q_19 = unbind_19[0] - k_19 = unbind_19[1] - v_19 = unbind_19[2] - unbind_19 = None - x_483 = torch._C._nn.scaled_dot_product_attention( - q_19, k_19, v_19, dropout_p=0.0 - ) - q_19 = k_19 = v_19 = None - transpose_22 = x_483.transpose(1, 2) - x_483 = None - x_484 = transpose_22.reshape(1, pow_53, 768) - transpose_22 = None - x_485 = torch._C._nn.linear( - x_484, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_484 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_486 = torch.nn.functional.dropout(x_485, 0.0, False, False) - x_485 = None - x_487 = mul_87 + x_486 - mul_87 = x_486 = None - view_39 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_88 = x_487 * view_39 - view_39 = None - layer_norm_73 = torch.nn.functional.layer_norm( - x_487, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_487 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_488 = torch._C._nn.linear( - layer_norm_73, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_73 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_49 = torch.nn.functional.relu(x_488, inplace=False) - x_488 = None - pow_50 = relu_49**2 - relu_49 = None - mul_89 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_50 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_50 - ) = None - x_489 = ( - mul_89 - + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_89 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_490 = torch.nn.functional.dropout(x_489, 0.0, False, False) - x_489 = None - x_491 = torch._C._nn.linear( - x_490, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_490 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_492 = torch.nn.functional.dropout(x_491, 0.0, False, False) - x_491 = None - x_493 = mul_88 + x_492 - mul_88 = x_492 = None - view_40 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_90 = x_493 * view_40 - view_40 = None - layer_norm_74 = torch.nn.functional.layer_norm( - x_493, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_493 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_80 = torch._C._nn.linear( - layer_norm_74, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_74 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_43 = linear_80.reshape(1, pow_53, 3, 24, 32) - linear_80 = None - qkv_20 = reshape_43.permute(2, 0, 3, 1, 4) - reshape_43 = None - unbind_20 = qkv_20.unbind(0) - qkv_20 = None - q_20 = unbind_20[0] - k_20 = unbind_20[1] - v_20 = unbind_20[2] - unbind_20 = None - x_494 = torch._C._nn.scaled_dot_product_attention( - q_20, k_20, v_20, dropout_p=0.0 - ) - q_20 = k_20 = v_20 = None - transpose_23 = x_494.transpose(1, 2) - x_494 = None - x_495 = transpose_23.reshape(1, pow_53, 768) - transpose_23 = pow_53 = None - x_496 = torch._C._nn.linear( - x_495, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_495 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_497 = torch.nn.functional.dropout(x_496, 0.0, False, False) - x_496 = None - x_498 = mul_90 + x_497 - mul_90 = x_497 = None - view_41 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_91 = x_498 * view_41 - view_41 = None - layer_norm_75 = torch.nn.functional.layer_norm( - x_498, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_498 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_499 = torch._C._nn.linear( - layer_norm_75, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_75 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_50 = torch.nn.functional.relu(x_499, inplace=False) - x_499 = None - pow_51 = relu_50**2 - relu_50 = None - mul_92 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_51 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_51 - ) = None - x_500 = ( - mul_92 - + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_92 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_501 = torch.nn.functional.dropout(x_500, 0.0, False, False) - x_500 = None - x_502 = torch._C._nn.linear( - x_501, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_501 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_503 = torch.nn.functional.dropout(x_502, 0.0, False, False) - x_502 = None - x_504 = mul_91 + x_503 - mul_91 = x_503 = None - transpose_24 = x_504.transpose(1, 2) - x_504 = None - x_505 = transpose_24.reshape(1, 768, sym_sum_4, sym_sum_4) - transpose_24 = sym_sum_4 = None - x_506 = torch.nn.functional.adaptive_avg_pool2d(x_505, 1) - x_505 = None - x_507 = x_506.permute(0, 2, 3, 1) - x_506 = None - x_508 = torch.nn.functional.layer_norm( - x_507, - (768,), - l_self_modules_head_modules_norm_parameters_weight_, - l_self_modules_head_modules_norm_parameters_bias_, - 1e-06, - ) - x_507 = ( - l_self_modules_head_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_norm_parameters_bias_ = None - x_509 = x_508.permute(0, 3, 1, 2) - x_508 = None - x_510 = x_509.flatten(1, -1) - x_509 = None - x_511 = torch.nn.functional.dropout(x_510, 0.0, False, False) - x_510 = None - x_512 = torch._C._nn.linear( - x_511, - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, - ) - x_511 = ( - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None - relu_51 = torch.nn.functional.relu(x_512, inplace=False) - x_512 = None - x_513 = torch.square(relu_51) - relu_51 = None - x_514 = torch.nn.functional.layer_norm( - x_513, - (3072,), - l_self_modules_head_modules_fc_modules_norm_parameters_weight_, - l_self_modules_head_modules_fc_modules_norm_parameters_bias_, - 1e-06, - ) - x_513 = ( - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None - x_515 = torch.nn.functional.dropout(x_514, 0.0, False, False) - x_514 = None - x_516 = torch._C._nn.linear( - x_515, - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, - ) - x_515 = ( - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None - return (x_516,) diff --git a/samples/timm/caformer_b36.sail_in1k_384/weight_meta.py b/samples/timm/caformer_b36.sail_in1k_384/weight_meta.py deleted file mode 100644 index 827d4a9370..0000000000 --- a/samples/timm/caformer_b36.sail_in1k_384/weight_meta.py +++ /dev/null @@ -1,3968 +0,0 @@ -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: - name = "L_self_modules_stem_modules_conv_parameters_weight_" - shape = [128, 3, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: - name = "L_self_modules_stem_modules_conv_parameters_bias_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_s1: - name = "s1" - shape = [] - dtype = "torch.int64" - device = "cpu" - mean = None - std = None - data = [4] - - -class Program_weight_tensor_meta_L_x_: - name = "L_x_" - shape = [1, 3, 224, 224] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.226 - std = 1.285 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: - name = "L_self_modules_stem_modules_norm_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [256, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [256, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [128, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [512, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [128, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [256, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [256, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [128, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [512, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [128, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [256, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [256, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [128, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [512, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [128, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" - shape = [256, 128, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" - shape = [512, 256, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" - shape = [768, 512, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [2304, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [768, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [3072, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [768, 3072] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [2304, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [768, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [3072, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [768, 3072] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [2304, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [768, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [3072, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [768, 3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_norm_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_norm_parameters_bias_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" - shape = [3072, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" - shape = [1000, 3072] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" - shape = [1000] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None diff --git a/samples/timm/caformer_b36.sail_in22k/graph_net.json b/samples/timm/caformer_b36.sail_in22k/graph_net.json deleted file mode 100644 index 1373fe3b58..0000000000 --- a/samples/timm/caformer_b36.sail_in22k/graph_net.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "framework": "torch", - "num_devices_required": 1, - "num_nodes_required": 1 -} \ No newline at end of file diff --git a/samples/timm/caformer_b36.sail_in22k/input_meta.py b/samples/timm/caformer_b36.sail_in22k/input_meta.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_b36.sail_in22k/input_tensor_constraints.py b/samples/timm/caformer_b36.sail_in22k/input_tensor_constraints.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_b36.sail_in22k/model.py b/samples/timm/caformer_b36.sail_in22k/model.py deleted file mode 100644 index 08d3de35cc..0000000000 --- a/samples/timm/caformer_b36.sail_in22k/model.py +++ /dev/null @@ -1,4758 +0,0 @@ -import torch - - -class GraphModule(torch.nn.Module): - def forward( - self, - L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - s1: torch.SymInt, - L_x_: torch.Tensor, - L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - ): - l_self_modules_stem_modules_conv_parameters_weight_ = ( - L_self_modules_stem_modules_conv_parameters_weight_ - ) - l_self_modules_stem_modules_conv_parameters_bias_ = ( - L_self_modules_stem_modules_conv_parameters_bias_ - ) - l_x_ = L_x_ - l_self_modules_stem_modules_norm_parameters_weight_ = ( - L_self_modules_stem_modules_norm_parameters_weight_ - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_head_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ - ) - x = torch.conv2d( - l_x_, - l_self_modules_stem_modules_conv_parameters_weight_, - l_self_modules_stem_modules_conv_parameters_bias_, - (4, 4), - (2, 2), - (1, 1), - 1, - ) - l_x_ = ( - l_self_modules_stem_modules_conv_parameters_weight_ - ) = l_self_modules_stem_modules_conv_parameters_bias_ = None - x_1 = x.permute(0, 2, 3, 1) - x = None - x_2 = torch.nn.functional.layer_norm( - x_1, - (128,), - l_self_modules_stem_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None - x_3 = x_2.permute(0, 3, 1, 2) - x_2 = None - sym_sum = torch.sym_sum([-3, s1]) - s1 = None - floordiv = sym_sum // 4 - sym_sum_1 = torch.sym_sum([1, floordiv]) - floordiv = sym_sum_1 = None - x_4 = x_3.permute(0, 2, 3, 1) - x_5 = torch.nn.functional.layer_norm( - x_4, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_6 = x_5.permute(0, 3, 1, 2) - x_5 = None - x_7 = torch.conv2d( - x_6, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu = torch.nn.functional.relu(x_7, inplace=False) - x_7 = None - pow_1 = relu**2 - relu = None - mul = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_1 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_1 - ) = None - x_8 = ( - mul - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_9 = torch.conv2d( - x_8, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 256, - ) - x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_10 = torch.conv2d( - x_9, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_11 = x_3 + x_10 - x_3 = x_10 = None - x_12 = x_11.permute(0, 2, 3, 1) - x_13 = torch.nn.functional.layer_norm( - x_12, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_14 = x_13.permute(0, 3, 1, 2) - x_13 = None - x_15 = torch.conv2d( - x_14, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_1 = torch.nn.functional.relu(x_15, inplace=False) - x_15 = None - pow_2 = relu_1**2 - relu_1 = None - mul_1 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_2 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_2 - ) = None - x_16 = ( - mul_1 - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) - x_16 = None - x_18 = torch.conv2d( - x_17, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) - x_18 = None - x_20 = x_11 + x_19 - x_11 = x_19 = None - x_21 = x_20.permute(0, 2, 3, 1) - x_22 = torch.nn.functional.layer_norm( - x_21, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_23 = x_22.permute(0, 3, 1, 2) - x_22 = None - x_24 = torch.conv2d( - x_23, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_2 = torch.nn.functional.relu(x_24, inplace=False) - x_24 = None - pow_3 = relu_2**2 - relu_2 = None - mul_2 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_3 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_3 - ) = None - x_25 = ( - mul_2 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_26 = torch.conv2d( - x_25, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 256, - ) - x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_27 = torch.conv2d( - x_26, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_28 = x_20 + x_27 - x_20 = x_27 = None - x_29 = x_28.permute(0, 2, 3, 1) - x_30 = torch.nn.functional.layer_norm( - x_29, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_31 = x_30.permute(0, 3, 1, 2) - x_30 = None - x_32 = torch.conv2d( - x_31, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_3 = torch.nn.functional.relu(x_32, inplace=False) - x_32 = None - pow_4 = relu_3**2 - relu_3 = None - mul_3 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_4 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_4 - ) = None - x_33 = ( - mul_3 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) - x_33 = None - x_35 = torch.conv2d( - x_34, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) - x_35 = None - x_37 = x_28 + x_36 - x_28 = x_36 = None - x_38 = x_37.permute(0, 2, 3, 1) - x_39 = torch.nn.functional.layer_norm( - x_38, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_40 = x_39.permute(0, 3, 1, 2) - x_39 = None - x_41 = torch.conv2d( - x_40, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_4 = torch.nn.functional.relu(x_41, inplace=False) - x_41 = None - pow_5 = relu_4**2 - relu_4 = None - mul_4 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_5 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_5 - ) = None - x_42 = ( - mul_4 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_43 = torch.conv2d( - x_42, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 256, - ) - x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_44 = torch.conv2d( - x_43, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_45 = x_37 + x_44 - x_37 = x_44 = None - x_46 = x_45.permute(0, 2, 3, 1) - x_47 = torch.nn.functional.layer_norm( - x_46, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_48 = x_47.permute(0, 3, 1, 2) - x_47 = None - x_49 = torch.conv2d( - x_48, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_5 = torch.nn.functional.relu(x_49, inplace=False) - x_49 = None - pow_6 = relu_5**2 - relu_5 = None - mul_5 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_6 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_6 - ) = None - x_50 = ( - mul_5 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) - x_50 = None - x_52 = torch.conv2d( - x_51, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) - x_52 = None - x_54 = x_45 + x_53 - x_45 = x_53 = None - x_55 = x_54.permute(0, 2, 3, 1) - x_54 = None - x_56 = torch.nn.functional.layer_norm( - x_55, - (128,), - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) - x_57 = x_56.permute(0, 3, 1, 2) - x_56 = None - x_58 = torch.conv2d( - x_57, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_1 = sym_sum // 8 - sym_sum_2 = torch.sym_sum([1, floordiv_1]) - floordiv_1 = sym_sum_2 = None - x_59 = x_58.permute(0, 2, 3, 1) - x_60 = torch.nn.functional.layer_norm( - x_59, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_61 = x_60.permute(0, 3, 1, 2) - x_60 = None - x_62 = torch.conv2d( - x_61, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_6 = torch.nn.functional.relu(x_62, inplace=False) - x_62 = None - pow_7 = relu_6**2 - relu_6 = None - mul_6 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_7 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_7 - ) = None - x_63 = ( - mul_6 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_64 = torch.conv2d( - x_63, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_65 = torch.conv2d( - x_64, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_66 = x_58 + x_65 - x_58 = x_65 = None - x_67 = x_66.permute(0, 2, 3, 1) - x_68 = torch.nn.functional.layer_norm( - x_67, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_69 = x_68.permute(0, 3, 1, 2) - x_68 = None - x_70 = torch.conv2d( - x_69, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_7 = torch.nn.functional.relu(x_70, inplace=False) - x_70 = None - pow_8 = relu_7**2 - relu_7 = None - mul_7 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_8 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_8 - ) = None - x_71 = ( - mul_7 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) - x_71 = None - x_73 = torch.conv2d( - x_72, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) - x_73 = None - x_75 = x_66 + x_74 - x_66 = x_74 = None - x_76 = x_75.permute(0, 2, 3, 1) - x_77 = torch.nn.functional.layer_norm( - x_76, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_78 = x_77.permute(0, 3, 1, 2) - x_77 = None - x_79 = torch.conv2d( - x_78, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_8 = torch.nn.functional.relu(x_79, inplace=False) - x_79 = None - pow_9 = relu_8**2 - relu_8 = None - mul_8 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_9 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_9 - ) = None - x_80 = ( - mul_8 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_81 = torch.conv2d( - x_80, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_82 = torch.conv2d( - x_81, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_83 = x_75 + x_82 - x_75 = x_82 = None - x_84 = x_83.permute(0, 2, 3, 1) - x_85 = torch.nn.functional.layer_norm( - x_84, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_86 = x_85.permute(0, 3, 1, 2) - x_85 = None - x_87 = torch.conv2d( - x_86, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_9 = torch.nn.functional.relu(x_87, inplace=False) - x_87 = None - pow_10 = relu_9**2 - relu_9 = None - mul_9 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_10 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_10 - ) = None - x_88 = ( - mul_9 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) - x_88 = None - x_90 = torch.conv2d( - x_89, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) - x_90 = None - x_92 = x_83 + x_91 - x_83 = x_91 = None - x_93 = x_92.permute(0, 2, 3, 1) - x_94 = torch.nn.functional.layer_norm( - x_93, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_95 = x_94.permute(0, 3, 1, 2) - x_94 = None - x_96 = torch.conv2d( - x_95, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_10 = torch.nn.functional.relu(x_96, inplace=False) - x_96 = None - pow_11 = relu_10**2 - relu_10 = None - mul_10 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_11 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_11 - ) = None - x_97 = ( - mul_10 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_98 = torch.conv2d( - x_97, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_99 = torch.conv2d( - x_98, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_100 = x_92 + x_99 - x_92 = x_99 = None - x_101 = x_100.permute(0, 2, 3, 1) - x_102 = torch.nn.functional.layer_norm( - x_101, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_103 = x_102.permute(0, 3, 1, 2) - x_102 = None - x_104 = torch.conv2d( - x_103, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_11 = torch.nn.functional.relu(x_104, inplace=False) - x_104 = None - pow_12 = relu_11**2 - relu_11 = None - mul_11 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_12 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_12 - ) = None - x_105 = ( - mul_11 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) - x_105 = None - x_107 = torch.conv2d( - x_106, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) - x_107 = None - x_109 = x_100 + x_108 - x_100 = x_108 = None - x_110 = x_109.permute(0, 2, 3, 1) - x_111 = torch.nn.functional.layer_norm( - x_110, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_110 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) - x_112 = x_111.permute(0, 3, 1, 2) - x_111 = None - x_113 = torch.conv2d( - x_112, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_112 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_12 = torch.nn.functional.relu(x_113, inplace=False) - x_113 = None - pow_13 = relu_12**2 - relu_12 = None - mul_12 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ - * pow_13 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_13 - ) = None - x_114 = ( - mul_12 - + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_12 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_115 = torch.conv2d( - x_114, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_114 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_116 = torch.conv2d( - x_115, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_115 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_117 = x_109 + x_116 - x_109 = x_116 = None - x_118 = x_117.permute(0, 2, 3, 1) - x_119 = torch.nn.functional.layer_norm( - x_118, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_118 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) - x_120 = x_119.permute(0, 3, 1, 2) - x_119 = None - x_121 = torch.conv2d( - x_120, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_120 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_13 = torch.nn.functional.relu(x_121, inplace=False) - x_121 = None - pow_14 = relu_13**2 - relu_13 = None - mul_13 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - * pow_14 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( - pow_14 - ) = None - x_122 = ( - mul_13 - + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - ) - mul_13 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) - x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) - x_122 = None - x_124 = torch.conv2d( - x_123, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_123 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) - x_124 = None - x_126 = x_117 + x_125 - x_117 = x_125 = None - x_127 = x_126.permute(0, 2, 3, 1) - x_128 = torch.nn.functional.layer_norm( - x_127, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_127 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) - x_129 = x_128.permute(0, 3, 1, 2) - x_128 = None - x_130 = torch.conv2d( - x_129, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_129 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_14 = torch.nn.functional.relu(x_130, inplace=False) - x_130 = None - pow_15 = relu_14**2 - relu_14 = None - mul_14 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ - * pow_15 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_15 - ) = None - x_131 = ( - mul_14 - + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_14 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_132 = torch.conv2d( - x_131, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_131 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_133 = torch.conv2d( - x_132, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_132 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_134 = x_126 + x_133 - x_126 = x_133 = None - x_135 = x_134.permute(0, 2, 3, 1) - x_136 = torch.nn.functional.layer_norm( - x_135, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_135 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) - x_137 = x_136.permute(0, 3, 1, 2) - x_136 = None - x_138 = torch.conv2d( - x_137, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_137 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_15 = torch.nn.functional.relu(x_138, inplace=False) - x_138 = None - pow_16 = relu_15**2 - relu_15 = None - mul_15 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - * pow_16 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( - pow_16 - ) = None - x_139 = ( - mul_15 - + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - ) - mul_15 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) - x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) - x_139 = None - x_141 = torch.conv2d( - x_140, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_140 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) - x_141 = None - x_143 = x_134 + x_142 - x_134 = x_142 = None - x_144 = x_143.permute(0, 2, 3, 1) - x_145 = torch.nn.functional.layer_norm( - x_144, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_144 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) - x_146 = x_145.permute(0, 3, 1, 2) - x_145 = None - x_147 = torch.conv2d( - x_146, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_146 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_16 = torch.nn.functional.relu(x_147, inplace=False) - x_147 = None - pow_17 = relu_16**2 - relu_16 = None - mul_16 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ - * pow_17 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_17 - ) = None - x_148 = ( - mul_16 - + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_16 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_149 = torch.conv2d( - x_148, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_148 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_150 = torch.conv2d( - x_149, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_149 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_151 = x_143 + x_150 - x_143 = x_150 = None - x_152 = x_151.permute(0, 2, 3, 1) - x_153 = torch.nn.functional.layer_norm( - x_152, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_152 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) - x_154 = x_153.permute(0, 3, 1, 2) - x_153 = None - x_155 = torch.conv2d( - x_154, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_154 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_17 = torch.nn.functional.relu(x_155, inplace=False) - x_155 = None - pow_18 = relu_17**2 - relu_17 = None - mul_17 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - * pow_18 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( - pow_18 - ) = None - x_156 = ( - mul_17 - + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - ) - mul_17 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) - x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) - x_156 = None - x_158 = torch.conv2d( - x_157, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_157 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) - x_158 = None - x_160 = x_151 + x_159 - x_151 = x_159 = None - x_161 = x_160.permute(0, 2, 3, 1) - x_162 = torch.nn.functional.layer_norm( - x_161, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_161 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) - x_163 = x_162.permute(0, 3, 1, 2) - x_162 = None - x_164 = torch.conv2d( - x_163, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_163 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_18 = torch.nn.functional.relu(x_164, inplace=False) - x_164 = None - pow_19 = relu_18**2 - relu_18 = None - mul_18 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ - * pow_19 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_19 - ) = None - x_165 = ( - mul_18 - + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_18 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_166 = torch.conv2d( - x_165, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_165 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_167 = torch.conv2d( - x_166, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_166 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_168 = x_160 + x_167 - x_160 = x_167 = None - x_169 = x_168.permute(0, 2, 3, 1) - x_170 = torch.nn.functional.layer_norm( - x_169, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_169 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) - x_171 = x_170.permute(0, 3, 1, 2) - x_170 = None - x_172 = torch.conv2d( - x_171, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_171 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_19 = torch.nn.functional.relu(x_172, inplace=False) - x_172 = None - pow_20 = relu_19**2 - relu_19 = None - mul_19 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - * pow_20 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( - pow_20 - ) = None - x_173 = ( - mul_19 - + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - ) - mul_19 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) - x_174 = torch.nn.functional.dropout(x_173, 0.0, False, False) - x_173 = None - x_175 = torch.conv2d( - x_174, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_174 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) - x_175 = None - x_177 = x_168 + x_176 - x_168 = x_176 = None - x_178 = x_177.permute(0, 2, 3, 1) - x_179 = torch.nn.functional.layer_norm( - x_178, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_178 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) - x_180 = x_179.permute(0, 3, 1, 2) - x_179 = None - x_181 = torch.conv2d( - x_180, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_180 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_20 = torch.nn.functional.relu(x_181, inplace=False) - x_181 = None - pow_21 = relu_20**2 - relu_20 = None - mul_20 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ - * pow_21 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_21 - ) = None - x_182 = ( - mul_20 - + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_20 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_183 = torch.conv2d( - x_182, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_182 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_184 = torch.conv2d( - x_183, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_183 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_185 = x_177 + x_184 - x_177 = x_184 = None - x_186 = x_185.permute(0, 2, 3, 1) - x_187 = torch.nn.functional.layer_norm( - x_186, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_186 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) - x_188 = x_187.permute(0, 3, 1, 2) - x_187 = None - x_189 = torch.conv2d( - x_188, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_188 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_21 = torch.nn.functional.relu(x_189, inplace=False) - x_189 = None - pow_22 = relu_21**2 - relu_21 = None - mul_21 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - * pow_22 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( - pow_22 - ) = None - x_190 = ( - mul_21 - + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - ) - mul_21 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) - x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) - x_190 = None - x_192 = torch.conv2d( - x_191, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_191 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) - x_192 = None - x_194 = x_185 + x_193 - x_185 = x_193 = None - x_195 = x_194.permute(0, 2, 3, 1) - x_196 = torch.nn.functional.layer_norm( - x_195, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_195 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) - x_197 = x_196.permute(0, 3, 1, 2) - x_196 = None - x_198 = torch.conv2d( - x_197, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_197 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_22 = torch.nn.functional.relu(x_198, inplace=False) - x_198 = None - pow_23 = relu_22**2 - relu_22 = None - mul_22 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ - * pow_23 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_23 - ) = None - x_199 = ( - mul_22 - + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_22 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_200 = torch.conv2d( - x_199, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_199 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_201 = torch.conv2d( - x_200, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_200 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_202 = x_194 + x_201 - x_194 = x_201 = None - x_203 = x_202.permute(0, 2, 3, 1) - x_204 = torch.nn.functional.layer_norm( - x_203, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_203 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) - x_205 = x_204.permute(0, 3, 1, 2) - x_204 = None - x_206 = torch.conv2d( - x_205, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_205 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_23 = torch.nn.functional.relu(x_206, inplace=False) - x_206 = None - pow_24 = relu_23**2 - relu_23 = None - mul_23 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - * pow_24 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( - pow_24 - ) = None - x_207 = ( - mul_23 - + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - ) - mul_23 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) - x_208 = torch.nn.functional.dropout(x_207, 0.0, False, False) - x_207 = None - x_209 = torch.conv2d( - x_208, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_208 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) - x_209 = None - x_211 = x_202 + x_210 - x_202 = x_210 = None - x_212 = x_211.permute(0, 2, 3, 1) - x_213 = torch.nn.functional.layer_norm( - x_212, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_212 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) - x_214 = x_213.permute(0, 3, 1, 2) - x_213 = None - x_215 = torch.conv2d( - x_214, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_214 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_24 = torch.nn.functional.relu(x_215, inplace=False) - x_215 = None - pow_25 = relu_24**2 - relu_24 = None - mul_24 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ - * pow_25 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_25 - ) = None - x_216 = ( - mul_24 - + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_24 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_217 = torch.conv2d( - x_216, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_216 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_218 = torch.conv2d( - x_217, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_217 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_219 = x_211 + x_218 - x_211 = x_218 = None - x_220 = x_219.permute(0, 2, 3, 1) - x_221 = torch.nn.functional.layer_norm( - x_220, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_220 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) - x_222 = x_221.permute(0, 3, 1, 2) - x_221 = None - x_223 = torch.conv2d( - x_222, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_222 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_25 = torch.nn.functional.relu(x_223, inplace=False) - x_223 = None - pow_26 = relu_25**2 - relu_25 = None - mul_25 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - * pow_26 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( - pow_26 - ) = None - x_224 = ( - mul_25 - + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - ) - mul_25 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) - x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) - x_224 = None - x_226 = torch.conv2d( - x_225, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_225 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) - x_226 = None - x_228 = x_219 + x_227 - x_219 = x_227 = None - x_229 = x_228.permute(0, 2, 3, 1) - x_230 = torch.nn.functional.layer_norm( - x_229, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_229 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) - x_231 = x_230.permute(0, 3, 1, 2) - x_230 = None - x_232 = torch.conv2d( - x_231, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_231 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_26 = torch.nn.functional.relu(x_232, inplace=False) - x_232 = None - pow_27 = relu_26**2 - relu_26 = None - mul_26 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ - * pow_27 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_27 - ) = None - x_233 = ( - mul_26 - + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_26 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_234 = torch.conv2d( - x_233, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_233 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_235 = torch.conv2d( - x_234, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_234 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_236 = x_228 + x_235 - x_228 = x_235 = None - x_237 = x_236.permute(0, 2, 3, 1) - x_238 = torch.nn.functional.layer_norm( - x_237, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_237 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) - x_239 = x_238.permute(0, 3, 1, 2) - x_238 = None - x_240 = torch.conv2d( - x_239, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_239 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_27 = torch.nn.functional.relu(x_240, inplace=False) - x_240 = None - pow_28 = relu_27**2 - relu_27 = None - mul_27 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - * pow_28 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( - pow_28 - ) = None - x_241 = ( - mul_27 - + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - ) - mul_27 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) - x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) - x_241 = None - x_243 = torch.conv2d( - x_242, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_242 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_244 = torch.nn.functional.dropout(x_243, 0.0, False, False) - x_243 = None - x_245 = x_236 + x_244 - x_236 = x_244 = None - x_246 = x_245.permute(0, 2, 3, 1) - x_247 = torch.nn.functional.layer_norm( - x_246, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_246 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) - x_248 = x_247.permute(0, 3, 1, 2) - x_247 = None - x_249 = torch.conv2d( - x_248, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_248 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_28 = torch.nn.functional.relu(x_249, inplace=False) - x_249 = None - pow_29 = relu_28**2 - relu_28 = None - mul_28 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ - * pow_29 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_29 - ) = None - x_250 = ( - mul_28 - + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_28 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_251 = torch.conv2d( - x_250, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_250 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_252 = torch.conv2d( - x_251, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_251 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_253 = x_245 + x_252 - x_245 = x_252 = None - x_254 = x_253.permute(0, 2, 3, 1) - x_255 = torch.nn.functional.layer_norm( - x_254, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_254 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) - x_256 = x_255.permute(0, 3, 1, 2) - x_255 = None - x_257 = torch.conv2d( - x_256, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_256 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_29 = torch.nn.functional.relu(x_257, inplace=False) - x_257 = None - pow_30 = relu_29**2 - relu_29 = None - mul_29 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - * pow_30 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( - pow_30 - ) = None - x_258 = ( - mul_29 - + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - ) - mul_29 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) - x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) - x_258 = None - x_260 = torch.conv2d( - x_259, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_259 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_261 = torch.nn.functional.dropout(x_260, 0.0, False, False) - x_260 = None - x_262 = x_253 + x_261 - x_253 = x_261 = None - x_263 = x_262.permute(0, 2, 3, 1) - x_262 = None - x_264 = torch.nn.functional.layer_norm( - x_263, - (256,), - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_263 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) - x_265 = x_264.permute(0, 3, 1, 2) - x_264 = None - x_266 = torch.conv2d( - x_265, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_265 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_2 = sym_sum // 16 - sym_sum_3 = torch.sym_sum([1, floordiv_2]) - floordiv_2 = None - reshape = x_266.reshape(1, 512, -1) - x_266 = None - x_267 = reshape.transpose(1, 2) - reshape = None - view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_30 = x_267 * view - view = None - layer_norm_33 = torch.nn.functional.layer_norm( - x_267, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_52 = sym_sum_3**2 - linear = torch._C._nn.linear( - layer_norm_33, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_33 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_1 = linear.reshape(1, pow_52, 3, 16, 32) - linear = None - qkv = reshape_1.permute(2, 0, 3, 1, 4) - reshape_1 = None - unbind = qkv.unbind(0) - qkv = None - q = unbind[0] - k = unbind[1] - v = unbind[2] - unbind = None - x_268 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) - q = k = v = None - transpose_1 = x_268.transpose(1, 2) - x_268 = None - x_269 = transpose_1.reshape(1, pow_52, 512) - transpose_1 = None - x_270 = torch._C._nn.linear( - x_269, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_269 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_271 = torch.nn.functional.dropout(x_270, 0.0, False, False) - x_270 = None - x_272 = mul_30 + x_271 - mul_30 = x_271 = None - view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_31 = x_272 * view_1 - view_1 = None - layer_norm_34 = torch.nn.functional.layer_norm( - x_272, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_272 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_273 = torch._C._nn.linear( - layer_norm_34, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_34 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_30 = torch.nn.functional.relu(x_273, inplace=False) - x_273 = None - pow_31 = relu_30**2 - relu_30 = None - mul_32 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_31 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_31 - ) = None - x_274 = ( - mul_32 - + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_275 = torch.nn.functional.dropout(x_274, 0.0, False, False) - x_274 = None - x_276 = torch._C._nn.linear( - x_275, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_277 = torch.nn.functional.dropout(x_276, 0.0, False, False) - x_276 = None - x_278 = mul_31 + x_277 - mul_31 = x_277 = None - view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_33 = x_278 * view_2 - view_2 = None - layer_norm_35 = torch.nn.functional.layer_norm( - x_278, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_4 = torch._C._nn.linear( - layer_norm_35, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_3 = linear_4.reshape(1, pow_52, 3, 16, 32) - linear_4 = None - qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) - reshape_3 = None - unbind_1 = qkv_1.unbind(0) - qkv_1 = None - q_1 = unbind_1[0] - k_1 = unbind_1[1] - v_1 = unbind_1[2] - unbind_1 = None - x_279 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) - q_1 = k_1 = v_1 = None - transpose_2 = x_279.transpose(1, 2) - x_279 = None - x_280 = transpose_2.reshape(1, pow_52, 512) - transpose_2 = None - x_281 = torch._C._nn.linear( - x_280, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_280 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) - x_281 = None - x_283 = mul_33 + x_282 - mul_33 = x_282 = None - view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_34 = x_283 * view_3 - view_3 = None - layer_norm_36 = torch.nn.functional.layer_norm( - x_283, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_283 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_284 = torch._C._nn.linear( - layer_norm_36, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_36 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_31 = torch.nn.functional.relu(x_284, inplace=False) - x_284 = None - pow_32 = relu_31**2 - relu_31 = None - mul_35 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_32 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_32 - ) = None - x_285 = ( - mul_35 - + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_286 = torch.nn.functional.dropout(x_285, 0.0, False, False) - x_285 = None - x_287 = torch._C._nn.linear( - x_286, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) - x_287 = None - x_289 = mul_34 + x_288 - mul_34 = x_288 = None - view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_36 = x_289 * view_4 - view_4 = None - layer_norm_37 = torch.nn.functional.layer_norm( - x_289, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_8 = torch._C._nn.linear( - layer_norm_37, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_37 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_5 = linear_8.reshape(1, pow_52, 3, 16, 32) - linear_8 = None - qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) - reshape_5 = None - unbind_2 = qkv_2.unbind(0) - qkv_2 = None - q_2 = unbind_2[0] - k_2 = unbind_2[1] - v_2 = unbind_2[2] - unbind_2 = None - x_290 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) - q_2 = k_2 = v_2 = None - transpose_3 = x_290.transpose(1, 2) - x_290 = None - x_291 = transpose_3.reshape(1, pow_52, 512) - transpose_3 = None - x_292 = torch._C._nn.linear( - x_291, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_291 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_293 = torch.nn.functional.dropout(x_292, 0.0, False, False) - x_292 = None - x_294 = mul_36 + x_293 - mul_36 = x_293 = None - view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_37 = x_294 * view_5 - view_5 = None - layer_norm_38 = torch.nn.functional.layer_norm( - x_294, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_294 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_295 = torch._C._nn.linear( - layer_norm_38, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_32 = torch.nn.functional.relu(x_295, inplace=False) - x_295 = None - pow_33 = relu_32**2 - relu_32 = None - mul_38 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_33 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_33 - ) = None - x_296 = ( - mul_38 - + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_297 = torch.nn.functional.dropout(x_296, 0.0, False, False) - x_296 = None - x_298 = torch._C._nn.linear( - x_297, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_299 = torch.nn.functional.dropout(x_298, 0.0, False, False) - x_298 = None - x_300 = mul_37 + x_299 - mul_37 = x_299 = None - view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_39 = x_300 * view_6 - view_6 = None - layer_norm_39 = torch.nn.functional.layer_norm( - x_300, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) - linear_12 = torch._C._nn.linear( - layer_norm_39, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_39 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_7 = linear_12.reshape(1, pow_52, 3, 16, 32) - linear_12 = None - qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) - reshape_7 = None - unbind_3 = qkv_3.unbind(0) - qkv_3 = None - q_3 = unbind_3[0] - k_3 = unbind_3[1] - v_3 = unbind_3[2] - unbind_3 = None - x_301 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) - q_3 = k_3 = v_3 = None - transpose_4 = x_301.transpose(1, 2) - x_301 = None - x_302 = transpose_4.reshape(1, pow_52, 512) - transpose_4 = None - x_303 = torch._C._nn.linear( - x_302, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_302 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_304 = torch.nn.functional.dropout(x_303, 0.0, False, False) - x_303 = None - x_305 = mul_39 + x_304 - mul_39 = x_304 = None - view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_40 = x_305 * view_7 - view_7 = None - layer_norm_40 = torch.nn.functional.layer_norm( - x_305, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_305 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) - x_306 = torch._C._nn.linear( - layer_norm_40, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_40 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_33 = torch.nn.functional.relu(x_306, inplace=False) - x_306 = None - pow_34 = relu_33**2 - relu_33 = None - mul_41 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - * pow_34 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( - pow_34 - ) = None - x_307 = ( - mul_41 - + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - ) - mul_41 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) - x_308 = torch.nn.functional.dropout(x_307, 0.0, False, False) - x_307 = None - x_309 = torch._C._nn.linear( - x_308, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) - x_309 = None - x_311 = mul_40 + x_310 - mul_40 = x_310 = None - view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_42 = x_311 * view_8 - view_8 = None - layer_norm_41 = torch.nn.functional.layer_norm( - x_311, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) - linear_16 = torch._C._nn.linear( - layer_norm_41, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_41 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_9 = linear_16.reshape(1, pow_52, 3, 16, 32) - linear_16 = None - qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) - reshape_9 = None - unbind_4 = qkv_4.unbind(0) - qkv_4 = None - q_4 = unbind_4[0] - k_4 = unbind_4[1] - v_4 = unbind_4[2] - unbind_4 = None - x_312 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) - q_4 = k_4 = v_4 = None - transpose_5 = x_312.transpose(1, 2) - x_312 = None - x_313 = transpose_5.reshape(1, pow_52, 512) - transpose_5 = None - x_314 = torch._C._nn.linear( - x_313, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_313 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) - x_314 = None - x_316 = mul_42 + x_315 - mul_42 = x_315 = None - view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_43 = x_316 * view_9 - view_9 = None - layer_norm_42 = torch.nn.functional.layer_norm( - x_316, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_316 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) - x_317 = torch._C._nn.linear( - layer_norm_42, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_42 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_34 = torch.nn.functional.relu(x_317, inplace=False) - x_317 = None - pow_35 = relu_34**2 - relu_34 = None - mul_44 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - * pow_35 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( - pow_35 - ) = None - x_318 = ( - mul_44 - + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - ) - mul_44 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) - x_319 = torch.nn.functional.dropout(x_318, 0.0, False, False) - x_318 = None - x_320 = torch._C._nn.linear( - x_319, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_321 = torch.nn.functional.dropout(x_320, 0.0, False, False) - x_320 = None - x_322 = mul_43 + x_321 - mul_43 = x_321 = None - view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_45 = x_322 * view_10 - view_10 = None - layer_norm_43 = torch.nn.functional.layer_norm( - x_322, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) - linear_20 = torch._C._nn.linear( - layer_norm_43, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_43 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_11 = linear_20.reshape(1, pow_52, 3, 16, 32) - linear_20 = None - qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) - reshape_11 = None - unbind_5 = qkv_5.unbind(0) - qkv_5 = None - q_5 = unbind_5[0] - k_5 = unbind_5[1] - v_5 = unbind_5[2] - unbind_5 = None - x_323 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) - q_5 = k_5 = v_5 = None - transpose_6 = x_323.transpose(1, 2) - x_323 = None - x_324 = transpose_6.reshape(1, pow_52, 512) - transpose_6 = None - x_325 = torch._C._nn.linear( - x_324, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_324 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_326 = torch.nn.functional.dropout(x_325, 0.0, False, False) - x_325 = None - x_327 = mul_45 + x_326 - mul_45 = x_326 = None - view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_46 = x_327 * view_11 - view_11 = None - layer_norm_44 = torch.nn.functional.layer_norm( - x_327, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_327 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) - x_328 = torch._C._nn.linear( - layer_norm_44, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_44 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_35 = torch.nn.functional.relu(x_328, inplace=False) - x_328 = None - pow_36 = relu_35**2 - relu_35 = None - mul_47 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - * pow_36 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( - pow_36 - ) = None - x_329 = ( - mul_47 - + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - ) - mul_47 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) - x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) - x_329 = None - x_331 = torch._C._nn.linear( - x_330, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_332 = torch.nn.functional.dropout(x_331, 0.0, False, False) - x_331 = None - x_333 = mul_46 + x_332 - mul_46 = x_332 = None - view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_48 = x_333 * view_12 - view_12 = None - layer_norm_45 = torch.nn.functional.layer_norm( - x_333, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) - linear_24 = torch._C._nn.linear( - layer_norm_45, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_45 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_13 = linear_24.reshape(1, pow_52, 3, 16, 32) - linear_24 = None - qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) - reshape_13 = None - unbind_6 = qkv_6.unbind(0) - qkv_6 = None - q_6 = unbind_6[0] - k_6 = unbind_6[1] - v_6 = unbind_6[2] - unbind_6 = None - x_334 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) - q_6 = k_6 = v_6 = None - transpose_7 = x_334.transpose(1, 2) - x_334 = None - x_335 = transpose_7.reshape(1, pow_52, 512) - transpose_7 = None - x_336 = torch._C._nn.linear( - x_335, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_335 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_337 = torch.nn.functional.dropout(x_336, 0.0, False, False) - x_336 = None - x_338 = mul_48 + x_337 - mul_48 = x_337 = None - view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_49 = x_338 * view_13 - view_13 = None - layer_norm_46 = torch.nn.functional.layer_norm( - x_338, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_338 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) - x_339 = torch._C._nn.linear( - layer_norm_46, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_46 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_36 = torch.nn.functional.relu(x_339, inplace=False) - x_339 = None - pow_37 = relu_36**2 - relu_36 = None - mul_50 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - * pow_37 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( - pow_37 - ) = None - x_340 = ( - mul_50 - + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - ) - mul_50 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) - x_341 = torch.nn.functional.dropout(x_340, 0.0, False, False) - x_340 = None - x_342 = torch._C._nn.linear( - x_341, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_343 = torch.nn.functional.dropout(x_342, 0.0, False, False) - x_342 = None - x_344 = mul_49 + x_343 - mul_49 = x_343 = None - view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_51 = x_344 * view_14 - view_14 = None - layer_norm_47 = torch.nn.functional.layer_norm( - x_344, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) - linear_28 = torch._C._nn.linear( - layer_norm_47, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_47 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_15 = linear_28.reshape(1, pow_52, 3, 16, 32) - linear_28 = None - qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) - reshape_15 = None - unbind_7 = qkv_7.unbind(0) - qkv_7 = None - q_7 = unbind_7[0] - k_7 = unbind_7[1] - v_7 = unbind_7[2] - unbind_7 = None - x_345 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) - q_7 = k_7 = v_7 = None - transpose_8 = x_345.transpose(1, 2) - x_345 = None - x_346 = transpose_8.reshape(1, pow_52, 512) - transpose_8 = None - x_347 = torch._C._nn.linear( - x_346, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_346 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) - x_347 = None - x_349 = mul_51 + x_348 - mul_51 = x_348 = None - view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_52 = x_349 * view_15 - view_15 = None - layer_norm_48 = torch.nn.functional.layer_norm( - x_349, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_349 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) - x_350 = torch._C._nn.linear( - layer_norm_48, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_48 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_37 = torch.nn.functional.relu(x_350, inplace=False) - x_350 = None - pow_38 = relu_37**2 - relu_37 = None - mul_53 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - * pow_38 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( - pow_38 - ) = None - x_351 = ( - mul_53 - + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - ) - mul_53 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) - x_352 = torch.nn.functional.dropout(x_351, 0.0, False, False) - x_351 = None - x_353 = torch._C._nn.linear( - x_352, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_354 = torch.nn.functional.dropout(x_353, 0.0, False, False) - x_353 = None - x_355 = mul_52 + x_354 - mul_52 = x_354 = None - view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_54 = x_355 * view_16 - view_16 = None - layer_norm_49 = torch.nn.functional.layer_norm( - x_355, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) - linear_32 = torch._C._nn.linear( - layer_norm_49, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_49 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_17 = linear_32.reshape(1, pow_52, 3, 16, 32) - linear_32 = None - qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) - reshape_17 = None - unbind_8 = qkv_8.unbind(0) - qkv_8 = None - q_8 = unbind_8[0] - k_8 = unbind_8[1] - v_8 = unbind_8[2] - unbind_8 = None - x_356 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) - q_8 = k_8 = v_8 = None - transpose_9 = x_356.transpose(1, 2) - x_356 = None - x_357 = transpose_9.reshape(1, pow_52, 512) - transpose_9 = None - x_358 = torch._C._nn.linear( - x_357, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_357 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_359 = torch.nn.functional.dropout(x_358, 0.0, False, False) - x_358 = None - x_360 = mul_54 + x_359 - mul_54 = x_359 = None - view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_55 = x_360 * view_17 - view_17 = None - layer_norm_50 = torch.nn.functional.layer_norm( - x_360, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_360 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) - x_361 = torch._C._nn.linear( - layer_norm_50, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_50 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_38 = torch.nn.functional.relu(x_361, inplace=False) - x_361 = None - pow_39 = relu_38**2 - relu_38 = None - mul_56 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - * pow_39 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( - pow_39 - ) = None - x_362 = ( - mul_56 - + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - ) - mul_56 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) - x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) - x_362 = None - x_364 = torch._C._nn.linear( - x_363, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_365 = torch.nn.functional.dropout(x_364, 0.0, False, False) - x_364 = None - x_366 = mul_55 + x_365 - mul_55 = x_365 = None - view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_57 = x_366 * view_18 - view_18 = None - layer_norm_51 = torch.nn.functional.layer_norm( - x_366, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) - linear_36 = torch._C._nn.linear( - layer_norm_51, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_51 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_19 = linear_36.reshape(1, pow_52, 3, 16, 32) - linear_36 = None - qkv_9 = reshape_19.permute(2, 0, 3, 1, 4) - reshape_19 = None - unbind_9 = qkv_9.unbind(0) - qkv_9 = None - q_9 = unbind_9[0] - k_9 = unbind_9[1] - v_9 = unbind_9[2] - unbind_9 = None - x_367 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) - q_9 = k_9 = v_9 = None - transpose_10 = x_367.transpose(1, 2) - x_367 = None - x_368 = transpose_10.reshape(1, pow_52, 512) - transpose_10 = None - x_369 = torch._C._nn.linear( - x_368, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_368 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_370 = torch.nn.functional.dropout(x_369, 0.0, False, False) - x_369 = None - x_371 = mul_57 + x_370 - mul_57 = x_370 = None - view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_58 = x_371 * view_19 - view_19 = None - layer_norm_52 = torch.nn.functional.layer_norm( - x_371, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_371 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) - x_372 = torch._C._nn.linear( - layer_norm_52, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_52 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_39 = torch.nn.functional.relu(x_372, inplace=False) - x_372 = None - pow_40 = relu_39**2 - relu_39 = None - mul_59 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - * pow_40 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( - pow_40 - ) = None - x_373 = ( - mul_59 - + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - ) - mul_59 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) - x_374 = torch.nn.functional.dropout(x_373, 0.0, False, False) - x_373 = None - x_375 = torch._C._nn.linear( - x_374, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_374 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_376 = torch.nn.functional.dropout(x_375, 0.0, False, False) - x_375 = None - x_377 = mul_58 + x_376 - mul_58 = x_376 = None - view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_60 = x_377 * view_20 - view_20 = None - layer_norm_53 = torch.nn.functional.layer_norm( - x_377, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_377 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) - linear_40 = torch._C._nn.linear( - layer_norm_53, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_53 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_21 = linear_40.reshape(1, pow_52, 3, 16, 32) - linear_40 = None - qkv_10 = reshape_21.permute(2, 0, 3, 1, 4) - reshape_21 = None - unbind_10 = qkv_10.unbind(0) - qkv_10 = None - q_10 = unbind_10[0] - k_10 = unbind_10[1] - v_10 = unbind_10[2] - unbind_10 = None - x_378 = torch._C._nn.scaled_dot_product_attention( - q_10, k_10, v_10, dropout_p=0.0 - ) - q_10 = k_10 = v_10 = None - transpose_11 = x_378.transpose(1, 2) - x_378 = None - x_379 = transpose_11.reshape(1, pow_52, 512) - transpose_11 = None - x_380 = torch._C._nn.linear( - x_379, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_379 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_381 = torch.nn.functional.dropout(x_380, 0.0, False, False) - x_380 = None - x_382 = mul_60 + x_381 - mul_60 = x_381 = None - view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_61 = x_382 * view_21 - view_21 = None - layer_norm_54 = torch.nn.functional.layer_norm( - x_382, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_382 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) - x_383 = torch._C._nn.linear( - layer_norm_54, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_54 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_40 = torch.nn.functional.relu(x_383, inplace=False) - x_383 = None - pow_41 = relu_40**2 - relu_40 = None - mul_62 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - * pow_41 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( - pow_41 - ) = None - x_384 = ( - mul_62 - + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - ) - mul_62 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) - x_385 = torch.nn.functional.dropout(x_384, 0.0, False, False) - x_384 = None - x_386 = torch._C._nn.linear( - x_385, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_385 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_387 = torch.nn.functional.dropout(x_386, 0.0, False, False) - x_386 = None - x_388 = mul_61 + x_387 - mul_61 = x_387 = None - view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_63 = x_388 * view_22 - view_22 = None - layer_norm_55 = torch.nn.functional.layer_norm( - x_388, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_388 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) - linear_44 = torch._C._nn.linear( - layer_norm_55, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_55 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_23 = linear_44.reshape(1, pow_52, 3, 16, 32) - linear_44 = None - qkv_11 = reshape_23.permute(2, 0, 3, 1, 4) - reshape_23 = None - unbind_11 = qkv_11.unbind(0) - qkv_11 = None - q_11 = unbind_11[0] - k_11 = unbind_11[1] - v_11 = unbind_11[2] - unbind_11 = None - x_389 = torch._C._nn.scaled_dot_product_attention( - q_11, k_11, v_11, dropout_p=0.0 - ) - q_11 = k_11 = v_11 = None - transpose_12 = x_389.transpose(1, 2) - x_389 = None - x_390 = transpose_12.reshape(1, pow_52, 512) - transpose_12 = None - x_391 = torch._C._nn.linear( - x_390, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_390 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_392 = torch.nn.functional.dropout(x_391, 0.0, False, False) - x_391 = None - x_393 = mul_63 + x_392 - mul_63 = x_392 = None - view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_64 = x_393 * view_23 - view_23 = None - layer_norm_56 = torch.nn.functional.layer_norm( - x_393, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_393 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) - x_394 = torch._C._nn.linear( - layer_norm_56, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_56 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_41 = torch.nn.functional.relu(x_394, inplace=False) - x_394 = None - pow_42 = relu_41**2 - relu_41 = None - mul_65 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - * pow_42 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( - pow_42 - ) = None - x_395 = ( - mul_65 - + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - ) - mul_65 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) - x_396 = torch.nn.functional.dropout(x_395, 0.0, False, False) - x_395 = None - x_397 = torch._C._nn.linear( - x_396, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_396 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_398 = torch.nn.functional.dropout(x_397, 0.0, False, False) - x_397 = None - x_399 = mul_64 + x_398 - mul_64 = x_398 = None - view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_66 = x_399 * view_24 - view_24 = None - layer_norm_57 = torch.nn.functional.layer_norm( - x_399, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_399 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = (None) - linear_48 = torch._C._nn.linear( - layer_norm_57, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_57 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_25 = linear_48.reshape(1, pow_52, 3, 16, 32) - linear_48 = None - qkv_12 = reshape_25.permute(2, 0, 3, 1, 4) - reshape_25 = None - unbind_12 = qkv_12.unbind(0) - qkv_12 = None - q_12 = unbind_12[0] - k_12 = unbind_12[1] - v_12 = unbind_12[2] - unbind_12 = None - x_400 = torch._C._nn.scaled_dot_product_attention( - q_12, k_12, v_12, dropout_p=0.0 - ) - q_12 = k_12 = v_12 = None - transpose_13 = x_400.transpose(1, 2) - x_400 = None - x_401 = transpose_13.reshape(1, pow_52, 512) - transpose_13 = None - x_402 = torch._C._nn.linear( - x_401, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_401 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_403 = torch.nn.functional.dropout(x_402, 0.0, False, False) - x_402 = None - x_404 = mul_66 + x_403 - mul_66 = x_403 = None - view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_67 = x_404 * view_25 - view_25 = None - layer_norm_58 = torch.nn.functional.layer_norm( - x_404, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_404 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = (None) - x_405 = torch._C._nn.linear( - layer_norm_58, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_58 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_42 = torch.nn.functional.relu(x_405, inplace=False) - x_405 = None - pow_43 = relu_42**2 - relu_42 = None - mul_68 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ - * pow_43 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = ( - pow_43 - ) = None - x_406 = ( - mul_68 - + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ - ) - mul_68 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = (None) - x_407 = torch.nn.functional.dropout(x_406, 0.0, False, False) - x_406 = None - x_408 = torch._C._nn.linear( - x_407, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_407 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_409 = torch.nn.functional.dropout(x_408, 0.0, False, False) - x_408 = None - x_410 = mul_67 + x_409 - mul_67 = x_409 = None - view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_69 = x_410 * view_26 - view_26 = None - layer_norm_59 = torch.nn.functional.layer_norm( - x_410, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_410 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = (None) - linear_52 = torch._C._nn.linear( - layer_norm_59, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_59 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_27 = linear_52.reshape(1, pow_52, 3, 16, 32) - linear_52 = None - qkv_13 = reshape_27.permute(2, 0, 3, 1, 4) - reshape_27 = None - unbind_13 = qkv_13.unbind(0) - qkv_13 = None - q_13 = unbind_13[0] - k_13 = unbind_13[1] - v_13 = unbind_13[2] - unbind_13 = None - x_411 = torch._C._nn.scaled_dot_product_attention( - q_13, k_13, v_13, dropout_p=0.0 - ) - q_13 = k_13 = v_13 = None - transpose_14 = x_411.transpose(1, 2) - x_411 = None - x_412 = transpose_14.reshape(1, pow_52, 512) - transpose_14 = None - x_413 = torch._C._nn.linear( - x_412, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_412 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_414 = torch.nn.functional.dropout(x_413, 0.0, False, False) - x_413 = None - x_415 = mul_69 + x_414 - mul_69 = x_414 = None - view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_70 = x_415 * view_27 - view_27 = None - layer_norm_60 = torch.nn.functional.layer_norm( - x_415, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_415 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = (None) - x_416 = torch._C._nn.linear( - layer_norm_60, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_60 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_43 = torch.nn.functional.relu(x_416, inplace=False) - x_416 = None - pow_44 = relu_43**2 - relu_43 = None - mul_71 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ - * pow_44 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = ( - pow_44 - ) = None - x_417 = ( - mul_71 - + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ - ) - mul_71 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = (None) - x_418 = torch.nn.functional.dropout(x_417, 0.0, False, False) - x_417 = None - x_419 = torch._C._nn.linear( - x_418, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_418 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_420 = torch.nn.functional.dropout(x_419, 0.0, False, False) - x_419 = None - x_421 = mul_70 + x_420 - mul_70 = x_420 = None - view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_72 = x_421 * view_28 - view_28 = None - layer_norm_61 = torch.nn.functional.layer_norm( - x_421, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_421 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = (None) - linear_56 = torch._C._nn.linear( - layer_norm_61, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_61 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_29 = linear_56.reshape(1, pow_52, 3, 16, 32) - linear_56 = None - qkv_14 = reshape_29.permute(2, 0, 3, 1, 4) - reshape_29 = None - unbind_14 = qkv_14.unbind(0) - qkv_14 = None - q_14 = unbind_14[0] - k_14 = unbind_14[1] - v_14 = unbind_14[2] - unbind_14 = None - x_422 = torch._C._nn.scaled_dot_product_attention( - q_14, k_14, v_14, dropout_p=0.0 - ) - q_14 = k_14 = v_14 = None - transpose_15 = x_422.transpose(1, 2) - x_422 = None - x_423 = transpose_15.reshape(1, pow_52, 512) - transpose_15 = None - x_424 = torch._C._nn.linear( - x_423, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_423 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_425 = torch.nn.functional.dropout(x_424, 0.0, False, False) - x_424 = None - x_426 = mul_72 + x_425 - mul_72 = x_425 = None - view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_73 = x_426 * view_29 - view_29 = None - layer_norm_62 = torch.nn.functional.layer_norm( - x_426, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_426 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = (None) - x_427 = torch._C._nn.linear( - layer_norm_62, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_62 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_44 = torch.nn.functional.relu(x_427, inplace=False) - x_427 = None - pow_45 = relu_44**2 - relu_44 = None - mul_74 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ - * pow_45 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = ( - pow_45 - ) = None - x_428 = ( - mul_74 - + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ - ) - mul_74 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = (None) - x_429 = torch.nn.functional.dropout(x_428, 0.0, False, False) - x_428 = None - x_430 = torch._C._nn.linear( - x_429, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_429 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_431 = torch.nn.functional.dropout(x_430, 0.0, False, False) - x_430 = None - x_432 = mul_73 + x_431 - mul_73 = x_431 = None - view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_75 = x_432 * view_30 - view_30 = None - layer_norm_63 = torch.nn.functional.layer_norm( - x_432, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_432 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = (None) - linear_60 = torch._C._nn.linear( - layer_norm_63, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_63 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_31 = linear_60.reshape(1, pow_52, 3, 16, 32) - linear_60 = None - qkv_15 = reshape_31.permute(2, 0, 3, 1, 4) - reshape_31 = None - unbind_15 = qkv_15.unbind(0) - qkv_15 = None - q_15 = unbind_15[0] - k_15 = unbind_15[1] - v_15 = unbind_15[2] - unbind_15 = None - x_433 = torch._C._nn.scaled_dot_product_attention( - q_15, k_15, v_15, dropout_p=0.0 - ) - q_15 = k_15 = v_15 = None - transpose_16 = x_433.transpose(1, 2) - x_433 = None - x_434 = transpose_16.reshape(1, pow_52, 512) - transpose_16 = None - x_435 = torch._C._nn.linear( - x_434, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_434 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_436 = torch.nn.functional.dropout(x_435, 0.0, False, False) - x_435 = None - x_437 = mul_75 + x_436 - mul_75 = x_436 = None - view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_76 = x_437 * view_31 - view_31 = None - layer_norm_64 = torch.nn.functional.layer_norm( - x_437, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_437 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = (None) - x_438 = torch._C._nn.linear( - layer_norm_64, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_64 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_45 = torch.nn.functional.relu(x_438, inplace=False) - x_438 = None - pow_46 = relu_45**2 - relu_45 = None - mul_77 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ - * pow_46 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = ( - pow_46 - ) = None - x_439 = ( - mul_77 - + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ - ) - mul_77 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = (None) - x_440 = torch.nn.functional.dropout(x_439, 0.0, False, False) - x_439 = None - x_441 = torch._C._nn.linear( - x_440, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_440 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_442 = torch.nn.functional.dropout(x_441, 0.0, False, False) - x_441 = None - x_443 = mul_76 + x_442 - mul_76 = x_442 = None - view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_78 = x_443 * view_32 - view_32 = None - layer_norm_65 = torch.nn.functional.layer_norm( - x_443, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_443 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = (None) - linear_64 = torch._C._nn.linear( - layer_norm_65, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_65 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_33 = linear_64.reshape(1, pow_52, 3, 16, 32) - linear_64 = None - qkv_16 = reshape_33.permute(2, 0, 3, 1, 4) - reshape_33 = None - unbind_16 = qkv_16.unbind(0) - qkv_16 = None - q_16 = unbind_16[0] - k_16 = unbind_16[1] - v_16 = unbind_16[2] - unbind_16 = None - x_444 = torch._C._nn.scaled_dot_product_attention( - q_16, k_16, v_16, dropout_p=0.0 - ) - q_16 = k_16 = v_16 = None - transpose_17 = x_444.transpose(1, 2) - x_444 = None - x_445 = transpose_17.reshape(1, pow_52, 512) - transpose_17 = None - x_446 = torch._C._nn.linear( - x_445, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_445 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_447 = torch.nn.functional.dropout(x_446, 0.0, False, False) - x_446 = None - x_448 = mul_78 + x_447 - mul_78 = x_447 = None - view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_79 = x_448 * view_33 - view_33 = None - layer_norm_66 = torch.nn.functional.layer_norm( - x_448, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_448 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = (None) - x_449 = torch._C._nn.linear( - layer_norm_66, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_66 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_46 = torch.nn.functional.relu(x_449, inplace=False) - x_449 = None - pow_47 = relu_46**2 - relu_46 = None - mul_80 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ - * pow_47 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = ( - pow_47 - ) = None - x_450 = ( - mul_80 - + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ - ) - mul_80 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = (None) - x_451 = torch.nn.functional.dropout(x_450, 0.0, False, False) - x_450 = None - x_452 = torch._C._nn.linear( - x_451, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_451 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_453 = torch.nn.functional.dropout(x_452, 0.0, False, False) - x_452 = None - x_454 = mul_79 + x_453 - mul_79 = x_453 = None - view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_81 = x_454 * view_34 - view_34 = None - layer_norm_67 = torch.nn.functional.layer_norm( - x_454, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_454 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = (None) - linear_68 = torch._C._nn.linear( - layer_norm_67, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_67 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_35 = linear_68.reshape(1, pow_52, 3, 16, 32) - linear_68 = None - qkv_17 = reshape_35.permute(2, 0, 3, 1, 4) - reshape_35 = None - unbind_17 = qkv_17.unbind(0) - qkv_17 = None - q_17 = unbind_17[0] - k_17 = unbind_17[1] - v_17 = unbind_17[2] - unbind_17 = None - x_455 = torch._C._nn.scaled_dot_product_attention( - q_17, k_17, v_17, dropout_p=0.0 - ) - q_17 = k_17 = v_17 = None - transpose_18 = x_455.transpose(1, 2) - x_455 = None - x_456 = transpose_18.reshape(1, pow_52, 512) - transpose_18 = pow_52 = None - x_457 = torch._C._nn.linear( - x_456, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_456 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_458 = torch.nn.functional.dropout(x_457, 0.0, False, False) - x_457 = None - x_459 = mul_81 + x_458 - mul_81 = x_458 = None - view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_82 = x_459 * view_35 - view_35 = None - layer_norm_68 = torch.nn.functional.layer_norm( - x_459, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_459 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = (None) - x_460 = torch._C._nn.linear( - layer_norm_68, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_68 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_47 = torch.nn.functional.relu(x_460, inplace=False) - x_460 = None - pow_48 = relu_47**2 - relu_47 = None - mul_83 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ - * pow_48 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = ( - pow_48 - ) = None - x_461 = ( - mul_83 - + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ - ) - mul_83 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = (None) - x_462 = torch.nn.functional.dropout(x_461, 0.0, False, False) - x_461 = None - x_463 = torch._C._nn.linear( - x_462, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_462 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_464 = torch.nn.functional.dropout(x_463, 0.0, False, False) - x_463 = None - x_465 = mul_82 + x_464 - mul_82 = x_464 = None - transpose_19 = x_465.transpose(1, 2) - x_465 = None - x_466 = transpose_19.reshape(1, 512, sym_sum_3, sym_sum_3) - transpose_19 = sym_sum_3 = None - x_467 = x_466.permute(0, 2, 3, 1) - x_466 = None - x_468 = torch.nn.functional.layer_norm( - x_467, - (512,), - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_467 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) - x_469 = x_468.permute(0, 3, 1, 2) - x_468 = None - x_470 = torch.conv2d( - x_469, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_469 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_3 = sym_sum // 32 - sym_sum = None - sym_sum_4 = torch.sym_sum([1, floordiv_3]) - floordiv_3 = None - reshape_38 = x_470.reshape(1, 768, -1) - x_470 = None - x_471 = reshape_38.transpose(1, 2) - reshape_38 = None - view_36 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_84 = x_471 * view_36 - view_36 = None - layer_norm_70 = torch.nn.functional.layer_norm( - x_471, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_471 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_53 = sym_sum_4**2 - linear_72 = torch._C._nn.linear( - layer_norm_70, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_70 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_39 = linear_72.reshape(1, pow_53, 3, 24, 32) - linear_72 = None - qkv_18 = reshape_39.permute(2, 0, 3, 1, 4) - reshape_39 = None - unbind_18 = qkv_18.unbind(0) - qkv_18 = None - q_18 = unbind_18[0] - k_18 = unbind_18[1] - v_18 = unbind_18[2] - unbind_18 = None - x_472 = torch._C._nn.scaled_dot_product_attention( - q_18, k_18, v_18, dropout_p=0.0 - ) - q_18 = k_18 = v_18 = None - transpose_21 = x_472.transpose(1, 2) - x_472 = None - x_473 = transpose_21.reshape(1, pow_53, 768) - transpose_21 = None - x_474 = torch._C._nn.linear( - x_473, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_473 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_475 = torch.nn.functional.dropout(x_474, 0.0, False, False) - x_474 = None - x_476 = mul_84 + x_475 - mul_84 = x_475 = None - view_37 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_85 = x_476 * view_37 - view_37 = None - layer_norm_71 = torch.nn.functional.layer_norm( - x_476, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_476 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_477 = torch._C._nn.linear( - layer_norm_71, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_71 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_48 = torch.nn.functional.relu(x_477, inplace=False) - x_477 = None - pow_49 = relu_48**2 - relu_48 = None - mul_86 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_49 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_49 - ) = None - x_478 = ( - mul_86 - + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_86 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_479 = torch.nn.functional.dropout(x_478, 0.0, False, False) - x_478 = None - x_480 = torch._C._nn.linear( - x_479, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_479 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_481 = torch.nn.functional.dropout(x_480, 0.0, False, False) - x_480 = None - x_482 = mul_85 + x_481 - mul_85 = x_481 = None - view_38 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_87 = x_482 * view_38 - view_38 = None - layer_norm_72 = torch.nn.functional.layer_norm( - x_482, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_482 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_76 = torch._C._nn.linear( - layer_norm_72, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_72 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_41 = linear_76.reshape(1, pow_53, 3, 24, 32) - linear_76 = None - qkv_19 = reshape_41.permute(2, 0, 3, 1, 4) - reshape_41 = None - unbind_19 = qkv_19.unbind(0) - qkv_19 = None - q_19 = unbind_19[0] - k_19 = unbind_19[1] - v_19 = unbind_19[2] - unbind_19 = None - x_483 = torch._C._nn.scaled_dot_product_attention( - q_19, k_19, v_19, dropout_p=0.0 - ) - q_19 = k_19 = v_19 = None - transpose_22 = x_483.transpose(1, 2) - x_483 = None - x_484 = transpose_22.reshape(1, pow_53, 768) - transpose_22 = None - x_485 = torch._C._nn.linear( - x_484, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_484 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_486 = torch.nn.functional.dropout(x_485, 0.0, False, False) - x_485 = None - x_487 = mul_87 + x_486 - mul_87 = x_486 = None - view_39 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_88 = x_487 * view_39 - view_39 = None - layer_norm_73 = torch.nn.functional.layer_norm( - x_487, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_487 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_488 = torch._C._nn.linear( - layer_norm_73, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_73 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_49 = torch.nn.functional.relu(x_488, inplace=False) - x_488 = None - pow_50 = relu_49**2 - relu_49 = None - mul_89 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_50 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_50 - ) = None - x_489 = ( - mul_89 - + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_89 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_490 = torch.nn.functional.dropout(x_489, 0.0, False, False) - x_489 = None - x_491 = torch._C._nn.linear( - x_490, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_490 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_492 = torch.nn.functional.dropout(x_491, 0.0, False, False) - x_491 = None - x_493 = mul_88 + x_492 - mul_88 = x_492 = None - view_40 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_90 = x_493 * view_40 - view_40 = None - layer_norm_74 = torch.nn.functional.layer_norm( - x_493, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_493 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_80 = torch._C._nn.linear( - layer_norm_74, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_74 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_43 = linear_80.reshape(1, pow_53, 3, 24, 32) - linear_80 = None - qkv_20 = reshape_43.permute(2, 0, 3, 1, 4) - reshape_43 = None - unbind_20 = qkv_20.unbind(0) - qkv_20 = None - q_20 = unbind_20[0] - k_20 = unbind_20[1] - v_20 = unbind_20[2] - unbind_20 = None - x_494 = torch._C._nn.scaled_dot_product_attention( - q_20, k_20, v_20, dropout_p=0.0 - ) - q_20 = k_20 = v_20 = None - transpose_23 = x_494.transpose(1, 2) - x_494 = None - x_495 = transpose_23.reshape(1, pow_53, 768) - transpose_23 = pow_53 = None - x_496 = torch._C._nn.linear( - x_495, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_495 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_497 = torch.nn.functional.dropout(x_496, 0.0, False, False) - x_496 = None - x_498 = mul_90 + x_497 - mul_90 = x_497 = None - view_41 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_91 = x_498 * view_41 - view_41 = None - layer_norm_75 = torch.nn.functional.layer_norm( - x_498, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_498 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_499 = torch._C._nn.linear( - layer_norm_75, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_75 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_50 = torch.nn.functional.relu(x_499, inplace=False) - x_499 = None - pow_51 = relu_50**2 - relu_50 = None - mul_92 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_51 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_51 - ) = None - x_500 = ( - mul_92 - + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_92 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_501 = torch.nn.functional.dropout(x_500, 0.0, False, False) - x_500 = None - x_502 = torch._C._nn.linear( - x_501, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_501 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_503 = torch.nn.functional.dropout(x_502, 0.0, False, False) - x_502 = None - x_504 = mul_91 + x_503 - mul_91 = x_503 = None - transpose_24 = x_504.transpose(1, 2) - x_504 = None - x_505 = transpose_24.reshape(1, 768, sym_sum_4, sym_sum_4) - transpose_24 = sym_sum_4 = None - x_506 = torch.nn.functional.adaptive_avg_pool2d(x_505, 1) - x_505 = None - x_507 = x_506.permute(0, 2, 3, 1) - x_506 = None - x_508 = torch.nn.functional.layer_norm( - x_507, - (768,), - l_self_modules_head_modules_norm_parameters_weight_, - l_self_modules_head_modules_norm_parameters_bias_, - 1e-06, - ) - x_507 = ( - l_self_modules_head_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_norm_parameters_bias_ = None - x_509 = x_508.permute(0, 3, 1, 2) - x_508 = None - x_510 = x_509.flatten(1, -1) - x_509 = None - x_511 = torch.nn.functional.dropout(x_510, 0.0, False, False) - x_510 = None - x_512 = torch._C._nn.linear( - x_511, - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, - ) - x_511 = ( - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None - relu_51 = torch.nn.functional.relu(x_512, inplace=False) - x_512 = None - x_513 = torch.square(relu_51) - relu_51 = None - x_514 = torch.nn.functional.layer_norm( - x_513, - (3072,), - l_self_modules_head_modules_fc_modules_norm_parameters_weight_, - l_self_modules_head_modules_fc_modules_norm_parameters_bias_, - 1e-06, - ) - x_513 = ( - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None - x_515 = torch.nn.functional.dropout(x_514, 0.0, False, False) - x_514 = None - x_516 = torch._C._nn.linear( - x_515, - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, - ) - x_515 = ( - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None - return (x_516,) diff --git a/samples/timm/caformer_b36.sail_in22k/weight_meta.py b/samples/timm/caformer_b36.sail_in22k/weight_meta.py deleted file mode 100644 index d7c23666a9..0000000000 --- a/samples/timm/caformer_b36.sail_in22k/weight_meta.py +++ /dev/null @@ -1,3968 +0,0 @@ -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: - name = "L_self_modules_stem_modules_conv_parameters_weight_" - shape = [128, 3, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: - name = "L_self_modules_stem_modules_conv_parameters_bias_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_s1: - name = "s1" - shape = [] - dtype = "torch.int64" - device = "cpu" - mean = None - std = None - data = [4] - - -class Program_weight_tensor_meta_L_x_: - name = "L_x_" - shape = [1, 3, 224, 224] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.227 - std = 1.283 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: - name = "L_self_modules_stem_modules_norm_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [256, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [256, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [128, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [512, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [128, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [256, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [256, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [128, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [512, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [128, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [256, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [256, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [128, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [512, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [128, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" - shape = [256, 128, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" - shape = [512, 256, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" - shape = [768, 512, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [2304, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [768, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [3072, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [768, 3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [2304, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [768, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [3072, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [768, 3072] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [2304, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [768, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [3072, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [768, 3072] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_norm_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_norm_parameters_bias_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" - shape = [3072, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" - shape = [21841, 3072] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" - shape = [21841] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None diff --git a/samples/timm/caformer_b36.sail_in22k_ft_in1k/graph_net.json b/samples/timm/caformer_b36.sail_in22k_ft_in1k/graph_net.json deleted file mode 100644 index 1373fe3b58..0000000000 --- a/samples/timm/caformer_b36.sail_in22k_ft_in1k/graph_net.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "framework": "torch", - "num_devices_required": 1, - "num_nodes_required": 1 -} \ No newline at end of file diff --git a/samples/timm/caformer_b36.sail_in22k_ft_in1k/input_meta.py b/samples/timm/caformer_b36.sail_in22k_ft_in1k/input_meta.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_b36.sail_in22k_ft_in1k/input_tensor_constraints.py b/samples/timm/caformer_b36.sail_in22k_ft_in1k/input_tensor_constraints.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_b36.sail_in22k_ft_in1k/model.py b/samples/timm/caformer_b36.sail_in22k_ft_in1k/model.py deleted file mode 100644 index 08d3de35cc..0000000000 --- a/samples/timm/caformer_b36.sail_in22k_ft_in1k/model.py +++ /dev/null @@ -1,4758 +0,0 @@ -import torch - - -class GraphModule(torch.nn.Module): - def forward( - self, - L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - s1: torch.SymInt, - L_x_: torch.Tensor, - L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - ): - l_self_modules_stem_modules_conv_parameters_weight_ = ( - L_self_modules_stem_modules_conv_parameters_weight_ - ) - l_self_modules_stem_modules_conv_parameters_bias_ = ( - L_self_modules_stem_modules_conv_parameters_bias_ - ) - l_x_ = L_x_ - l_self_modules_stem_modules_norm_parameters_weight_ = ( - L_self_modules_stem_modules_norm_parameters_weight_ - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_head_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ - ) - x = torch.conv2d( - l_x_, - l_self_modules_stem_modules_conv_parameters_weight_, - l_self_modules_stem_modules_conv_parameters_bias_, - (4, 4), - (2, 2), - (1, 1), - 1, - ) - l_x_ = ( - l_self_modules_stem_modules_conv_parameters_weight_ - ) = l_self_modules_stem_modules_conv_parameters_bias_ = None - x_1 = x.permute(0, 2, 3, 1) - x = None - x_2 = torch.nn.functional.layer_norm( - x_1, - (128,), - l_self_modules_stem_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None - x_3 = x_2.permute(0, 3, 1, 2) - x_2 = None - sym_sum = torch.sym_sum([-3, s1]) - s1 = None - floordiv = sym_sum // 4 - sym_sum_1 = torch.sym_sum([1, floordiv]) - floordiv = sym_sum_1 = None - x_4 = x_3.permute(0, 2, 3, 1) - x_5 = torch.nn.functional.layer_norm( - x_4, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_6 = x_5.permute(0, 3, 1, 2) - x_5 = None - x_7 = torch.conv2d( - x_6, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu = torch.nn.functional.relu(x_7, inplace=False) - x_7 = None - pow_1 = relu**2 - relu = None - mul = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_1 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_1 - ) = None - x_8 = ( - mul - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_9 = torch.conv2d( - x_8, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 256, - ) - x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_10 = torch.conv2d( - x_9, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_11 = x_3 + x_10 - x_3 = x_10 = None - x_12 = x_11.permute(0, 2, 3, 1) - x_13 = torch.nn.functional.layer_norm( - x_12, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_14 = x_13.permute(0, 3, 1, 2) - x_13 = None - x_15 = torch.conv2d( - x_14, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_1 = torch.nn.functional.relu(x_15, inplace=False) - x_15 = None - pow_2 = relu_1**2 - relu_1 = None - mul_1 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_2 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_2 - ) = None - x_16 = ( - mul_1 - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) - x_16 = None - x_18 = torch.conv2d( - x_17, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) - x_18 = None - x_20 = x_11 + x_19 - x_11 = x_19 = None - x_21 = x_20.permute(0, 2, 3, 1) - x_22 = torch.nn.functional.layer_norm( - x_21, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_23 = x_22.permute(0, 3, 1, 2) - x_22 = None - x_24 = torch.conv2d( - x_23, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_2 = torch.nn.functional.relu(x_24, inplace=False) - x_24 = None - pow_3 = relu_2**2 - relu_2 = None - mul_2 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_3 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_3 - ) = None - x_25 = ( - mul_2 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_26 = torch.conv2d( - x_25, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 256, - ) - x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_27 = torch.conv2d( - x_26, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_28 = x_20 + x_27 - x_20 = x_27 = None - x_29 = x_28.permute(0, 2, 3, 1) - x_30 = torch.nn.functional.layer_norm( - x_29, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_31 = x_30.permute(0, 3, 1, 2) - x_30 = None - x_32 = torch.conv2d( - x_31, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_3 = torch.nn.functional.relu(x_32, inplace=False) - x_32 = None - pow_4 = relu_3**2 - relu_3 = None - mul_3 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_4 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_4 - ) = None - x_33 = ( - mul_3 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) - x_33 = None - x_35 = torch.conv2d( - x_34, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) - x_35 = None - x_37 = x_28 + x_36 - x_28 = x_36 = None - x_38 = x_37.permute(0, 2, 3, 1) - x_39 = torch.nn.functional.layer_norm( - x_38, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_40 = x_39.permute(0, 3, 1, 2) - x_39 = None - x_41 = torch.conv2d( - x_40, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_4 = torch.nn.functional.relu(x_41, inplace=False) - x_41 = None - pow_5 = relu_4**2 - relu_4 = None - mul_4 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_5 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_5 - ) = None - x_42 = ( - mul_4 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_43 = torch.conv2d( - x_42, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 256, - ) - x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_44 = torch.conv2d( - x_43, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_45 = x_37 + x_44 - x_37 = x_44 = None - x_46 = x_45.permute(0, 2, 3, 1) - x_47 = torch.nn.functional.layer_norm( - x_46, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_48 = x_47.permute(0, 3, 1, 2) - x_47 = None - x_49 = torch.conv2d( - x_48, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_5 = torch.nn.functional.relu(x_49, inplace=False) - x_49 = None - pow_6 = relu_5**2 - relu_5 = None - mul_5 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_6 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_6 - ) = None - x_50 = ( - mul_5 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) - x_50 = None - x_52 = torch.conv2d( - x_51, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) - x_52 = None - x_54 = x_45 + x_53 - x_45 = x_53 = None - x_55 = x_54.permute(0, 2, 3, 1) - x_54 = None - x_56 = torch.nn.functional.layer_norm( - x_55, - (128,), - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) - x_57 = x_56.permute(0, 3, 1, 2) - x_56 = None - x_58 = torch.conv2d( - x_57, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_1 = sym_sum // 8 - sym_sum_2 = torch.sym_sum([1, floordiv_1]) - floordiv_1 = sym_sum_2 = None - x_59 = x_58.permute(0, 2, 3, 1) - x_60 = torch.nn.functional.layer_norm( - x_59, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_61 = x_60.permute(0, 3, 1, 2) - x_60 = None - x_62 = torch.conv2d( - x_61, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_6 = torch.nn.functional.relu(x_62, inplace=False) - x_62 = None - pow_7 = relu_6**2 - relu_6 = None - mul_6 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_7 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_7 - ) = None - x_63 = ( - mul_6 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_64 = torch.conv2d( - x_63, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_65 = torch.conv2d( - x_64, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_66 = x_58 + x_65 - x_58 = x_65 = None - x_67 = x_66.permute(0, 2, 3, 1) - x_68 = torch.nn.functional.layer_norm( - x_67, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_69 = x_68.permute(0, 3, 1, 2) - x_68 = None - x_70 = torch.conv2d( - x_69, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_7 = torch.nn.functional.relu(x_70, inplace=False) - x_70 = None - pow_8 = relu_7**2 - relu_7 = None - mul_7 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_8 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_8 - ) = None - x_71 = ( - mul_7 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) - x_71 = None - x_73 = torch.conv2d( - x_72, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) - x_73 = None - x_75 = x_66 + x_74 - x_66 = x_74 = None - x_76 = x_75.permute(0, 2, 3, 1) - x_77 = torch.nn.functional.layer_norm( - x_76, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_78 = x_77.permute(0, 3, 1, 2) - x_77 = None - x_79 = torch.conv2d( - x_78, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_8 = torch.nn.functional.relu(x_79, inplace=False) - x_79 = None - pow_9 = relu_8**2 - relu_8 = None - mul_8 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_9 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_9 - ) = None - x_80 = ( - mul_8 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_81 = torch.conv2d( - x_80, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_82 = torch.conv2d( - x_81, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_83 = x_75 + x_82 - x_75 = x_82 = None - x_84 = x_83.permute(0, 2, 3, 1) - x_85 = torch.nn.functional.layer_norm( - x_84, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_86 = x_85.permute(0, 3, 1, 2) - x_85 = None - x_87 = torch.conv2d( - x_86, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_9 = torch.nn.functional.relu(x_87, inplace=False) - x_87 = None - pow_10 = relu_9**2 - relu_9 = None - mul_9 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_10 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_10 - ) = None - x_88 = ( - mul_9 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) - x_88 = None - x_90 = torch.conv2d( - x_89, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) - x_90 = None - x_92 = x_83 + x_91 - x_83 = x_91 = None - x_93 = x_92.permute(0, 2, 3, 1) - x_94 = torch.nn.functional.layer_norm( - x_93, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_95 = x_94.permute(0, 3, 1, 2) - x_94 = None - x_96 = torch.conv2d( - x_95, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_10 = torch.nn.functional.relu(x_96, inplace=False) - x_96 = None - pow_11 = relu_10**2 - relu_10 = None - mul_10 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_11 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_11 - ) = None - x_97 = ( - mul_10 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_98 = torch.conv2d( - x_97, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_99 = torch.conv2d( - x_98, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_100 = x_92 + x_99 - x_92 = x_99 = None - x_101 = x_100.permute(0, 2, 3, 1) - x_102 = torch.nn.functional.layer_norm( - x_101, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_103 = x_102.permute(0, 3, 1, 2) - x_102 = None - x_104 = torch.conv2d( - x_103, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_11 = torch.nn.functional.relu(x_104, inplace=False) - x_104 = None - pow_12 = relu_11**2 - relu_11 = None - mul_11 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_12 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_12 - ) = None - x_105 = ( - mul_11 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) - x_105 = None - x_107 = torch.conv2d( - x_106, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) - x_107 = None - x_109 = x_100 + x_108 - x_100 = x_108 = None - x_110 = x_109.permute(0, 2, 3, 1) - x_111 = torch.nn.functional.layer_norm( - x_110, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_110 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) - x_112 = x_111.permute(0, 3, 1, 2) - x_111 = None - x_113 = torch.conv2d( - x_112, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_112 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_12 = torch.nn.functional.relu(x_113, inplace=False) - x_113 = None - pow_13 = relu_12**2 - relu_12 = None - mul_12 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ - * pow_13 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_13 - ) = None - x_114 = ( - mul_12 - + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_12 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_115 = torch.conv2d( - x_114, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_114 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_116 = torch.conv2d( - x_115, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_115 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_117 = x_109 + x_116 - x_109 = x_116 = None - x_118 = x_117.permute(0, 2, 3, 1) - x_119 = torch.nn.functional.layer_norm( - x_118, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_118 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) - x_120 = x_119.permute(0, 3, 1, 2) - x_119 = None - x_121 = torch.conv2d( - x_120, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_120 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_13 = torch.nn.functional.relu(x_121, inplace=False) - x_121 = None - pow_14 = relu_13**2 - relu_13 = None - mul_13 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - * pow_14 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( - pow_14 - ) = None - x_122 = ( - mul_13 - + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - ) - mul_13 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) - x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) - x_122 = None - x_124 = torch.conv2d( - x_123, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_123 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) - x_124 = None - x_126 = x_117 + x_125 - x_117 = x_125 = None - x_127 = x_126.permute(0, 2, 3, 1) - x_128 = torch.nn.functional.layer_norm( - x_127, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_127 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) - x_129 = x_128.permute(0, 3, 1, 2) - x_128 = None - x_130 = torch.conv2d( - x_129, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_129 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_14 = torch.nn.functional.relu(x_130, inplace=False) - x_130 = None - pow_15 = relu_14**2 - relu_14 = None - mul_14 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ - * pow_15 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_15 - ) = None - x_131 = ( - mul_14 - + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_14 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_132 = torch.conv2d( - x_131, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_131 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_133 = torch.conv2d( - x_132, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_132 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_134 = x_126 + x_133 - x_126 = x_133 = None - x_135 = x_134.permute(0, 2, 3, 1) - x_136 = torch.nn.functional.layer_norm( - x_135, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_135 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) - x_137 = x_136.permute(0, 3, 1, 2) - x_136 = None - x_138 = torch.conv2d( - x_137, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_137 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_15 = torch.nn.functional.relu(x_138, inplace=False) - x_138 = None - pow_16 = relu_15**2 - relu_15 = None - mul_15 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - * pow_16 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( - pow_16 - ) = None - x_139 = ( - mul_15 - + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - ) - mul_15 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) - x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) - x_139 = None - x_141 = torch.conv2d( - x_140, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_140 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) - x_141 = None - x_143 = x_134 + x_142 - x_134 = x_142 = None - x_144 = x_143.permute(0, 2, 3, 1) - x_145 = torch.nn.functional.layer_norm( - x_144, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_144 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) - x_146 = x_145.permute(0, 3, 1, 2) - x_145 = None - x_147 = torch.conv2d( - x_146, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_146 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_16 = torch.nn.functional.relu(x_147, inplace=False) - x_147 = None - pow_17 = relu_16**2 - relu_16 = None - mul_16 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ - * pow_17 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_17 - ) = None - x_148 = ( - mul_16 - + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_16 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_149 = torch.conv2d( - x_148, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_148 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_150 = torch.conv2d( - x_149, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_149 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_151 = x_143 + x_150 - x_143 = x_150 = None - x_152 = x_151.permute(0, 2, 3, 1) - x_153 = torch.nn.functional.layer_norm( - x_152, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_152 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) - x_154 = x_153.permute(0, 3, 1, 2) - x_153 = None - x_155 = torch.conv2d( - x_154, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_154 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_17 = torch.nn.functional.relu(x_155, inplace=False) - x_155 = None - pow_18 = relu_17**2 - relu_17 = None - mul_17 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - * pow_18 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( - pow_18 - ) = None - x_156 = ( - mul_17 - + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - ) - mul_17 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) - x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) - x_156 = None - x_158 = torch.conv2d( - x_157, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_157 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) - x_158 = None - x_160 = x_151 + x_159 - x_151 = x_159 = None - x_161 = x_160.permute(0, 2, 3, 1) - x_162 = torch.nn.functional.layer_norm( - x_161, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_161 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) - x_163 = x_162.permute(0, 3, 1, 2) - x_162 = None - x_164 = torch.conv2d( - x_163, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_163 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_18 = torch.nn.functional.relu(x_164, inplace=False) - x_164 = None - pow_19 = relu_18**2 - relu_18 = None - mul_18 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ - * pow_19 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_19 - ) = None - x_165 = ( - mul_18 - + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_18 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_166 = torch.conv2d( - x_165, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_165 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_167 = torch.conv2d( - x_166, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_166 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_168 = x_160 + x_167 - x_160 = x_167 = None - x_169 = x_168.permute(0, 2, 3, 1) - x_170 = torch.nn.functional.layer_norm( - x_169, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_169 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) - x_171 = x_170.permute(0, 3, 1, 2) - x_170 = None - x_172 = torch.conv2d( - x_171, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_171 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_19 = torch.nn.functional.relu(x_172, inplace=False) - x_172 = None - pow_20 = relu_19**2 - relu_19 = None - mul_19 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - * pow_20 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( - pow_20 - ) = None - x_173 = ( - mul_19 - + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - ) - mul_19 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) - x_174 = torch.nn.functional.dropout(x_173, 0.0, False, False) - x_173 = None - x_175 = torch.conv2d( - x_174, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_174 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) - x_175 = None - x_177 = x_168 + x_176 - x_168 = x_176 = None - x_178 = x_177.permute(0, 2, 3, 1) - x_179 = torch.nn.functional.layer_norm( - x_178, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_178 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) - x_180 = x_179.permute(0, 3, 1, 2) - x_179 = None - x_181 = torch.conv2d( - x_180, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_180 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_20 = torch.nn.functional.relu(x_181, inplace=False) - x_181 = None - pow_21 = relu_20**2 - relu_20 = None - mul_20 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ - * pow_21 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_21 - ) = None - x_182 = ( - mul_20 - + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_20 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_183 = torch.conv2d( - x_182, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_182 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_184 = torch.conv2d( - x_183, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_183 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_185 = x_177 + x_184 - x_177 = x_184 = None - x_186 = x_185.permute(0, 2, 3, 1) - x_187 = torch.nn.functional.layer_norm( - x_186, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_186 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) - x_188 = x_187.permute(0, 3, 1, 2) - x_187 = None - x_189 = torch.conv2d( - x_188, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_188 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_21 = torch.nn.functional.relu(x_189, inplace=False) - x_189 = None - pow_22 = relu_21**2 - relu_21 = None - mul_21 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - * pow_22 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( - pow_22 - ) = None - x_190 = ( - mul_21 - + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - ) - mul_21 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) - x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) - x_190 = None - x_192 = torch.conv2d( - x_191, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_191 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) - x_192 = None - x_194 = x_185 + x_193 - x_185 = x_193 = None - x_195 = x_194.permute(0, 2, 3, 1) - x_196 = torch.nn.functional.layer_norm( - x_195, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_195 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) - x_197 = x_196.permute(0, 3, 1, 2) - x_196 = None - x_198 = torch.conv2d( - x_197, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_197 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_22 = torch.nn.functional.relu(x_198, inplace=False) - x_198 = None - pow_23 = relu_22**2 - relu_22 = None - mul_22 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ - * pow_23 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_23 - ) = None - x_199 = ( - mul_22 - + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_22 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_200 = torch.conv2d( - x_199, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_199 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_201 = torch.conv2d( - x_200, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_200 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_202 = x_194 + x_201 - x_194 = x_201 = None - x_203 = x_202.permute(0, 2, 3, 1) - x_204 = torch.nn.functional.layer_norm( - x_203, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_203 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) - x_205 = x_204.permute(0, 3, 1, 2) - x_204 = None - x_206 = torch.conv2d( - x_205, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_205 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_23 = torch.nn.functional.relu(x_206, inplace=False) - x_206 = None - pow_24 = relu_23**2 - relu_23 = None - mul_23 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - * pow_24 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( - pow_24 - ) = None - x_207 = ( - mul_23 - + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - ) - mul_23 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) - x_208 = torch.nn.functional.dropout(x_207, 0.0, False, False) - x_207 = None - x_209 = torch.conv2d( - x_208, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_208 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) - x_209 = None - x_211 = x_202 + x_210 - x_202 = x_210 = None - x_212 = x_211.permute(0, 2, 3, 1) - x_213 = torch.nn.functional.layer_norm( - x_212, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_212 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) - x_214 = x_213.permute(0, 3, 1, 2) - x_213 = None - x_215 = torch.conv2d( - x_214, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_214 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_24 = torch.nn.functional.relu(x_215, inplace=False) - x_215 = None - pow_25 = relu_24**2 - relu_24 = None - mul_24 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ - * pow_25 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_25 - ) = None - x_216 = ( - mul_24 - + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_24 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_217 = torch.conv2d( - x_216, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_216 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_218 = torch.conv2d( - x_217, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_217 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_219 = x_211 + x_218 - x_211 = x_218 = None - x_220 = x_219.permute(0, 2, 3, 1) - x_221 = torch.nn.functional.layer_norm( - x_220, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_220 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) - x_222 = x_221.permute(0, 3, 1, 2) - x_221 = None - x_223 = torch.conv2d( - x_222, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_222 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_25 = torch.nn.functional.relu(x_223, inplace=False) - x_223 = None - pow_26 = relu_25**2 - relu_25 = None - mul_25 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - * pow_26 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( - pow_26 - ) = None - x_224 = ( - mul_25 - + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - ) - mul_25 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) - x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) - x_224 = None - x_226 = torch.conv2d( - x_225, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_225 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) - x_226 = None - x_228 = x_219 + x_227 - x_219 = x_227 = None - x_229 = x_228.permute(0, 2, 3, 1) - x_230 = torch.nn.functional.layer_norm( - x_229, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_229 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) - x_231 = x_230.permute(0, 3, 1, 2) - x_230 = None - x_232 = torch.conv2d( - x_231, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_231 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_26 = torch.nn.functional.relu(x_232, inplace=False) - x_232 = None - pow_27 = relu_26**2 - relu_26 = None - mul_26 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ - * pow_27 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_27 - ) = None - x_233 = ( - mul_26 - + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_26 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_234 = torch.conv2d( - x_233, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_233 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_235 = torch.conv2d( - x_234, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_234 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_236 = x_228 + x_235 - x_228 = x_235 = None - x_237 = x_236.permute(0, 2, 3, 1) - x_238 = torch.nn.functional.layer_norm( - x_237, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_237 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) - x_239 = x_238.permute(0, 3, 1, 2) - x_238 = None - x_240 = torch.conv2d( - x_239, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_239 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_27 = torch.nn.functional.relu(x_240, inplace=False) - x_240 = None - pow_28 = relu_27**2 - relu_27 = None - mul_27 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - * pow_28 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( - pow_28 - ) = None - x_241 = ( - mul_27 - + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - ) - mul_27 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) - x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) - x_241 = None - x_243 = torch.conv2d( - x_242, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_242 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_244 = torch.nn.functional.dropout(x_243, 0.0, False, False) - x_243 = None - x_245 = x_236 + x_244 - x_236 = x_244 = None - x_246 = x_245.permute(0, 2, 3, 1) - x_247 = torch.nn.functional.layer_norm( - x_246, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_246 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) - x_248 = x_247.permute(0, 3, 1, 2) - x_247 = None - x_249 = torch.conv2d( - x_248, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_248 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_28 = torch.nn.functional.relu(x_249, inplace=False) - x_249 = None - pow_29 = relu_28**2 - relu_28 = None - mul_28 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ - * pow_29 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_29 - ) = None - x_250 = ( - mul_28 - + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_28 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_251 = torch.conv2d( - x_250, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_250 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_252 = torch.conv2d( - x_251, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_251 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_253 = x_245 + x_252 - x_245 = x_252 = None - x_254 = x_253.permute(0, 2, 3, 1) - x_255 = torch.nn.functional.layer_norm( - x_254, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_254 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) - x_256 = x_255.permute(0, 3, 1, 2) - x_255 = None - x_257 = torch.conv2d( - x_256, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_256 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_29 = torch.nn.functional.relu(x_257, inplace=False) - x_257 = None - pow_30 = relu_29**2 - relu_29 = None - mul_29 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - * pow_30 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( - pow_30 - ) = None - x_258 = ( - mul_29 - + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - ) - mul_29 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) - x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) - x_258 = None - x_260 = torch.conv2d( - x_259, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_259 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_261 = torch.nn.functional.dropout(x_260, 0.0, False, False) - x_260 = None - x_262 = x_253 + x_261 - x_253 = x_261 = None - x_263 = x_262.permute(0, 2, 3, 1) - x_262 = None - x_264 = torch.nn.functional.layer_norm( - x_263, - (256,), - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_263 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) - x_265 = x_264.permute(0, 3, 1, 2) - x_264 = None - x_266 = torch.conv2d( - x_265, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_265 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_2 = sym_sum // 16 - sym_sum_3 = torch.sym_sum([1, floordiv_2]) - floordiv_2 = None - reshape = x_266.reshape(1, 512, -1) - x_266 = None - x_267 = reshape.transpose(1, 2) - reshape = None - view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_30 = x_267 * view - view = None - layer_norm_33 = torch.nn.functional.layer_norm( - x_267, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_52 = sym_sum_3**2 - linear = torch._C._nn.linear( - layer_norm_33, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_33 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_1 = linear.reshape(1, pow_52, 3, 16, 32) - linear = None - qkv = reshape_1.permute(2, 0, 3, 1, 4) - reshape_1 = None - unbind = qkv.unbind(0) - qkv = None - q = unbind[0] - k = unbind[1] - v = unbind[2] - unbind = None - x_268 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) - q = k = v = None - transpose_1 = x_268.transpose(1, 2) - x_268 = None - x_269 = transpose_1.reshape(1, pow_52, 512) - transpose_1 = None - x_270 = torch._C._nn.linear( - x_269, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_269 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_271 = torch.nn.functional.dropout(x_270, 0.0, False, False) - x_270 = None - x_272 = mul_30 + x_271 - mul_30 = x_271 = None - view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_31 = x_272 * view_1 - view_1 = None - layer_norm_34 = torch.nn.functional.layer_norm( - x_272, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_272 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_273 = torch._C._nn.linear( - layer_norm_34, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_34 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_30 = torch.nn.functional.relu(x_273, inplace=False) - x_273 = None - pow_31 = relu_30**2 - relu_30 = None - mul_32 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_31 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_31 - ) = None - x_274 = ( - mul_32 - + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_275 = torch.nn.functional.dropout(x_274, 0.0, False, False) - x_274 = None - x_276 = torch._C._nn.linear( - x_275, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_277 = torch.nn.functional.dropout(x_276, 0.0, False, False) - x_276 = None - x_278 = mul_31 + x_277 - mul_31 = x_277 = None - view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_33 = x_278 * view_2 - view_2 = None - layer_norm_35 = torch.nn.functional.layer_norm( - x_278, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_4 = torch._C._nn.linear( - layer_norm_35, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_3 = linear_4.reshape(1, pow_52, 3, 16, 32) - linear_4 = None - qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) - reshape_3 = None - unbind_1 = qkv_1.unbind(0) - qkv_1 = None - q_1 = unbind_1[0] - k_1 = unbind_1[1] - v_1 = unbind_1[2] - unbind_1 = None - x_279 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) - q_1 = k_1 = v_1 = None - transpose_2 = x_279.transpose(1, 2) - x_279 = None - x_280 = transpose_2.reshape(1, pow_52, 512) - transpose_2 = None - x_281 = torch._C._nn.linear( - x_280, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_280 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) - x_281 = None - x_283 = mul_33 + x_282 - mul_33 = x_282 = None - view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_34 = x_283 * view_3 - view_3 = None - layer_norm_36 = torch.nn.functional.layer_norm( - x_283, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_283 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_284 = torch._C._nn.linear( - layer_norm_36, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_36 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_31 = torch.nn.functional.relu(x_284, inplace=False) - x_284 = None - pow_32 = relu_31**2 - relu_31 = None - mul_35 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_32 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_32 - ) = None - x_285 = ( - mul_35 - + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_286 = torch.nn.functional.dropout(x_285, 0.0, False, False) - x_285 = None - x_287 = torch._C._nn.linear( - x_286, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) - x_287 = None - x_289 = mul_34 + x_288 - mul_34 = x_288 = None - view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_36 = x_289 * view_4 - view_4 = None - layer_norm_37 = torch.nn.functional.layer_norm( - x_289, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_8 = torch._C._nn.linear( - layer_norm_37, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_37 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_5 = linear_8.reshape(1, pow_52, 3, 16, 32) - linear_8 = None - qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) - reshape_5 = None - unbind_2 = qkv_2.unbind(0) - qkv_2 = None - q_2 = unbind_2[0] - k_2 = unbind_2[1] - v_2 = unbind_2[2] - unbind_2 = None - x_290 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) - q_2 = k_2 = v_2 = None - transpose_3 = x_290.transpose(1, 2) - x_290 = None - x_291 = transpose_3.reshape(1, pow_52, 512) - transpose_3 = None - x_292 = torch._C._nn.linear( - x_291, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_291 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_293 = torch.nn.functional.dropout(x_292, 0.0, False, False) - x_292 = None - x_294 = mul_36 + x_293 - mul_36 = x_293 = None - view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_37 = x_294 * view_5 - view_5 = None - layer_norm_38 = torch.nn.functional.layer_norm( - x_294, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_294 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_295 = torch._C._nn.linear( - layer_norm_38, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_32 = torch.nn.functional.relu(x_295, inplace=False) - x_295 = None - pow_33 = relu_32**2 - relu_32 = None - mul_38 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_33 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_33 - ) = None - x_296 = ( - mul_38 - + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_297 = torch.nn.functional.dropout(x_296, 0.0, False, False) - x_296 = None - x_298 = torch._C._nn.linear( - x_297, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_299 = torch.nn.functional.dropout(x_298, 0.0, False, False) - x_298 = None - x_300 = mul_37 + x_299 - mul_37 = x_299 = None - view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_39 = x_300 * view_6 - view_6 = None - layer_norm_39 = torch.nn.functional.layer_norm( - x_300, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) - linear_12 = torch._C._nn.linear( - layer_norm_39, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_39 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_7 = linear_12.reshape(1, pow_52, 3, 16, 32) - linear_12 = None - qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) - reshape_7 = None - unbind_3 = qkv_3.unbind(0) - qkv_3 = None - q_3 = unbind_3[0] - k_3 = unbind_3[1] - v_3 = unbind_3[2] - unbind_3 = None - x_301 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) - q_3 = k_3 = v_3 = None - transpose_4 = x_301.transpose(1, 2) - x_301 = None - x_302 = transpose_4.reshape(1, pow_52, 512) - transpose_4 = None - x_303 = torch._C._nn.linear( - x_302, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_302 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_304 = torch.nn.functional.dropout(x_303, 0.0, False, False) - x_303 = None - x_305 = mul_39 + x_304 - mul_39 = x_304 = None - view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_40 = x_305 * view_7 - view_7 = None - layer_norm_40 = torch.nn.functional.layer_norm( - x_305, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_305 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) - x_306 = torch._C._nn.linear( - layer_norm_40, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_40 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_33 = torch.nn.functional.relu(x_306, inplace=False) - x_306 = None - pow_34 = relu_33**2 - relu_33 = None - mul_41 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - * pow_34 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( - pow_34 - ) = None - x_307 = ( - mul_41 - + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - ) - mul_41 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) - x_308 = torch.nn.functional.dropout(x_307, 0.0, False, False) - x_307 = None - x_309 = torch._C._nn.linear( - x_308, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) - x_309 = None - x_311 = mul_40 + x_310 - mul_40 = x_310 = None - view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_42 = x_311 * view_8 - view_8 = None - layer_norm_41 = torch.nn.functional.layer_norm( - x_311, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) - linear_16 = torch._C._nn.linear( - layer_norm_41, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_41 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_9 = linear_16.reshape(1, pow_52, 3, 16, 32) - linear_16 = None - qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) - reshape_9 = None - unbind_4 = qkv_4.unbind(0) - qkv_4 = None - q_4 = unbind_4[0] - k_4 = unbind_4[1] - v_4 = unbind_4[2] - unbind_4 = None - x_312 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) - q_4 = k_4 = v_4 = None - transpose_5 = x_312.transpose(1, 2) - x_312 = None - x_313 = transpose_5.reshape(1, pow_52, 512) - transpose_5 = None - x_314 = torch._C._nn.linear( - x_313, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_313 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) - x_314 = None - x_316 = mul_42 + x_315 - mul_42 = x_315 = None - view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_43 = x_316 * view_9 - view_9 = None - layer_norm_42 = torch.nn.functional.layer_norm( - x_316, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_316 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) - x_317 = torch._C._nn.linear( - layer_norm_42, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_42 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_34 = torch.nn.functional.relu(x_317, inplace=False) - x_317 = None - pow_35 = relu_34**2 - relu_34 = None - mul_44 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - * pow_35 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( - pow_35 - ) = None - x_318 = ( - mul_44 - + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - ) - mul_44 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) - x_319 = torch.nn.functional.dropout(x_318, 0.0, False, False) - x_318 = None - x_320 = torch._C._nn.linear( - x_319, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_321 = torch.nn.functional.dropout(x_320, 0.0, False, False) - x_320 = None - x_322 = mul_43 + x_321 - mul_43 = x_321 = None - view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_45 = x_322 * view_10 - view_10 = None - layer_norm_43 = torch.nn.functional.layer_norm( - x_322, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) - linear_20 = torch._C._nn.linear( - layer_norm_43, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_43 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_11 = linear_20.reshape(1, pow_52, 3, 16, 32) - linear_20 = None - qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) - reshape_11 = None - unbind_5 = qkv_5.unbind(0) - qkv_5 = None - q_5 = unbind_5[0] - k_5 = unbind_5[1] - v_5 = unbind_5[2] - unbind_5 = None - x_323 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) - q_5 = k_5 = v_5 = None - transpose_6 = x_323.transpose(1, 2) - x_323 = None - x_324 = transpose_6.reshape(1, pow_52, 512) - transpose_6 = None - x_325 = torch._C._nn.linear( - x_324, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_324 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_326 = torch.nn.functional.dropout(x_325, 0.0, False, False) - x_325 = None - x_327 = mul_45 + x_326 - mul_45 = x_326 = None - view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_46 = x_327 * view_11 - view_11 = None - layer_norm_44 = torch.nn.functional.layer_norm( - x_327, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_327 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) - x_328 = torch._C._nn.linear( - layer_norm_44, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_44 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_35 = torch.nn.functional.relu(x_328, inplace=False) - x_328 = None - pow_36 = relu_35**2 - relu_35 = None - mul_47 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - * pow_36 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( - pow_36 - ) = None - x_329 = ( - mul_47 - + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - ) - mul_47 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) - x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) - x_329 = None - x_331 = torch._C._nn.linear( - x_330, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_332 = torch.nn.functional.dropout(x_331, 0.0, False, False) - x_331 = None - x_333 = mul_46 + x_332 - mul_46 = x_332 = None - view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_48 = x_333 * view_12 - view_12 = None - layer_norm_45 = torch.nn.functional.layer_norm( - x_333, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) - linear_24 = torch._C._nn.linear( - layer_norm_45, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_45 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_13 = linear_24.reshape(1, pow_52, 3, 16, 32) - linear_24 = None - qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) - reshape_13 = None - unbind_6 = qkv_6.unbind(0) - qkv_6 = None - q_6 = unbind_6[0] - k_6 = unbind_6[1] - v_6 = unbind_6[2] - unbind_6 = None - x_334 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) - q_6 = k_6 = v_6 = None - transpose_7 = x_334.transpose(1, 2) - x_334 = None - x_335 = transpose_7.reshape(1, pow_52, 512) - transpose_7 = None - x_336 = torch._C._nn.linear( - x_335, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_335 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_337 = torch.nn.functional.dropout(x_336, 0.0, False, False) - x_336 = None - x_338 = mul_48 + x_337 - mul_48 = x_337 = None - view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_49 = x_338 * view_13 - view_13 = None - layer_norm_46 = torch.nn.functional.layer_norm( - x_338, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_338 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) - x_339 = torch._C._nn.linear( - layer_norm_46, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_46 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_36 = torch.nn.functional.relu(x_339, inplace=False) - x_339 = None - pow_37 = relu_36**2 - relu_36 = None - mul_50 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - * pow_37 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( - pow_37 - ) = None - x_340 = ( - mul_50 - + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - ) - mul_50 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) - x_341 = torch.nn.functional.dropout(x_340, 0.0, False, False) - x_340 = None - x_342 = torch._C._nn.linear( - x_341, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_343 = torch.nn.functional.dropout(x_342, 0.0, False, False) - x_342 = None - x_344 = mul_49 + x_343 - mul_49 = x_343 = None - view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_51 = x_344 * view_14 - view_14 = None - layer_norm_47 = torch.nn.functional.layer_norm( - x_344, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) - linear_28 = torch._C._nn.linear( - layer_norm_47, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_47 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_15 = linear_28.reshape(1, pow_52, 3, 16, 32) - linear_28 = None - qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) - reshape_15 = None - unbind_7 = qkv_7.unbind(0) - qkv_7 = None - q_7 = unbind_7[0] - k_7 = unbind_7[1] - v_7 = unbind_7[2] - unbind_7 = None - x_345 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) - q_7 = k_7 = v_7 = None - transpose_8 = x_345.transpose(1, 2) - x_345 = None - x_346 = transpose_8.reshape(1, pow_52, 512) - transpose_8 = None - x_347 = torch._C._nn.linear( - x_346, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_346 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) - x_347 = None - x_349 = mul_51 + x_348 - mul_51 = x_348 = None - view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_52 = x_349 * view_15 - view_15 = None - layer_norm_48 = torch.nn.functional.layer_norm( - x_349, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_349 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) - x_350 = torch._C._nn.linear( - layer_norm_48, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_48 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_37 = torch.nn.functional.relu(x_350, inplace=False) - x_350 = None - pow_38 = relu_37**2 - relu_37 = None - mul_53 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - * pow_38 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( - pow_38 - ) = None - x_351 = ( - mul_53 - + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - ) - mul_53 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) - x_352 = torch.nn.functional.dropout(x_351, 0.0, False, False) - x_351 = None - x_353 = torch._C._nn.linear( - x_352, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_354 = torch.nn.functional.dropout(x_353, 0.0, False, False) - x_353 = None - x_355 = mul_52 + x_354 - mul_52 = x_354 = None - view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_54 = x_355 * view_16 - view_16 = None - layer_norm_49 = torch.nn.functional.layer_norm( - x_355, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) - linear_32 = torch._C._nn.linear( - layer_norm_49, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_49 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_17 = linear_32.reshape(1, pow_52, 3, 16, 32) - linear_32 = None - qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) - reshape_17 = None - unbind_8 = qkv_8.unbind(0) - qkv_8 = None - q_8 = unbind_8[0] - k_8 = unbind_8[1] - v_8 = unbind_8[2] - unbind_8 = None - x_356 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) - q_8 = k_8 = v_8 = None - transpose_9 = x_356.transpose(1, 2) - x_356 = None - x_357 = transpose_9.reshape(1, pow_52, 512) - transpose_9 = None - x_358 = torch._C._nn.linear( - x_357, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_357 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_359 = torch.nn.functional.dropout(x_358, 0.0, False, False) - x_358 = None - x_360 = mul_54 + x_359 - mul_54 = x_359 = None - view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_55 = x_360 * view_17 - view_17 = None - layer_norm_50 = torch.nn.functional.layer_norm( - x_360, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_360 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) - x_361 = torch._C._nn.linear( - layer_norm_50, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_50 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_38 = torch.nn.functional.relu(x_361, inplace=False) - x_361 = None - pow_39 = relu_38**2 - relu_38 = None - mul_56 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - * pow_39 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( - pow_39 - ) = None - x_362 = ( - mul_56 - + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - ) - mul_56 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) - x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) - x_362 = None - x_364 = torch._C._nn.linear( - x_363, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_365 = torch.nn.functional.dropout(x_364, 0.0, False, False) - x_364 = None - x_366 = mul_55 + x_365 - mul_55 = x_365 = None - view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_57 = x_366 * view_18 - view_18 = None - layer_norm_51 = torch.nn.functional.layer_norm( - x_366, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) - linear_36 = torch._C._nn.linear( - layer_norm_51, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_51 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_19 = linear_36.reshape(1, pow_52, 3, 16, 32) - linear_36 = None - qkv_9 = reshape_19.permute(2, 0, 3, 1, 4) - reshape_19 = None - unbind_9 = qkv_9.unbind(0) - qkv_9 = None - q_9 = unbind_9[0] - k_9 = unbind_9[1] - v_9 = unbind_9[2] - unbind_9 = None - x_367 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) - q_9 = k_9 = v_9 = None - transpose_10 = x_367.transpose(1, 2) - x_367 = None - x_368 = transpose_10.reshape(1, pow_52, 512) - transpose_10 = None - x_369 = torch._C._nn.linear( - x_368, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_368 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_370 = torch.nn.functional.dropout(x_369, 0.0, False, False) - x_369 = None - x_371 = mul_57 + x_370 - mul_57 = x_370 = None - view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_58 = x_371 * view_19 - view_19 = None - layer_norm_52 = torch.nn.functional.layer_norm( - x_371, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_371 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) - x_372 = torch._C._nn.linear( - layer_norm_52, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_52 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_39 = torch.nn.functional.relu(x_372, inplace=False) - x_372 = None - pow_40 = relu_39**2 - relu_39 = None - mul_59 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - * pow_40 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( - pow_40 - ) = None - x_373 = ( - mul_59 - + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - ) - mul_59 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) - x_374 = torch.nn.functional.dropout(x_373, 0.0, False, False) - x_373 = None - x_375 = torch._C._nn.linear( - x_374, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_374 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_376 = torch.nn.functional.dropout(x_375, 0.0, False, False) - x_375 = None - x_377 = mul_58 + x_376 - mul_58 = x_376 = None - view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_60 = x_377 * view_20 - view_20 = None - layer_norm_53 = torch.nn.functional.layer_norm( - x_377, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_377 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) - linear_40 = torch._C._nn.linear( - layer_norm_53, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_53 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_21 = linear_40.reshape(1, pow_52, 3, 16, 32) - linear_40 = None - qkv_10 = reshape_21.permute(2, 0, 3, 1, 4) - reshape_21 = None - unbind_10 = qkv_10.unbind(0) - qkv_10 = None - q_10 = unbind_10[0] - k_10 = unbind_10[1] - v_10 = unbind_10[2] - unbind_10 = None - x_378 = torch._C._nn.scaled_dot_product_attention( - q_10, k_10, v_10, dropout_p=0.0 - ) - q_10 = k_10 = v_10 = None - transpose_11 = x_378.transpose(1, 2) - x_378 = None - x_379 = transpose_11.reshape(1, pow_52, 512) - transpose_11 = None - x_380 = torch._C._nn.linear( - x_379, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_379 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_381 = torch.nn.functional.dropout(x_380, 0.0, False, False) - x_380 = None - x_382 = mul_60 + x_381 - mul_60 = x_381 = None - view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_61 = x_382 * view_21 - view_21 = None - layer_norm_54 = torch.nn.functional.layer_norm( - x_382, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_382 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) - x_383 = torch._C._nn.linear( - layer_norm_54, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_54 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_40 = torch.nn.functional.relu(x_383, inplace=False) - x_383 = None - pow_41 = relu_40**2 - relu_40 = None - mul_62 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - * pow_41 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( - pow_41 - ) = None - x_384 = ( - mul_62 - + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - ) - mul_62 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) - x_385 = torch.nn.functional.dropout(x_384, 0.0, False, False) - x_384 = None - x_386 = torch._C._nn.linear( - x_385, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_385 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_387 = torch.nn.functional.dropout(x_386, 0.0, False, False) - x_386 = None - x_388 = mul_61 + x_387 - mul_61 = x_387 = None - view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_63 = x_388 * view_22 - view_22 = None - layer_norm_55 = torch.nn.functional.layer_norm( - x_388, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_388 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) - linear_44 = torch._C._nn.linear( - layer_norm_55, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_55 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_23 = linear_44.reshape(1, pow_52, 3, 16, 32) - linear_44 = None - qkv_11 = reshape_23.permute(2, 0, 3, 1, 4) - reshape_23 = None - unbind_11 = qkv_11.unbind(0) - qkv_11 = None - q_11 = unbind_11[0] - k_11 = unbind_11[1] - v_11 = unbind_11[2] - unbind_11 = None - x_389 = torch._C._nn.scaled_dot_product_attention( - q_11, k_11, v_11, dropout_p=0.0 - ) - q_11 = k_11 = v_11 = None - transpose_12 = x_389.transpose(1, 2) - x_389 = None - x_390 = transpose_12.reshape(1, pow_52, 512) - transpose_12 = None - x_391 = torch._C._nn.linear( - x_390, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_390 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_392 = torch.nn.functional.dropout(x_391, 0.0, False, False) - x_391 = None - x_393 = mul_63 + x_392 - mul_63 = x_392 = None - view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_64 = x_393 * view_23 - view_23 = None - layer_norm_56 = torch.nn.functional.layer_norm( - x_393, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_393 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) - x_394 = torch._C._nn.linear( - layer_norm_56, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_56 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_41 = torch.nn.functional.relu(x_394, inplace=False) - x_394 = None - pow_42 = relu_41**2 - relu_41 = None - mul_65 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - * pow_42 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( - pow_42 - ) = None - x_395 = ( - mul_65 - + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - ) - mul_65 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) - x_396 = torch.nn.functional.dropout(x_395, 0.0, False, False) - x_395 = None - x_397 = torch._C._nn.linear( - x_396, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_396 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_398 = torch.nn.functional.dropout(x_397, 0.0, False, False) - x_397 = None - x_399 = mul_64 + x_398 - mul_64 = x_398 = None - view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_66 = x_399 * view_24 - view_24 = None - layer_norm_57 = torch.nn.functional.layer_norm( - x_399, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_399 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = (None) - linear_48 = torch._C._nn.linear( - layer_norm_57, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_57 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_25 = linear_48.reshape(1, pow_52, 3, 16, 32) - linear_48 = None - qkv_12 = reshape_25.permute(2, 0, 3, 1, 4) - reshape_25 = None - unbind_12 = qkv_12.unbind(0) - qkv_12 = None - q_12 = unbind_12[0] - k_12 = unbind_12[1] - v_12 = unbind_12[2] - unbind_12 = None - x_400 = torch._C._nn.scaled_dot_product_attention( - q_12, k_12, v_12, dropout_p=0.0 - ) - q_12 = k_12 = v_12 = None - transpose_13 = x_400.transpose(1, 2) - x_400 = None - x_401 = transpose_13.reshape(1, pow_52, 512) - transpose_13 = None - x_402 = torch._C._nn.linear( - x_401, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_401 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_403 = torch.nn.functional.dropout(x_402, 0.0, False, False) - x_402 = None - x_404 = mul_66 + x_403 - mul_66 = x_403 = None - view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_67 = x_404 * view_25 - view_25 = None - layer_norm_58 = torch.nn.functional.layer_norm( - x_404, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_404 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = (None) - x_405 = torch._C._nn.linear( - layer_norm_58, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_58 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_42 = torch.nn.functional.relu(x_405, inplace=False) - x_405 = None - pow_43 = relu_42**2 - relu_42 = None - mul_68 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ - * pow_43 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = ( - pow_43 - ) = None - x_406 = ( - mul_68 - + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ - ) - mul_68 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = (None) - x_407 = torch.nn.functional.dropout(x_406, 0.0, False, False) - x_406 = None - x_408 = torch._C._nn.linear( - x_407, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_407 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_409 = torch.nn.functional.dropout(x_408, 0.0, False, False) - x_408 = None - x_410 = mul_67 + x_409 - mul_67 = x_409 = None - view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_69 = x_410 * view_26 - view_26 = None - layer_norm_59 = torch.nn.functional.layer_norm( - x_410, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_410 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = (None) - linear_52 = torch._C._nn.linear( - layer_norm_59, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_59 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_27 = linear_52.reshape(1, pow_52, 3, 16, 32) - linear_52 = None - qkv_13 = reshape_27.permute(2, 0, 3, 1, 4) - reshape_27 = None - unbind_13 = qkv_13.unbind(0) - qkv_13 = None - q_13 = unbind_13[0] - k_13 = unbind_13[1] - v_13 = unbind_13[2] - unbind_13 = None - x_411 = torch._C._nn.scaled_dot_product_attention( - q_13, k_13, v_13, dropout_p=0.0 - ) - q_13 = k_13 = v_13 = None - transpose_14 = x_411.transpose(1, 2) - x_411 = None - x_412 = transpose_14.reshape(1, pow_52, 512) - transpose_14 = None - x_413 = torch._C._nn.linear( - x_412, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_412 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_414 = torch.nn.functional.dropout(x_413, 0.0, False, False) - x_413 = None - x_415 = mul_69 + x_414 - mul_69 = x_414 = None - view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_70 = x_415 * view_27 - view_27 = None - layer_norm_60 = torch.nn.functional.layer_norm( - x_415, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_415 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = (None) - x_416 = torch._C._nn.linear( - layer_norm_60, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_60 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_43 = torch.nn.functional.relu(x_416, inplace=False) - x_416 = None - pow_44 = relu_43**2 - relu_43 = None - mul_71 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ - * pow_44 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = ( - pow_44 - ) = None - x_417 = ( - mul_71 - + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ - ) - mul_71 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = (None) - x_418 = torch.nn.functional.dropout(x_417, 0.0, False, False) - x_417 = None - x_419 = torch._C._nn.linear( - x_418, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_418 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_420 = torch.nn.functional.dropout(x_419, 0.0, False, False) - x_419 = None - x_421 = mul_70 + x_420 - mul_70 = x_420 = None - view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_72 = x_421 * view_28 - view_28 = None - layer_norm_61 = torch.nn.functional.layer_norm( - x_421, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_421 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = (None) - linear_56 = torch._C._nn.linear( - layer_norm_61, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_61 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_29 = linear_56.reshape(1, pow_52, 3, 16, 32) - linear_56 = None - qkv_14 = reshape_29.permute(2, 0, 3, 1, 4) - reshape_29 = None - unbind_14 = qkv_14.unbind(0) - qkv_14 = None - q_14 = unbind_14[0] - k_14 = unbind_14[1] - v_14 = unbind_14[2] - unbind_14 = None - x_422 = torch._C._nn.scaled_dot_product_attention( - q_14, k_14, v_14, dropout_p=0.0 - ) - q_14 = k_14 = v_14 = None - transpose_15 = x_422.transpose(1, 2) - x_422 = None - x_423 = transpose_15.reshape(1, pow_52, 512) - transpose_15 = None - x_424 = torch._C._nn.linear( - x_423, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_423 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_425 = torch.nn.functional.dropout(x_424, 0.0, False, False) - x_424 = None - x_426 = mul_72 + x_425 - mul_72 = x_425 = None - view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_73 = x_426 * view_29 - view_29 = None - layer_norm_62 = torch.nn.functional.layer_norm( - x_426, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_426 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = (None) - x_427 = torch._C._nn.linear( - layer_norm_62, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_62 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_44 = torch.nn.functional.relu(x_427, inplace=False) - x_427 = None - pow_45 = relu_44**2 - relu_44 = None - mul_74 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ - * pow_45 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = ( - pow_45 - ) = None - x_428 = ( - mul_74 - + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ - ) - mul_74 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = (None) - x_429 = torch.nn.functional.dropout(x_428, 0.0, False, False) - x_428 = None - x_430 = torch._C._nn.linear( - x_429, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_429 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_431 = torch.nn.functional.dropout(x_430, 0.0, False, False) - x_430 = None - x_432 = mul_73 + x_431 - mul_73 = x_431 = None - view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_75 = x_432 * view_30 - view_30 = None - layer_norm_63 = torch.nn.functional.layer_norm( - x_432, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_432 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = (None) - linear_60 = torch._C._nn.linear( - layer_norm_63, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_63 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_31 = linear_60.reshape(1, pow_52, 3, 16, 32) - linear_60 = None - qkv_15 = reshape_31.permute(2, 0, 3, 1, 4) - reshape_31 = None - unbind_15 = qkv_15.unbind(0) - qkv_15 = None - q_15 = unbind_15[0] - k_15 = unbind_15[1] - v_15 = unbind_15[2] - unbind_15 = None - x_433 = torch._C._nn.scaled_dot_product_attention( - q_15, k_15, v_15, dropout_p=0.0 - ) - q_15 = k_15 = v_15 = None - transpose_16 = x_433.transpose(1, 2) - x_433 = None - x_434 = transpose_16.reshape(1, pow_52, 512) - transpose_16 = None - x_435 = torch._C._nn.linear( - x_434, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_434 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_436 = torch.nn.functional.dropout(x_435, 0.0, False, False) - x_435 = None - x_437 = mul_75 + x_436 - mul_75 = x_436 = None - view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_76 = x_437 * view_31 - view_31 = None - layer_norm_64 = torch.nn.functional.layer_norm( - x_437, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_437 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = (None) - x_438 = torch._C._nn.linear( - layer_norm_64, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_64 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_45 = torch.nn.functional.relu(x_438, inplace=False) - x_438 = None - pow_46 = relu_45**2 - relu_45 = None - mul_77 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ - * pow_46 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = ( - pow_46 - ) = None - x_439 = ( - mul_77 - + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ - ) - mul_77 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = (None) - x_440 = torch.nn.functional.dropout(x_439, 0.0, False, False) - x_439 = None - x_441 = torch._C._nn.linear( - x_440, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_440 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_442 = torch.nn.functional.dropout(x_441, 0.0, False, False) - x_441 = None - x_443 = mul_76 + x_442 - mul_76 = x_442 = None - view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_78 = x_443 * view_32 - view_32 = None - layer_norm_65 = torch.nn.functional.layer_norm( - x_443, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_443 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = (None) - linear_64 = torch._C._nn.linear( - layer_norm_65, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_65 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_33 = linear_64.reshape(1, pow_52, 3, 16, 32) - linear_64 = None - qkv_16 = reshape_33.permute(2, 0, 3, 1, 4) - reshape_33 = None - unbind_16 = qkv_16.unbind(0) - qkv_16 = None - q_16 = unbind_16[0] - k_16 = unbind_16[1] - v_16 = unbind_16[2] - unbind_16 = None - x_444 = torch._C._nn.scaled_dot_product_attention( - q_16, k_16, v_16, dropout_p=0.0 - ) - q_16 = k_16 = v_16 = None - transpose_17 = x_444.transpose(1, 2) - x_444 = None - x_445 = transpose_17.reshape(1, pow_52, 512) - transpose_17 = None - x_446 = torch._C._nn.linear( - x_445, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_445 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_447 = torch.nn.functional.dropout(x_446, 0.0, False, False) - x_446 = None - x_448 = mul_78 + x_447 - mul_78 = x_447 = None - view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_79 = x_448 * view_33 - view_33 = None - layer_norm_66 = torch.nn.functional.layer_norm( - x_448, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_448 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = (None) - x_449 = torch._C._nn.linear( - layer_norm_66, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_66 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_46 = torch.nn.functional.relu(x_449, inplace=False) - x_449 = None - pow_47 = relu_46**2 - relu_46 = None - mul_80 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ - * pow_47 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = ( - pow_47 - ) = None - x_450 = ( - mul_80 - + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ - ) - mul_80 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = (None) - x_451 = torch.nn.functional.dropout(x_450, 0.0, False, False) - x_450 = None - x_452 = torch._C._nn.linear( - x_451, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_451 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_453 = torch.nn.functional.dropout(x_452, 0.0, False, False) - x_452 = None - x_454 = mul_79 + x_453 - mul_79 = x_453 = None - view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_81 = x_454 * view_34 - view_34 = None - layer_norm_67 = torch.nn.functional.layer_norm( - x_454, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_454 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = (None) - linear_68 = torch._C._nn.linear( - layer_norm_67, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_67 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_35 = linear_68.reshape(1, pow_52, 3, 16, 32) - linear_68 = None - qkv_17 = reshape_35.permute(2, 0, 3, 1, 4) - reshape_35 = None - unbind_17 = qkv_17.unbind(0) - qkv_17 = None - q_17 = unbind_17[0] - k_17 = unbind_17[1] - v_17 = unbind_17[2] - unbind_17 = None - x_455 = torch._C._nn.scaled_dot_product_attention( - q_17, k_17, v_17, dropout_p=0.0 - ) - q_17 = k_17 = v_17 = None - transpose_18 = x_455.transpose(1, 2) - x_455 = None - x_456 = transpose_18.reshape(1, pow_52, 512) - transpose_18 = pow_52 = None - x_457 = torch._C._nn.linear( - x_456, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_456 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_458 = torch.nn.functional.dropout(x_457, 0.0, False, False) - x_457 = None - x_459 = mul_81 + x_458 - mul_81 = x_458 = None - view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_82 = x_459 * view_35 - view_35 = None - layer_norm_68 = torch.nn.functional.layer_norm( - x_459, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_459 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = (None) - x_460 = torch._C._nn.linear( - layer_norm_68, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_68 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_47 = torch.nn.functional.relu(x_460, inplace=False) - x_460 = None - pow_48 = relu_47**2 - relu_47 = None - mul_83 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ - * pow_48 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = ( - pow_48 - ) = None - x_461 = ( - mul_83 - + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ - ) - mul_83 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = (None) - x_462 = torch.nn.functional.dropout(x_461, 0.0, False, False) - x_461 = None - x_463 = torch._C._nn.linear( - x_462, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_462 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_464 = torch.nn.functional.dropout(x_463, 0.0, False, False) - x_463 = None - x_465 = mul_82 + x_464 - mul_82 = x_464 = None - transpose_19 = x_465.transpose(1, 2) - x_465 = None - x_466 = transpose_19.reshape(1, 512, sym_sum_3, sym_sum_3) - transpose_19 = sym_sum_3 = None - x_467 = x_466.permute(0, 2, 3, 1) - x_466 = None - x_468 = torch.nn.functional.layer_norm( - x_467, - (512,), - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_467 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) - x_469 = x_468.permute(0, 3, 1, 2) - x_468 = None - x_470 = torch.conv2d( - x_469, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_469 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_3 = sym_sum // 32 - sym_sum = None - sym_sum_4 = torch.sym_sum([1, floordiv_3]) - floordiv_3 = None - reshape_38 = x_470.reshape(1, 768, -1) - x_470 = None - x_471 = reshape_38.transpose(1, 2) - reshape_38 = None - view_36 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_84 = x_471 * view_36 - view_36 = None - layer_norm_70 = torch.nn.functional.layer_norm( - x_471, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_471 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_53 = sym_sum_4**2 - linear_72 = torch._C._nn.linear( - layer_norm_70, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_70 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_39 = linear_72.reshape(1, pow_53, 3, 24, 32) - linear_72 = None - qkv_18 = reshape_39.permute(2, 0, 3, 1, 4) - reshape_39 = None - unbind_18 = qkv_18.unbind(0) - qkv_18 = None - q_18 = unbind_18[0] - k_18 = unbind_18[1] - v_18 = unbind_18[2] - unbind_18 = None - x_472 = torch._C._nn.scaled_dot_product_attention( - q_18, k_18, v_18, dropout_p=0.0 - ) - q_18 = k_18 = v_18 = None - transpose_21 = x_472.transpose(1, 2) - x_472 = None - x_473 = transpose_21.reshape(1, pow_53, 768) - transpose_21 = None - x_474 = torch._C._nn.linear( - x_473, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_473 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_475 = torch.nn.functional.dropout(x_474, 0.0, False, False) - x_474 = None - x_476 = mul_84 + x_475 - mul_84 = x_475 = None - view_37 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_85 = x_476 * view_37 - view_37 = None - layer_norm_71 = torch.nn.functional.layer_norm( - x_476, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_476 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_477 = torch._C._nn.linear( - layer_norm_71, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_71 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_48 = torch.nn.functional.relu(x_477, inplace=False) - x_477 = None - pow_49 = relu_48**2 - relu_48 = None - mul_86 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_49 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_49 - ) = None - x_478 = ( - mul_86 - + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_86 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_479 = torch.nn.functional.dropout(x_478, 0.0, False, False) - x_478 = None - x_480 = torch._C._nn.linear( - x_479, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_479 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_481 = torch.nn.functional.dropout(x_480, 0.0, False, False) - x_480 = None - x_482 = mul_85 + x_481 - mul_85 = x_481 = None - view_38 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_87 = x_482 * view_38 - view_38 = None - layer_norm_72 = torch.nn.functional.layer_norm( - x_482, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_482 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_76 = torch._C._nn.linear( - layer_norm_72, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_72 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_41 = linear_76.reshape(1, pow_53, 3, 24, 32) - linear_76 = None - qkv_19 = reshape_41.permute(2, 0, 3, 1, 4) - reshape_41 = None - unbind_19 = qkv_19.unbind(0) - qkv_19 = None - q_19 = unbind_19[0] - k_19 = unbind_19[1] - v_19 = unbind_19[2] - unbind_19 = None - x_483 = torch._C._nn.scaled_dot_product_attention( - q_19, k_19, v_19, dropout_p=0.0 - ) - q_19 = k_19 = v_19 = None - transpose_22 = x_483.transpose(1, 2) - x_483 = None - x_484 = transpose_22.reshape(1, pow_53, 768) - transpose_22 = None - x_485 = torch._C._nn.linear( - x_484, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_484 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_486 = torch.nn.functional.dropout(x_485, 0.0, False, False) - x_485 = None - x_487 = mul_87 + x_486 - mul_87 = x_486 = None - view_39 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_88 = x_487 * view_39 - view_39 = None - layer_norm_73 = torch.nn.functional.layer_norm( - x_487, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_487 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_488 = torch._C._nn.linear( - layer_norm_73, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_73 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_49 = torch.nn.functional.relu(x_488, inplace=False) - x_488 = None - pow_50 = relu_49**2 - relu_49 = None - mul_89 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_50 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_50 - ) = None - x_489 = ( - mul_89 - + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_89 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_490 = torch.nn.functional.dropout(x_489, 0.0, False, False) - x_489 = None - x_491 = torch._C._nn.linear( - x_490, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_490 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_492 = torch.nn.functional.dropout(x_491, 0.0, False, False) - x_491 = None - x_493 = mul_88 + x_492 - mul_88 = x_492 = None - view_40 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_90 = x_493 * view_40 - view_40 = None - layer_norm_74 = torch.nn.functional.layer_norm( - x_493, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_493 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_80 = torch._C._nn.linear( - layer_norm_74, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_74 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_43 = linear_80.reshape(1, pow_53, 3, 24, 32) - linear_80 = None - qkv_20 = reshape_43.permute(2, 0, 3, 1, 4) - reshape_43 = None - unbind_20 = qkv_20.unbind(0) - qkv_20 = None - q_20 = unbind_20[0] - k_20 = unbind_20[1] - v_20 = unbind_20[2] - unbind_20 = None - x_494 = torch._C._nn.scaled_dot_product_attention( - q_20, k_20, v_20, dropout_p=0.0 - ) - q_20 = k_20 = v_20 = None - transpose_23 = x_494.transpose(1, 2) - x_494 = None - x_495 = transpose_23.reshape(1, pow_53, 768) - transpose_23 = pow_53 = None - x_496 = torch._C._nn.linear( - x_495, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_495 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_497 = torch.nn.functional.dropout(x_496, 0.0, False, False) - x_496 = None - x_498 = mul_90 + x_497 - mul_90 = x_497 = None - view_41 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_91 = x_498 * view_41 - view_41 = None - layer_norm_75 = torch.nn.functional.layer_norm( - x_498, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_498 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_499 = torch._C._nn.linear( - layer_norm_75, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_75 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_50 = torch.nn.functional.relu(x_499, inplace=False) - x_499 = None - pow_51 = relu_50**2 - relu_50 = None - mul_92 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_51 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_51 - ) = None - x_500 = ( - mul_92 - + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_92 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_501 = torch.nn.functional.dropout(x_500, 0.0, False, False) - x_500 = None - x_502 = torch._C._nn.linear( - x_501, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_501 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_503 = torch.nn.functional.dropout(x_502, 0.0, False, False) - x_502 = None - x_504 = mul_91 + x_503 - mul_91 = x_503 = None - transpose_24 = x_504.transpose(1, 2) - x_504 = None - x_505 = transpose_24.reshape(1, 768, sym_sum_4, sym_sum_4) - transpose_24 = sym_sum_4 = None - x_506 = torch.nn.functional.adaptive_avg_pool2d(x_505, 1) - x_505 = None - x_507 = x_506.permute(0, 2, 3, 1) - x_506 = None - x_508 = torch.nn.functional.layer_norm( - x_507, - (768,), - l_self_modules_head_modules_norm_parameters_weight_, - l_self_modules_head_modules_norm_parameters_bias_, - 1e-06, - ) - x_507 = ( - l_self_modules_head_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_norm_parameters_bias_ = None - x_509 = x_508.permute(0, 3, 1, 2) - x_508 = None - x_510 = x_509.flatten(1, -1) - x_509 = None - x_511 = torch.nn.functional.dropout(x_510, 0.0, False, False) - x_510 = None - x_512 = torch._C._nn.linear( - x_511, - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, - ) - x_511 = ( - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None - relu_51 = torch.nn.functional.relu(x_512, inplace=False) - x_512 = None - x_513 = torch.square(relu_51) - relu_51 = None - x_514 = torch.nn.functional.layer_norm( - x_513, - (3072,), - l_self_modules_head_modules_fc_modules_norm_parameters_weight_, - l_self_modules_head_modules_fc_modules_norm_parameters_bias_, - 1e-06, - ) - x_513 = ( - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None - x_515 = torch.nn.functional.dropout(x_514, 0.0, False, False) - x_514 = None - x_516 = torch._C._nn.linear( - x_515, - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, - ) - x_515 = ( - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None - return (x_516,) diff --git a/samples/timm/caformer_b36.sail_in22k_ft_in1k/weight_meta.py b/samples/timm/caformer_b36.sail_in22k_ft_in1k/weight_meta.py deleted file mode 100644 index 279c30e841..0000000000 --- a/samples/timm/caformer_b36.sail_in22k_ft_in1k/weight_meta.py +++ /dev/null @@ -1,3968 +0,0 @@ -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: - name = "L_self_modules_stem_modules_conv_parameters_weight_" - shape = [128, 3, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: - name = "L_self_modules_stem_modules_conv_parameters_bias_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_s1: - name = "s1" - shape = [] - dtype = "torch.int64" - device = "cpu" - mean = None - std = None - data = [4] - - -class Program_weight_tensor_meta_L_x_: - name = "L_x_" - shape = [1, 3, 224, 224] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.231 - std = 1.285 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: - name = "L_self_modules_stem_modules_norm_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [256, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [256, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [128, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [512, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [128, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [256, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [256, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [128, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [512, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [128, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [256, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [256, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [128, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [512, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [128, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" - shape = [256, 128, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" - shape = [512, 256, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" - shape = [768, 512, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [2304, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [768, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [3072, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [768, 3072] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [2304, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [768, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [3072, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [768, 3072] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [2304, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [768, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [3072, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [768, 3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_norm_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_norm_parameters_bias_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" - shape = [3072, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" - shape = [1000, 3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" - shape = [1000] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None diff --git a/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/graph_net.json b/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/graph_net.json deleted file mode 100644 index 1373fe3b58..0000000000 --- a/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/graph_net.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "framework": "torch", - "num_devices_required": 1, - "num_nodes_required": 1 -} \ No newline at end of file diff --git a/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/input_meta.py b/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/input_meta.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/input_tensor_constraints.py b/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/input_tensor_constraints.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/model.py b/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/model.py deleted file mode 100644 index 08d3de35cc..0000000000 --- a/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/model.py +++ /dev/null @@ -1,4758 +0,0 @@ -import torch - - -class GraphModule(torch.nn.Module): - def forward( - self, - L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - s1: torch.SymInt, - L_x_: torch.Tensor, - L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - ): - l_self_modules_stem_modules_conv_parameters_weight_ = ( - L_self_modules_stem_modules_conv_parameters_weight_ - ) - l_self_modules_stem_modules_conv_parameters_bias_ = ( - L_self_modules_stem_modules_conv_parameters_bias_ - ) - l_x_ = L_x_ - l_self_modules_stem_modules_norm_parameters_weight_ = ( - L_self_modules_stem_modules_norm_parameters_weight_ - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_head_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ - ) - x = torch.conv2d( - l_x_, - l_self_modules_stem_modules_conv_parameters_weight_, - l_self_modules_stem_modules_conv_parameters_bias_, - (4, 4), - (2, 2), - (1, 1), - 1, - ) - l_x_ = ( - l_self_modules_stem_modules_conv_parameters_weight_ - ) = l_self_modules_stem_modules_conv_parameters_bias_ = None - x_1 = x.permute(0, 2, 3, 1) - x = None - x_2 = torch.nn.functional.layer_norm( - x_1, - (128,), - l_self_modules_stem_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None - x_3 = x_2.permute(0, 3, 1, 2) - x_2 = None - sym_sum = torch.sym_sum([-3, s1]) - s1 = None - floordiv = sym_sum // 4 - sym_sum_1 = torch.sym_sum([1, floordiv]) - floordiv = sym_sum_1 = None - x_4 = x_3.permute(0, 2, 3, 1) - x_5 = torch.nn.functional.layer_norm( - x_4, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_6 = x_5.permute(0, 3, 1, 2) - x_5 = None - x_7 = torch.conv2d( - x_6, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu = torch.nn.functional.relu(x_7, inplace=False) - x_7 = None - pow_1 = relu**2 - relu = None - mul = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_1 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_1 - ) = None - x_8 = ( - mul - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_9 = torch.conv2d( - x_8, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 256, - ) - x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_10 = torch.conv2d( - x_9, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_11 = x_3 + x_10 - x_3 = x_10 = None - x_12 = x_11.permute(0, 2, 3, 1) - x_13 = torch.nn.functional.layer_norm( - x_12, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_14 = x_13.permute(0, 3, 1, 2) - x_13 = None - x_15 = torch.conv2d( - x_14, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_1 = torch.nn.functional.relu(x_15, inplace=False) - x_15 = None - pow_2 = relu_1**2 - relu_1 = None - mul_1 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_2 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_2 - ) = None - x_16 = ( - mul_1 - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) - x_16 = None - x_18 = torch.conv2d( - x_17, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) - x_18 = None - x_20 = x_11 + x_19 - x_11 = x_19 = None - x_21 = x_20.permute(0, 2, 3, 1) - x_22 = torch.nn.functional.layer_norm( - x_21, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_23 = x_22.permute(0, 3, 1, 2) - x_22 = None - x_24 = torch.conv2d( - x_23, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_2 = torch.nn.functional.relu(x_24, inplace=False) - x_24 = None - pow_3 = relu_2**2 - relu_2 = None - mul_2 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_3 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_3 - ) = None - x_25 = ( - mul_2 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_26 = torch.conv2d( - x_25, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 256, - ) - x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_27 = torch.conv2d( - x_26, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_28 = x_20 + x_27 - x_20 = x_27 = None - x_29 = x_28.permute(0, 2, 3, 1) - x_30 = torch.nn.functional.layer_norm( - x_29, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_31 = x_30.permute(0, 3, 1, 2) - x_30 = None - x_32 = torch.conv2d( - x_31, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_3 = torch.nn.functional.relu(x_32, inplace=False) - x_32 = None - pow_4 = relu_3**2 - relu_3 = None - mul_3 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_4 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_4 - ) = None - x_33 = ( - mul_3 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) - x_33 = None - x_35 = torch.conv2d( - x_34, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) - x_35 = None - x_37 = x_28 + x_36 - x_28 = x_36 = None - x_38 = x_37.permute(0, 2, 3, 1) - x_39 = torch.nn.functional.layer_norm( - x_38, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_40 = x_39.permute(0, 3, 1, 2) - x_39 = None - x_41 = torch.conv2d( - x_40, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_4 = torch.nn.functional.relu(x_41, inplace=False) - x_41 = None - pow_5 = relu_4**2 - relu_4 = None - mul_4 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_5 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_5 - ) = None - x_42 = ( - mul_4 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_43 = torch.conv2d( - x_42, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 256, - ) - x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_44 = torch.conv2d( - x_43, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_45 = x_37 + x_44 - x_37 = x_44 = None - x_46 = x_45.permute(0, 2, 3, 1) - x_47 = torch.nn.functional.layer_norm( - x_46, - (128,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_48 = x_47.permute(0, 3, 1, 2) - x_47 = None - x_49 = torch.conv2d( - x_48, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_5 = torch.nn.functional.relu(x_49, inplace=False) - x_49 = None - pow_6 = relu_5**2 - relu_5 = None - mul_5 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_6 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_6 - ) = None - x_50 = ( - mul_5 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) - x_50 = None - x_52 = torch.conv2d( - x_51, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) - x_52 = None - x_54 = x_45 + x_53 - x_45 = x_53 = None - x_55 = x_54.permute(0, 2, 3, 1) - x_54 = None - x_56 = torch.nn.functional.layer_norm( - x_55, - (128,), - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) - x_57 = x_56.permute(0, 3, 1, 2) - x_56 = None - x_58 = torch.conv2d( - x_57, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_1 = sym_sum // 8 - sym_sum_2 = torch.sym_sum([1, floordiv_1]) - floordiv_1 = sym_sum_2 = None - x_59 = x_58.permute(0, 2, 3, 1) - x_60 = torch.nn.functional.layer_norm( - x_59, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_61 = x_60.permute(0, 3, 1, 2) - x_60 = None - x_62 = torch.conv2d( - x_61, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_6 = torch.nn.functional.relu(x_62, inplace=False) - x_62 = None - pow_7 = relu_6**2 - relu_6 = None - mul_6 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_7 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_7 - ) = None - x_63 = ( - mul_6 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_64 = torch.conv2d( - x_63, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_65 = torch.conv2d( - x_64, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_66 = x_58 + x_65 - x_58 = x_65 = None - x_67 = x_66.permute(0, 2, 3, 1) - x_68 = torch.nn.functional.layer_norm( - x_67, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_69 = x_68.permute(0, 3, 1, 2) - x_68 = None - x_70 = torch.conv2d( - x_69, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_7 = torch.nn.functional.relu(x_70, inplace=False) - x_70 = None - pow_8 = relu_7**2 - relu_7 = None - mul_7 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_8 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_8 - ) = None - x_71 = ( - mul_7 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) - x_71 = None - x_73 = torch.conv2d( - x_72, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) - x_73 = None - x_75 = x_66 + x_74 - x_66 = x_74 = None - x_76 = x_75.permute(0, 2, 3, 1) - x_77 = torch.nn.functional.layer_norm( - x_76, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_78 = x_77.permute(0, 3, 1, 2) - x_77 = None - x_79 = torch.conv2d( - x_78, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_8 = torch.nn.functional.relu(x_79, inplace=False) - x_79 = None - pow_9 = relu_8**2 - relu_8 = None - mul_8 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_9 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_9 - ) = None - x_80 = ( - mul_8 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_81 = torch.conv2d( - x_80, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_82 = torch.conv2d( - x_81, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_83 = x_75 + x_82 - x_75 = x_82 = None - x_84 = x_83.permute(0, 2, 3, 1) - x_85 = torch.nn.functional.layer_norm( - x_84, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_86 = x_85.permute(0, 3, 1, 2) - x_85 = None - x_87 = torch.conv2d( - x_86, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_9 = torch.nn.functional.relu(x_87, inplace=False) - x_87 = None - pow_10 = relu_9**2 - relu_9 = None - mul_9 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_10 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_10 - ) = None - x_88 = ( - mul_9 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) - x_88 = None - x_90 = torch.conv2d( - x_89, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) - x_90 = None - x_92 = x_83 + x_91 - x_83 = x_91 = None - x_93 = x_92.permute(0, 2, 3, 1) - x_94 = torch.nn.functional.layer_norm( - x_93, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_95 = x_94.permute(0, 3, 1, 2) - x_94 = None - x_96 = torch.conv2d( - x_95, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_10 = torch.nn.functional.relu(x_96, inplace=False) - x_96 = None - pow_11 = relu_10**2 - relu_10 = None - mul_10 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_11 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_11 - ) = None - x_97 = ( - mul_10 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_98 = torch.conv2d( - x_97, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_99 = torch.conv2d( - x_98, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_100 = x_92 + x_99 - x_92 = x_99 = None - x_101 = x_100.permute(0, 2, 3, 1) - x_102 = torch.nn.functional.layer_norm( - x_101, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_103 = x_102.permute(0, 3, 1, 2) - x_102 = None - x_104 = torch.conv2d( - x_103, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_11 = torch.nn.functional.relu(x_104, inplace=False) - x_104 = None - pow_12 = relu_11**2 - relu_11 = None - mul_11 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_12 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_12 - ) = None - x_105 = ( - mul_11 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) - x_105 = None - x_107 = torch.conv2d( - x_106, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) - x_107 = None - x_109 = x_100 + x_108 - x_100 = x_108 = None - x_110 = x_109.permute(0, 2, 3, 1) - x_111 = torch.nn.functional.layer_norm( - x_110, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_110 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) - x_112 = x_111.permute(0, 3, 1, 2) - x_111 = None - x_113 = torch.conv2d( - x_112, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_112 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_12 = torch.nn.functional.relu(x_113, inplace=False) - x_113 = None - pow_13 = relu_12**2 - relu_12 = None - mul_12 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ - * pow_13 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_13 - ) = None - x_114 = ( - mul_12 - + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_12 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_115 = torch.conv2d( - x_114, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_114 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_116 = torch.conv2d( - x_115, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_115 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_117 = x_109 + x_116 - x_109 = x_116 = None - x_118 = x_117.permute(0, 2, 3, 1) - x_119 = torch.nn.functional.layer_norm( - x_118, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_118 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) - x_120 = x_119.permute(0, 3, 1, 2) - x_119 = None - x_121 = torch.conv2d( - x_120, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_120 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_13 = torch.nn.functional.relu(x_121, inplace=False) - x_121 = None - pow_14 = relu_13**2 - relu_13 = None - mul_13 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - * pow_14 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( - pow_14 - ) = None - x_122 = ( - mul_13 - + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - ) - mul_13 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) - x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) - x_122 = None - x_124 = torch.conv2d( - x_123, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_123 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) - x_124 = None - x_126 = x_117 + x_125 - x_117 = x_125 = None - x_127 = x_126.permute(0, 2, 3, 1) - x_128 = torch.nn.functional.layer_norm( - x_127, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_127 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) - x_129 = x_128.permute(0, 3, 1, 2) - x_128 = None - x_130 = torch.conv2d( - x_129, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_129 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_14 = torch.nn.functional.relu(x_130, inplace=False) - x_130 = None - pow_15 = relu_14**2 - relu_14 = None - mul_14 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ - * pow_15 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_15 - ) = None - x_131 = ( - mul_14 - + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_14 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_132 = torch.conv2d( - x_131, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_131 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_133 = torch.conv2d( - x_132, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_132 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_134 = x_126 + x_133 - x_126 = x_133 = None - x_135 = x_134.permute(0, 2, 3, 1) - x_136 = torch.nn.functional.layer_norm( - x_135, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_135 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) - x_137 = x_136.permute(0, 3, 1, 2) - x_136 = None - x_138 = torch.conv2d( - x_137, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_137 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_15 = torch.nn.functional.relu(x_138, inplace=False) - x_138 = None - pow_16 = relu_15**2 - relu_15 = None - mul_15 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - * pow_16 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( - pow_16 - ) = None - x_139 = ( - mul_15 - + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - ) - mul_15 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) - x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) - x_139 = None - x_141 = torch.conv2d( - x_140, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_140 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) - x_141 = None - x_143 = x_134 + x_142 - x_134 = x_142 = None - x_144 = x_143.permute(0, 2, 3, 1) - x_145 = torch.nn.functional.layer_norm( - x_144, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_144 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) - x_146 = x_145.permute(0, 3, 1, 2) - x_145 = None - x_147 = torch.conv2d( - x_146, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_146 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_16 = torch.nn.functional.relu(x_147, inplace=False) - x_147 = None - pow_17 = relu_16**2 - relu_16 = None - mul_16 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ - * pow_17 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_17 - ) = None - x_148 = ( - mul_16 - + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_16 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_149 = torch.conv2d( - x_148, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_148 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_150 = torch.conv2d( - x_149, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_149 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_151 = x_143 + x_150 - x_143 = x_150 = None - x_152 = x_151.permute(0, 2, 3, 1) - x_153 = torch.nn.functional.layer_norm( - x_152, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_152 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) - x_154 = x_153.permute(0, 3, 1, 2) - x_153 = None - x_155 = torch.conv2d( - x_154, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_154 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_17 = torch.nn.functional.relu(x_155, inplace=False) - x_155 = None - pow_18 = relu_17**2 - relu_17 = None - mul_17 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - * pow_18 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( - pow_18 - ) = None - x_156 = ( - mul_17 - + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - ) - mul_17 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) - x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) - x_156 = None - x_158 = torch.conv2d( - x_157, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_157 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) - x_158 = None - x_160 = x_151 + x_159 - x_151 = x_159 = None - x_161 = x_160.permute(0, 2, 3, 1) - x_162 = torch.nn.functional.layer_norm( - x_161, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_161 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) - x_163 = x_162.permute(0, 3, 1, 2) - x_162 = None - x_164 = torch.conv2d( - x_163, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_163 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_18 = torch.nn.functional.relu(x_164, inplace=False) - x_164 = None - pow_19 = relu_18**2 - relu_18 = None - mul_18 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ - * pow_19 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_19 - ) = None - x_165 = ( - mul_18 - + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_18 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_166 = torch.conv2d( - x_165, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_165 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_167 = torch.conv2d( - x_166, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_166 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_168 = x_160 + x_167 - x_160 = x_167 = None - x_169 = x_168.permute(0, 2, 3, 1) - x_170 = torch.nn.functional.layer_norm( - x_169, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_169 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) - x_171 = x_170.permute(0, 3, 1, 2) - x_170 = None - x_172 = torch.conv2d( - x_171, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_171 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_19 = torch.nn.functional.relu(x_172, inplace=False) - x_172 = None - pow_20 = relu_19**2 - relu_19 = None - mul_19 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - * pow_20 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( - pow_20 - ) = None - x_173 = ( - mul_19 - + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - ) - mul_19 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) - x_174 = torch.nn.functional.dropout(x_173, 0.0, False, False) - x_173 = None - x_175 = torch.conv2d( - x_174, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_174 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) - x_175 = None - x_177 = x_168 + x_176 - x_168 = x_176 = None - x_178 = x_177.permute(0, 2, 3, 1) - x_179 = torch.nn.functional.layer_norm( - x_178, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_178 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) - x_180 = x_179.permute(0, 3, 1, 2) - x_179 = None - x_181 = torch.conv2d( - x_180, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_180 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_20 = torch.nn.functional.relu(x_181, inplace=False) - x_181 = None - pow_21 = relu_20**2 - relu_20 = None - mul_20 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ - * pow_21 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_21 - ) = None - x_182 = ( - mul_20 - + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_20 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_183 = torch.conv2d( - x_182, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_182 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_184 = torch.conv2d( - x_183, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_183 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_185 = x_177 + x_184 - x_177 = x_184 = None - x_186 = x_185.permute(0, 2, 3, 1) - x_187 = torch.nn.functional.layer_norm( - x_186, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_186 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) - x_188 = x_187.permute(0, 3, 1, 2) - x_187 = None - x_189 = torch.conv2d( - x_188, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_188 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_21 = torch.nn.functional.relu(x_189, inplace=False) - x_189 = None - pow_22 = relu_21**2 - relu_21 = None - mul_21 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - * pow_22 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( - pow_22 - ) = None - x_190 = ( - mul_21 - + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - ) - mul_21 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) - x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) - x_190 = None - x_192 = torch.conv2d( - x_191, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_191 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) - x_192 = None - x_194 = x_185 + x_193 - x_185 = x_193 = None - x_195 = x_194.permute(0, 2, 3, 1) - x_196 = torch.nn.functional.layer_norm( - x_195, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_195 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) - x_197 = x_196.permute(0, 3, 1, 2) - x_196 = None - x_198 = torch.conv2d( - x_197, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_197 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_22 = torch.nn.functional.relu(x_198, inplace=False) - x_198 = None - pow_23 = relu_22**2 - relu_22 = None - mul_22 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ - * pow_23 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_23 - ) = None - x_199 = ( - mul_22 - + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_22 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_200 = torch.conv2d( - x_199, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_199 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_201 = torch.conv2d( - x_200, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_200 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_202 = x_194 + x_201 - x_194 = x_201 = None - x_203 = x_202.permute(0, 2, 3, 1) - x_204 = torch.nn.functional.layer_norm( - x_203, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_203 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) - x_205 = x_204.permute(0, 3, 1, 2) - x_204 = None - x_206 = torch.conv2d( - x_205, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_205 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_23 = torch.nn.functional.relu(x_206, inplace=False) - x_206 = None - pow_24 = relu_23**2 - relu_23 = None - mul_23 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - * pow_24 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( - pow_24 - ) = None - x_207 = ( - mul_23 - + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - ) - mul_23 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) - x_208 = torch.nn.functional.dropout(x_207, 0.0, False, False) - x_207 = None - x_209 = torch.conv2d( - x_208, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_208 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) - x_209 = None - x_211 = x_202 + x_210 - x_202 = x_210 = None - x_212 = x_211.permute(0, 2, 3, 1) - x_213 = torch.nn.functional.layer_norm( - x_212, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_212 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) - x_214 = x_213.permute(0, 3, 1, 2) - x_213 = None - x_215 = torch.conv2d( - x_214, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_214 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_24 = torch.nn.functional.relu(x_215, inplace=False) - x_215 = None - pow_25 = relu_24**2 - relu_24 = None - mul_24 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ - * pow_25 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_25 - ) = None - x_216 = ( - mul_24 - + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_24 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_217 = torch.conv2d( - x_216, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_216 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_218 = torch.conv2d( - x_217, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_217 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_219 = x_211 + x_218 - x_211 = x_218 = None - x_220 = x_219.permute(0, 2, 3, 1) - x_221 = torch.nn.functional.layer_norm( - x_220, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_220 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) - x_222 = x_221.permute(0, 3, 1, 2) - x_221 = None - x_223 = torch.conv2d( - x_222, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_222 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_25 = torch.nn.functional.relu(x_223, inplace=False) - x_223 = None - pow_26 = relu_25**2 - relu_25 = None - mul_25 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - * pow_26 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( - pow_26 - ) = None - x_224 = ( - mul_25 - + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - ) - mul_25 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) - x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) - x_224 = None - x_226 = torch.conv2d( - x_225, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_225 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) - x_226 = None - x_228 = x_219 + x_227 - x_219 = x_227 = None - x_229 = x_228.permute(0, 2, 3, 1) - x_230 = torch.nn.functional.layer_norm( - x_229, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_229 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) - x_231 = x_230.permute(0, 3, 1, 2) - x_230 = None - x_232 = torch.conv2d( - x_231, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_231 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_26 = torch.nn.functional.relu(x_232, inplace=False) - x_232 = None - pow_27 = relu_26**2 - relu_26 = None - mul_26 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ - * pow_27 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_27 - ) = None - x_233 = ( - mul_26 - + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_26 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_234 = torch.conv2d( - x_233, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_233 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_235 = torch.conv2d( - x_234, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_234 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_236 = x_228 + x_235 - x_228 = x_235 = None - x_237 = x_236.permute(0, 2, 3, 1) - x_238 = torch.nn.functional.layer_norm( - x_237, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_237 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) - x_239 = x_238.permute(0, 3, 1, 2) - x_238 = None - x_240 = torch.conv2d( - x_239, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_239 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_27 = torch.nn.functional.relu(x_240, inplace=False) - x_240 = None - pow_28 = relu_27**2 - relu_27 = None - mul_27 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - * pow_28 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( - pow_28 - ) = None - x_241 = ( - mul_27 - + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - ) - mul_27 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) - x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) - x_241 = None - x_243 = torch.conv2d( - x_242, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_242 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_244 = torch.nn.functional.dropout(x_243, 0.0, False, False) - x_243 = None - x_245 = x_236 + x_244 - x_236 = x_244 = None - x_246 = x_245.permute(0, 2, 3, 1) - x_247 = torch.nn.functional.layer_norm( - x_246, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_246 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) - x_248 = x_247.permute(0, 3, 1, 2) - x_247 = None - x_249 = torch.conv2d( - x_248, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_248 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_28 = torch.nn.functional.relu(x_249, inplace=False) - x_249 = None - pow_29 = relu_28**2 - relu_28 = None - mul_28 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ - * pow_29 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_29 - ) = None - x_250 = ( - mul_28 - + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_28 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_251 = torch.conv2d( - x_250, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 512, - ) - x_250 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_252 = torch.conv2d( - x_251, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_251 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_253 = x_245 + x_252 - x_245 = x_252 = None - x_254 = x_253.permute(0, 2, 3, 1) - x_255 = torch.nn.functional.layer_norm( - x_254, - (256,), - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_254 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) - x_256 = x_255.permute(0, 3, 1, 2) - x_255 = None - x_257 = torch.conv2d( - x_256, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_256 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_29 = torch.nn.functional.relu(x_257, inplace=False) - x_257 = None - pow_30 = relu_29**2 - relu_29 = None - mul_29 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - * pow_30 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( - pow_30 - ) = None - x_258 = ( - mul_29 - + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - ) - mul_29 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) - x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) - x_258 = None - x_260 = torch.conv2d( - x_259, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_259 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_261 = torch.nn.functional.dropout(x_260, 0.0, False, False) - x_260 = None - x_262 = x_253 + x_261 - x_253 = x_261 = None - x_263 = x_262.permute(0, 2, 3, 1) - x_262 = None - x_264 = torch.nn.functional.layer_norm( - x_263, - (256,), - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_263 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) - x_265 = x_264.permute(0, 3, 1, 2) - x_264 = None - x_266 = torch.conv2d( - x_265, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_265 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_2 = sym_sum // 16 - sym_sum_3 = torch.sym_sum([1, floordiv_2]) - floordiv_2 = None - reshape = x_266.reshape(1, 512, -1) - x_266 = None - x_267 = reshape.transpose(1, 2) - reshape = None - view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_30 = x_267 * view - view = None - layer_norm_33 = torch.nn.functional.layer_norm( - x_267, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_52 = sym_sum_3**2 - linear = torch._C._nn.linear( - layer_norm_33, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_33 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_1 = linear.reshape(1, pow_52, 3, 16, 32) - linear = None - qkv = reshape_1.permute(2, 0, 3, 1, 4) - reshape_1 = None - unbind = qkv.unbind(0) - qkv = None - q = unbind[0] - k = unbind[1] - v = unbind[2] - unbind = None - x_268 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) - q = k = v = None - transpose_1 = x_268.transpose(1, 2) - x_268 = None - x_269 = transpose_1.reshape(1, pow_52, 512) - transpose_1 = None - x_270 = torch._C._nn.linear( - x_269, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_269 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_271 = torch.nn.functional.dropout(x_270, 0.0, False, False) - x_270 = None - x_272 = mul_30 + x_271 - mul_30 = x_271 = None - view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_31 = x_272 * view_1 - view_1 = None - layer_norm_34 = torch.nn.functional.layer_norm( - x_272, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_272 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_273 = torch._C._nn.linear( - layer_norm_34, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_34 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_30 = torch.nn.functional.relu(x_273, inplace=False) - x_273 = None - pow_31 = relu_30**2 - relu_30 = None - mul_32 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_31 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_31 - ) = None - x_274 = ( - mul_32 - + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_275 = torch.nn.functional.dropout(x_274, 0.0, False, False) - x_274 = None - x_276 = torch._C._nn.linear( - x_275, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_277 = torch.nn.functional.dropout(x_276, 0.0, False, False) - x_276 = None - x_278 = mul_31 + x_277 - mul_31 = x_277 = None - view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_33 = x_278 * view_2 - view_2 = None - layer_norm_35 = torch.nn.functional.layer_norm( - x_278, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_4 = torch._C._nn.linear( - layer_norm_35, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_3 = linear_4.reshape(1, pow_52, 3, 16, 32) - linear_4 = None - qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) - reshape_3 = None - unbind_1 = qkv_1.unbind(0) - qkv_1 = None - q_1 = unbind_1[0] - k_1 = unbind_1[1] - v_1 = unbind_1[2] - unbind_1 = None - x_279 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) - q_1 = k_1 = v_1 = None - transpose_2 = x_279.transpose(1, 2) - x_279 = None - x_280 = transpose_2.reshape(1, pow_52, 512) - transpose_2 = None - x_281 = torch._C._nn.linear( - x_280, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_280 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) - x_281 = None - x_283 = mul_33 + x_282 - mul_33 = x_282 = None - view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_34 = x_283 * view_3 - view_3 = None - layer_norm_36 = torch.nn.functional.layer_norm( - x_283, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_283 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_284 = torch._C._nn.linear( - layer_norm_36, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_36 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_31 = torch.nn.functional.relu(x_284, inplace=False) - x_284 = None - pow_32 = relu_31**2 - relu_31 = None - mul_35 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_32 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_32 - ) = None - x_285 = ( - mul_35 - + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_286 = torch.nn.functional.dropout(x_285, 0.0, False, False) - x_285 = None - x_287 = torch._C._nn.linear( - x_286, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) - x_287 = None - x_289 = mul_34 + x_288 - mul_34 = x_288 = None - view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_36 = x_289 * view_4 - view_4 = None - layer_norm_37 = torch.nn.functional.layer_norm( - x_289, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_8 = torch._C._nn.linear( - layer_norm_37, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_37 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_5 = linear_8.reshape(1, pow_52, 3, 16, 32) - linear_8 = None - qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) - reshape_5 = None - unbind_2 = qkv_2.unbind(0) - qkv_2 = None - q_2 = unbind_2[0] - k_2 = unbind_2[1] - v_2 = unbind_2[2] - unbind_2 = None - x_290 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) - q_2 = k_2 = v_2 = None - transpose_3 = x_290.transpose(1, 2) - x_290 = None - x_291 = transpose_3.reshape(1, pow_52, 512) - transpose_3 = None - x_292 = torch._C._nn.linear( - x_291, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_291 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_293 = torch.nn.functional.dropout(x_292, 0.0, False, False) - x_292 = None - x_294 = mul_36 + x_293 - mul_36 = x_293 = None - view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_37 = x_294 * view_5 - view_5 = None - layer_norm_38 = torch.nn.functional.layer_norm( - x_294, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_294 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_295 = torch._C._nn.linear( - layer_norm_38, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_32 = torch.nn.functional.relu(x_295, inplace=False) - x_295 = None - pow_33 = relu_32**2 - relu_32 = None - mul_38 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_33 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_33 - ) = None - x_296 = ( - mul_38 - + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_297 = torch.nn.functional.dropout(x_296, 0.0, False, False) - x_296 = None - x_298 = torch._C._nn.linear( - x_297, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_299 = torch.nn.functional.dropout(x_298, 0.0, False, False) - x_298 = None - x_300 = mul_37 + x_299 - mul_37 = x_299 = None - view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_39 = x_300 * view_6 - view_6 = None - layer_norm_39 = torch.nn.functional.layer_norm( - x_300, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) - linear_12 = torch._C._nn.linear( - layer_norm_39, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_39 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_7 = linear_12.reshape(1, pow_52, 3, 16, 32) - linear_12 = None - qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) - reshape_7 = None - unbind_3 = qkv_3.unbind(0) - qkv_3 = None - q_3 = unbind_3[0] - k_3 = unbind_3[1] - v_3 = unbind_3[2] - unbind_3 = None - x_301 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) - q_3 = k_3 = v_3 = None - transpose_4 = x_301.transpose(1, 2) - x_301 = None - x_302 = transpose_4.reshape(1, pow_52, 512) - transpose_4 = None - x_303 = torch._C._nn.linear( - x_302, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_302 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_304 = torch.nn.functional.dropout(x_303, 0.0, False, False) - x_303 = None - x_305 = mul_39 + x_304 - mul_39 = x_304 = None - view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_40 = x_305 * view_7 - view_7 = None - layer_norm_40 = torch.nn.functional.layer_norm( - x_305, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_305 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) - x_306 = torch._C._nn.linear( - layer_norm_40, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_40 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_33 = torch.nn.functional.relu(x_306, inplace=False) - x_306 = None - pow_34 = relu_33**2 - relu_33 = None - mul_41 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - * pow_34 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( - pow_34 - ) = None - x_307 = ( - mul_41 - + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - ) - mul_41 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) - x_308 = torch.nn.functional.dropout(x_307, 0.0, False, False) - x_307 = None - x_309 = torch._C._nn.linear( - x_308, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) - x_309 = None - x_311 = mul_40 + x_310 - mul_40 = x_310 = None - view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_42 = x_311 * view_8 - view_8 = None - layer_norm_41 = torch.nn.functional.layer_norm( - x_311, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) - linear_16 = torch._C._nn.linear( - layer_norm_41, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_41 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_9 = linear_16.reshape(1, pow_52, 3, 16, 32) - linear_16 = None - qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) - reshape_9 = None - unbind_4 = qkv_4.unbind(0) - qkv_4 = None - q_4 = unbind_4[0] - k_4 = unbind_4[1] - v_4 = unbind_4[2] - unbind_4 = None - x_312 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) - q_4 = k_4 = v_4 = None - transpose_5 = x_312.transpose(1, 2) - x_312 = None - x_313 = transpose_5.reshape(1, pow_52, 512) - transpose_5 = None - x_314 = torch._C._nn.linear( - x_313, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_313 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) - x_314 = None - x_316 = mul_42 + x_315 - mul_42 = x_315 = None - view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_43 = x_316 * view_9 - view_9 = None - layer_norm_42 = torch.nn.functional.layer_norm( - x_316, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_316 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) - x_317 = torch._C._nn.linear( - layer_norm_42, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_42 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_34 = torch.nn.functional.relu(x_317, inplace=False) - x_317 = None - pow_35 = relu_34**2 - relu_34 = None - mul_44 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - * pow_35 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( - pow_35 - ) = None - x_318 = ( - mul_44 - + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - ) - mul_44 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) - x_319 = torch.nn.functional.dropout(x_318, 0.0, False, False) - x_318 = None - x_320 = torch._C._nn.linear( - x_319, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_321 = torch.nn.functional.dropout(x_320, 0.0, False, False) - x_320 = None - x_322 = mul_43 + x_321 - mul_43 = x_321 = None - view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_45 = x_322 * view_10 - view_10 = None - layer_norm_43 = torch.nn.functional.layer_norm( - x_322, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) - linear_20 = torch._C._nn.linear( - layer_norm_43, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_43 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_11 = linear_20.reshape(1, pow_52, 3, 16, 32) - linear_20 = None - qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) - reshape_11 = None - unbind_5 = qkv_5.unbind(0) - qkv_5 = None - q_5 = unbind_5[0] - k_5 = unbind_5[1] - v_5 = unbind_5[2] - unbind_5 = None - x_323 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) - q_5 = k_5 = v_5 = None - transpose_6 = x_323.transpose(1, 2) - x_323 = None - x_324 = transpose_6.reshape(1, pow_52, 512) - transpose_6 = None - x_325 = torch._C._nn.linear( - x_324, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_324 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_326 = torch.nn.functional.dropout(x_325, 0.0, False, False) - x_325 = None - x_327 = mul_45 + x_326 - mul_45 = x_326 = None - view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_46 = x_327 * view_11 - view_11 = None - layer_norm_44 = torch.nn.functional.layer_norm( - x_327, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_327 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) - x_328 = torch._C._nn.linear( - layer_norm_44, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_44 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_35 = torch.nn.functional.relu(x_328, inplace=False) - x_328 = None - pow_36 = relu_35**2 - relu_35 = None - mul_47 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - * pow_36 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( - pow_36 - ) = None - x_329 = ( - mul_47 - + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - ) - mul_47 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) - x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) - x_329 = None - x_331 = torch._C._nn.linear( - x_330, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_332 = torch.nn.functional.dropout(x_331, 0.0, False, False) - x_331 = None - x_333 = mul_46 + x_332 - mul_46 = x_332 = None - view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_48 = x_333 * view_12 - view_12 = None - layer_norm_45 = torch.nn.functional.layer_norm( - x_333, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) - linear_24 = torch._C._nn.linear( - layer_norm_45, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_45 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_13 = linear_24.reshape(1, pow_52, 3, 16, 32) - linear_24 = None - qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) - reshape_13 = None - unbind_6 = qkv_6.unbind(0) - qkv_6 = None - q_6 = unbind_6[0] - k_6 = unbind_6[1] - v_6 = unbind_6[2] - unbind_6 = None - x_334 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) - q_6 = k_6 = v_6 = None - transpose_7 = x_334.transpose(1, 2) - x_334 = None - x_335 = transpose_7.reshape(1, pow_52, 512) - transpose_7 = None - x_336 = torch._C._nn.linear( - x_335, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_335 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_337 = torch.nn.functional.dropout(x_336, 0.0, False, False) - x_336 = None - x_338 = mul_48 + x_337 - mul_48 = x_337 = None - view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_49 = x_338 * view_13 - view_13 = None - layer_norm_46 = torch.nn.functional.layer_norm( - x_338, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_338 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) - x_339 = torch._C._nn.linear( - layer_norm_46, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_46 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_36 = torch.nn.functional.relu(x_339, inplace=False) - x_339 = None - pow_37 = relu_36**2 - relu_36 = None - mul_50 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - * pow_37 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( - pow_37 - ) = None - x_340 = ( - mul_50 - + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - ) - mul_50 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) - x_341 = torch.nn.functional.dropout(x_340, 0.0, False, False) - x_340 = None - x_342 = torch._C._nn.linear( - x_341, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_343 = torch.nn.functional.dropout(x_342, 0.0, False, False) - x_342 = None - x_344 = mul_49 + x_343 - mul_49 = x_343 = None - view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_51 = x_344 * view_14 - view_14 = None - layer_norm_47 = torch.nn.functional.layer_norm( - x_344, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) - linear_28 = torch._C._nn.linear( - layer_norm_47, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_47 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_15 = linear_28.reshape(1, pow_52, 3, 16, 32) - linear_28 = None - qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) - reshape_15 = None - unbind_7 = qkv_7.unbind(0) - qkv_7 = None - q_7 = unbind_7[0] - k_7 = unbind_7[1] - v_7 = unbind_7[2] - unbind_7 = None - x_345 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) - q_7 = k_7 = v_7 = None - transpose_8 = x_345.transpose(1, 2) - x_345 = None - x_346 = transpose_8.reshape(1, pow_52, 512) - transpose_8 = None - x_347 = torch._C._nn.linear( - x_346, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_346 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) - x_347 = None - x_349 = mul_51 + x_348 - mul_51 = x_348 = None - view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_52 = x_349 * view_15 - view_15 = None - layer_norm_48 = torch.nn.functional.layer_norm( - x_349, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_349 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) - x_350 = torch._C._nn.linear( - layer_norm_48, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_48 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_37 = torch.nn.functional.relu(x_350, inplace=False) - x_350 = None - pow_38 = relu_37**2 - relu_37 = None - mul_53 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - * pow_38 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( - pow_38 - ) = None - x_351 = ( - mul_53 - + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - ) - mul_53 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) - x_352 = torch.nn.functional.dropout(x_351, 0.0, False, False) - x_351 = None - x_353 = torch._C._nn.linear( - x_352, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_354 = torch.nn.functional.dropout(x_353, 0.0, False, False) - x_353 = None - x_355 = mul_52 + x_354 - mul_52 = x_354 = None - view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_54 = x_355 * view_16 - view_16 = None - layer_norm_49 = torch.nn.functional.layer_norm( - x_355, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) - linear_32 = torch._C._nn.linear( - layer_norm_49, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_49 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_17 = linear_32.reshape(1, pow_52, 3, 16, 32) - linear_32 = None - qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) - reshape_17 = None - unbind_8 = qkv_8.unbind(0) - qkv_8 = None - q_8 = unbind_8[0] - k_8 = unbind_8[1] - v_8 = unbind_8[2] - unbind_8 = None - x_356 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) - q_8 = k_8 = v_8 = None - transpose_9 = x_356.transpose(1, 2) - x_356 = None - x_357 = transpose_9.reshape(1, pow_52, 512) - transpose_9 = None - x_358 = torch._C._nn.linear( - x_357, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_357 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_359 = torch.nn.functional.dropout(x_358, 0.0, False, False) - x_358 = None - x_360 = mul_54 + x_359 - mul_54 = x_359 = None - view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_55 = x_360 * view_17 - view_17 = None - layer_norm_50 = torch.nn.functional.layer_norm( - x_360, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_360 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) - x_361 = torch._C._nn.linear( - layer_norm_50, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_50 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_38 = torch.nn.functional.relu(x_361, inplace=False) - x_361 = None - pow_39 = relu_38**2 - relu_38 = None - mul_56 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - * pow_39 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( - pow_39 - ) = None - x_362 = ( - mul_56 - + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - ) - mul_56 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) - x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) - x_362 = None - x_364 = torch._C._nn.linear( - x_363, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_365 = torch.nn.functional.dropout(x_364, 0.0, False, False) - x_364 = None - x_366 = mul_55 + x_365 - mul_55 = x_365 = None - view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_57 = x_366 * view_18 - view_18 = None - layer_norm_51 = torch.nn.functional.layer_norm( - x_366, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) - linear_36 = torch._C._nn.linear( - layer_norm_51, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_51 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_19 = linear_36.reshape(1, pow_52, 3, 16, 32) - linear_36 = None - qkv_9 = reshape_19.permute(2, 0, 3, 1, 4) - reshape_19 = None - unbind_9 = qkv_9.unbind(0) - qkv_9 = None - q_9 = unbind_9[0] - k_9 = unbind_9[1] - v_9 = unbind_9[2] - unbind_9 = None - x_367 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) - q_9 = k_9 = v_9 = None - transpose_10 = x_367.transpose(1, 2) - x_367 = None - x_368 = transpose_10.reshape(1, pow_52, 512) - transpose_10 = None - x_369 = torch._C._nn.linear( - x_368, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_368 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_370 = torch.nn.functional.dropout(x_369, 0.0, False, False) - x_369 = None - x_371 = mul_57 + x_370 - mul_57 = x_370 = None - view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_58 = x_371 * view_19 - view_19 = None - layer_norm_52 = torch.nn.functional.layer_norm( - x_371, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_371 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) - x_372 = torch._C._nn.linear( - layer_norm_52, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_52 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_39 = torch.nn.functional.relu(x_372, inplace=False) - x_372 = None - pow_40 = relu_39**2 - relu_39 = None - mul_59 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - * pow_40 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( - pow_40 - ) = None - x_373 = ( - mul_59 - + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - ) - mul_59 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) - x_374 = torch.nn.functional.dropout(x_373, 0.0, False, False) - x_373 = None - x_375 = torch._C._nn.linear( - x_374, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_374 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_376 = torch.nn.functional.dropout(x_375, 0.0, False, False) - x_375 = None - x_377 = mul_58 + x_376 - mul_58 = x_376 = None - view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_60 = x_377 * view_20 - view_20 = None - layer_norm_53 = torch.nn.functional.layer_norm( - x_377, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_377 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) - linear_40 = torch._C._nn.linear( - layer_norm_53, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_53 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_21 = linear_40.reshape(1, pow_52, 3, 16, 32) - linear_40 = None - qkv_10 = reshape_21.permute(2, 0, 3, 1, 4) - reshape_21 = None - unbind_10 = qkv_10.unbind(0) - qkv_10 = None - q_10 = unbind_10[0] - k_10 = unbind_10[1] - v_10 = unbind_10[2] - unbind_10 = None - x_378 = torch._C._nn.scaled_dot_product_attention( - q_10, k_10, v_10, dropout_p=0.0 - ) - q_10 = k_10 = v_10 = None - transpose_11 = x_378.transpose(1, 2) - x_378 = None - x_379 = transpose_11.reshape(1, pow_52, 512) - transpose_11 = None - x_380 = torch._C._nn.linear( - x_379, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_379 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_381 = torch.nn.functional.dropout(x_380, 0.0, False, False) - x_380 = None - x_382 = mul_60 + x_381 - mul_60 = x_381 = None - view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_61 = x_382 * view_21 - view_21 = None - layer_norm_54 = torch.nn.functional.layer_norm( - x_382, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_382 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) - x_383 = torch._C._nn.linear( - layer_norm_54, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_54 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_40 = torch.nn.functional.relu(x_383, inplace=False) - x_383 = None - pow_41 = relu_40**2 - relu_40 = None - mul_62 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - * pow_41 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( - pow_41 - ) = None - x_384 = ( - mul_62 - + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - ) - mul_62 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) - x_385 = torch.nn.functional.dropout(x_384, 0.0, False, False) - x_384 = None - x_386 = torch._C._nn.linear( - x_385, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_385 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_387 = torch.nn.functional.dropout(x_386, 0.0, False, False) - x_386 = None - x_388 = mul_61 + x_387 - mul_61 = x_387 = None - view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_63 = x_388 * view_22 - view_22 = None - layer_norm_55 = torch.nn.functional.layer_norm( - x_388, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_388 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) - linear_44 = torch._C._nn.linear( - layer_norm_55, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_55 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_23 = linear_44.reshape(1, pow_52, 3, 16, 32) - linear_44 = None - qkv_11 = reshape_23.permute(2, 0, 3, 1, 4) - reshape_23 = None - unbind_11 = qkv_11.unbind(0) - qkv_11 = None - q_11 = unbind_11[0] - k_11 = unbind_11[1] - v_11 = unbind_11[2] - unbind_11 = None - x_389 = torch._C._nn.scaled_dot_product_attention( - q_11, k_11, v_11, dropout_p=0.0 - ) - q_11 = k_11 = v_11 = None - transpose_12 = x_389.transpose(1, 2) - x_389 = None - x_390 = transpose_12.reshape(1, pow_52, 512) - transpose_12 = None - x_391 = torch._C._nn.linear( - x_390, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_390 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_392 = torch.nn.functional.dropout(x_391, 0.0, False, False) - x_391 = None - x_393 = mul_63 + x_392 - mul_63 = x_392 = None - view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_64 = x_393 * view_23 - view_23 = None - layer_norm_56 = torch.nn.functional.layer_norm( - x_393, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_393 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) - x_394 = torch._C._nn.linear( - layer_norm_56, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_56 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_41 = torch.nn.functional.relu(x_394, inplace=False) - x_394 = None - pow_42 = relu_41**2 - relu_41 = None - mul_65 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - * pow_42 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( - pow_42 - ) = None - x_395 = ( - mul_65 - + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - ) - mul_65 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) - x_396 = torch.nn.functional.dropout(x_395, 0.0, False, False) - x_395 = None - x_397 = torch._C._nn.linear( - x_396, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_396 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_398 = torch.nn.functional.dropout(x_397, 0.0, False, False) - x_397 = None - x_399 = mul_64 + x_398 - mul_64 = x_398 = None - view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_66 = x_399 * view_24 - view_24 = None - layer_norm_57 = torch.nn.functional.layer_norm( - x_399, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_399 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = (None) - linear_48 = torch._C._nn.linear( - layer_norm_57, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_57 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_25 = linear_48.reshape(1, pow_52, 3, 16, 32) - linear_48 = None - qkv_12 = reshape_25.permute(2, 0, 3, 1, 4) - reshape_25 = None - unbind_12 = qkv_12.unbind(0) - qkv_12 = None - q_12 = unbind_12[0] - k_12 = unbind_12[1] - v_12 = unbind_12[2] - unbind_12 = None - x_400 = torch._C._nn.scaled_dot_product_attention( - q_12, k_12, v_12, dropout_p=0.0 - ) - q_12 = k_12 = v_12 = None - transpose_13 = x_400.transpose(1, 2) - x_400 = None - x_401 = transpose_13.reshape(1, pow_52, 512) - transpose_13 = None - x_402 = torch._C._nn.linear( - x_401, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_401 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_403 = torch.nn.functional.dropout(x_402, 0.0, False, False) - x_402 = None - x_404 = mul_66 + x_403 - mul_66 = x_403 = None - view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_67 = x_404 * view_25 - view_25 = None - layer_norm_58 = torch.nn.functional.layer_norm( - x_404, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_404 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = (None) - x_405 = torch._C._nn.linear( - layer_norm_58, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_58 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_42 = torch.nn.functional.relu(x_405, inplace=False) - x_405 = None - pow_43 = relu_42**2 - relu_42 = None - mul_68 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ - * pow_43 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = ( - pow_43 - ) = None - x_406 = ( - mul_68 - + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ - ) - mul_68 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = (None) - x_407 = torch.nn.functional.dropout(x_406, 0.0, False, False) - x_406 = None - x_408 = torch._C._nn.linear( - x_407, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_407 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_409 = torch.nn.functional.dropout(x_408, 0.0, False, False) - x_408 = None - x_410 = mul_67 + x_409 - mul_67 = x_409 = None - view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_69 = x_410 * view_26 - view_26 = None - layer_norm_59 = torch.nn.functional.layer_norm( - x_410, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_410 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = (None) - linear_52 = torch._C._nn.linear( - layer_norm_59, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_59 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_27 = linear_52.reshape(1, pow_52, 3, 16, 32) - linear_52 = None - qkv_13 = reshape_27.permute(2, 0, 3, 1, 4) - reshape_27 = None - unbind_13 = qkv_13.unbind(0) - qkv_13 = None - q_13 = unbind_13[0] - k_13 = unbind_13[1] - v_13 = unbind_13[2] - unbind_13 = None - x_411 = torch._C._nn.scaled_dot_product_attention( - q_13, k_13, v_13, dropout_p=0.0 - ) - q_13 = k_13 = v_13 = None - transpose_14 = x_411.transpose(1, 2) - x_411 = None - x_412 = transpose_14.reshape(1, pow_52, 512) - transpose_14 = None - x_413 = torch._C._nn.linear( - x_412, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_412 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_414 = torch.nn.functional.dropout(x_413, 0.0, False, False) - x_413 = None - x_415 = mul_69 + x_414 - mul_69 = x_414 = None - view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_70 = x_415 * view_27 - view_27 = None - layer_norm_60 = torch.nn.functional.layer_norm( - x_415, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_415 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = (None) - x_416 = torch._C._nn.linear( - layer_norm_60, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_60 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_43 = torch.nn.functional.relu(x_416, inplace=False) - x_416 = None - pow_44 = relu_43**2 - relu_43 = None - mul_71 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ - * pow_44 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = ( - pow_44 - ) = None - x_417 = ( - mul_71 - + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ - ) - mul_71 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = (None) - x_418 = torch.nn.functional.dropout(x_417, 0.0, False, False) - x_417 = None - x_419 = torch._C._nn.linear( - x_418, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_418 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_420 = torch.nn.functional.dropout(x_419, 0.0, False, False) - x_419 = None - x_421 = mul_70 + x_420 - mul_70 = x_420 = None - view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_72 = x_421 * view_28 - view_28 = None - layer_norm_61 = torch.nn.functional.layer_norm( - x_421, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_421 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = (None) - linear_56 = torch._C._nn.linear( - layer_norm_61, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_61 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_29 = linear_56.reshape(1, pow_52, 3, 16, 32) - linear_56 = None - qkv_14 = reshape_29.permute(2, 0, 3, 1, 4) - reshape_29 = None - unbind_14 = qkv_14.unbind(0) - qkv_14 = None - q_14 = unbind_14[0] - k_14 = unbind_14[1] - v_14 = unbind_14[2] - unbind_14 = None - x_422 = torch._C._nn.scaled_dot_product_attention( - q_14, k_14, v_14, dropout_p=0.0 - ) - q_14 = k_14 = v_14 = None - transpose_15 = x_422.transpose(1, 2) - x_422 = None - x_423 = transpose_15.reshape(1, pow_52, 512) - transpose_15 = None - x_424 = torch._C._nn.linear( - x_423, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_423 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_425 = torch.nn.functional.dropout(x_424, 0.0, False, False) - x_424 = None - x_426 = mul_72 + x_425 - mul_72 = x_425 = None - view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_73 = x_426 * view_29 - view_29 = None - layer_norm_62 = torch.nn.functional.layer_norm( - x_426, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_426 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = (None) - x_427 = torch._C._nn.linear( - layer_norm_62, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_62 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_44 = torch.nn.functional.relu(x_427, inplace=False) - x_427 = None - pow_45 = relu_44**2 - relu_44 = None - mul_74 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ - * pow_45 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = ( - pow_45 - ) = None - x_428 = ( - mul_74 - + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ - ) - mul_74 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = (None) - x_429 = torch.nn.functional.dropout(x_428, 0.0, False, False) - x_428 = None - x_430 = torch._C._nn.linear( - x_429, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_429 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_431 = torch.nn.functional.dropout(x_430, 0.0, False, False) - x_430 = None - x_432 = mul_73 + x_431 - mul_73 = x_431 = None - view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_75 = x_432 * view_30 - view_30 = None - layer_norm_63 = torch.nn.functional.layer_norm( - x_432, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_432 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = (None) - linear_60 = torch._C._nn.linear( - layer_norm_63, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_63 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_31 = linear_60.reshape(1, pow_52, 3, 16, 32) - linear_60 = None - qkv_15 = reshape_31.permute(2, 0, 3, 1, 4) - reshape_31 = None - unbind_15 = qkv_15.unbind(0) - qkv_15 = None - q_15 = unbind_15[0] - k_15 = unbind_15[1] - v_15 = unbind_15[2] - unbind_15 = None - x_433 = torch._C._nn.scaled_dot_product_attention( - q_15, k_15, v_15, dropout_p=0.0 - ) - q_15 = k_15 = v_15 = None - transpose_16 = x_433.transpose(1, 2) - x_433 = None - x_434 = transpose_16.reshape(1, pow_52, 512) - transpose_16 = None - x_435 = torch._C._nn.linear( - x_434, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_434 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_436 = torch.nn.functional.dropout(x_435, 0.0, False, False) - x_435 = None - x_437 = mul_75 + x_436 - mul_75 = x_436 = None - view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_76 = x_437 * view_31 - view_31 = None - layer_norm_64 = torch.nn.functional.layer_norm( - x_437, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_437 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = (None) - x_438 = torch._C._nn.linear( - layer_norm_64, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_64 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_45 = torch.nn.functional.relu(x_438, inplace=False) - x_438 = None - pow_46 = relu_45**2 - relu_45 = None - mul_77 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ - * pow_46 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = ( - pow_46 - ) = None - x_439 = ( - mul_77 - + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ - ) - mul_77 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = (None) - x_440 = torch.nn.functional.dropout(x_439, 0.0, False, False) - x_439 = None - x_441 = torch._C._nn.linear( - x_440, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_440 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_442 = torch.nn.functional.dropout(x_441, 0.0, False, False) - x_441 = None - x_443 = mul_76 + x_442 - mul_76 = x_442 = None - view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_78 = x_443 * view_32 - view_32 = None - layer_norm_65 = torch.nn.functional.layer_norm( - x_443, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_443 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = (None) - linear_64 = torch._C._nn.linear( - layer_norm_65, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_65 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_33 = linear_64.reshape(1, pow_52, 3, 16, 32) - linear_64 = None - qkv_16 = reshape_33.permute(2, 0, 3, 1, 4) - reshape_33 = None - unbind_16 = qkv_16.unbind(0) - qkv_16 = None - q_16 = unbind_16[0] - k_16 = unbind_16[1] - v_16 = unbind_16[2] - unbind_16 = None - x_444 = torch._C._nn.scaled_dot_product_attention( - q_16, k_16, v_16, dropout_p=0.0 - ) - q_16 = k_16 = v_16 = None - transpose_17 = x_444.transpose(1, 2) - x_444 = None - x_445 = transpose_17.reshape(1, pow_52, 512) - transpose_17 = None - x_446 = torch._C._nn.linear( - x_445, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_445 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_447 = torch.nn.functional.dropout(x_446, 0.0, False, False) - x_446 = None - x_448 = mul_78 + x_447 - mul_78 = x_447 = None - view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_79 = x_448 * view_33 - view_33 = None - layer_norm_66 = torch.nn.functional.layer_norm( - x_448, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_448 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = (None) - x_449 = torch._C._nn.linear( - layer_norm_66, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_66 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_46 = torch.nn.functional.relu(x_449, inplace=False) - x_449 = None - pow_47 = relu_46**2 - relu_46 = None - mul_80 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ - * pow_47 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = ( - pow_47 - ) = None - x_450 = ( - mul_80 - + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ - ) - mul_80 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = (None) - x_451 = torch.nn.functional.dropout(x_450, 0.0, False, False) - x_450 = None - x_452 = torch._C._nn.linear( - x_451, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_451 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_453 = torch.nn.functional.dropout(x_452, 0.0, False, False) - x_452 = None - x_454 = mul_79 + x_453 - mul_79 = x_453 = None - view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_81 = x_454 * view_34 - view_34 = None - layer_norm_67 = torch.nn.functional.layer_norm( - x_454, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_454 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = (None) - linear_68 = torch._C._nn.linear( - layer_norm_67, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_67 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_35 = linear_68.reshape(1, pow_52, 3, 16, 32) - linear_68 = None - qkv_17 = reshape_35.permute(2, 0, 3, 1, 4) - reshape_35 = None - unbind_17 = qkv_17.unbind(0) - qkv_17 = None - q_17 = unbind_17[0] - k_17 = unbind_17[1] - v_17 = unbind_17[2] - unbind_17 = None - x_455 = torch._C._nn.scaled_dot_product_attention( - q_17, k_17, v_17, dropout_p=0.0 - ) - q_17 = k_17 = v_17 = None - transpose_18 = x_455.transpose(1, 2) - x_455 = None - x_456 = transpose_18.reshape(1, pow_52, 512) - transpose_18 = pow_52 = None - x_457 = torch._C._nn.linear( - x_456, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_456 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_458 = torch.nn.functional.dropout(x_457, 0.0, False, False) - x_457 = None - x_459 = mul_81 + x_458 - mul_81 = x_458 = None - view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_82 = x_459 * view_35 - view_35 = None - layer_norm_68 = torch.nn.functional.layer_norm( - x_459, - (512,), - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_459 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = (None) - x_460 = torch._C._nn.linear( - layer_norm_68, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_68 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_47 = torch.nn.functional.relu(x_460, inplace=False) - x_460 = None - pow_48 = relu_47**2 - relu_47 = None - mul_83 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ - * pow_48 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = ( - pow_48 - ) = None - x_461 = ( - mul_83 - + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ - ) - mul_83 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = (None) - x_462 = torch.nn.functional.dropout(x_461, 0.0, False, False) - x_461 = None - x_463 = torch._C._nn.linear( - x_462, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_462 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_464 = torch.nn.functional.dropout(x_463, 0.0, False, False) - x_463 = None - x_465 = mul_82 + x_464 - mul_82 = x_464 = None - transpose_19 = x_465.transpose(1, 2) - x_465 = None - x_466 = transpose_19.reshape(1, 512, sym_sum_3, sym_sum_3) - transpose_19 = sym_sum_3 = None - x_467 = x_466.permute(0, 2, 3, 1) - x_466 = None - x_468 = torch.nn.functional.layer_norm( - x_467, - (512,), - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_467 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) - x_469 = x_468.permute(0, 3, 1, 2) - x_468 = None - x_470 = torch.conv2d( - x_469, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_469 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_3 = sym_sum // 32 - sym_sum = None - sym_sum_4 = torch.sym_sum([1, floordiv_3]) - floordiv_3 = None - reshape_38 = x_470.reshape(1, 768, -1) - x_470 = None - x_471 = reshape_38.transpose(1, 2) - reshape_38 = None - view_36 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_84 = x_471 * view_36 - view_36 = None - layer_norm_70 = torch.nn.functional.layer_norm( - x_471, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_471 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_53 = sym_sum_4**2 - linear_72 = torch._C._nn.linear( - layer_norm_70, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_70 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_39 = linear_72.reshape(1, pow_53, 3, 24, 32) - linear_72 = None - qkv_18 = reshape_39.permute(2, 0, 3, 1, 4) - reshape_39 = None - unbind_18 = qkv_18.unbind(0) - qkv_18 = None - q_18 = unbind_18[0] - k_18 = unbind_18[1] - v_18 = unbind_18[2] - unbind_18 = None - x_472 = torch._C._nn.scaled_dot_product_attention( - q_18, k_18, v_18, dropout_p=0.0 - ) - q_18 = k_18 = v_18 = None - transpose_21 = x_472.transpose(1, 2) - x_472 = None - x_473 = transpose_21.reshape(1, pow_53, 768) - transpose_21 = None - x_474 = torch._C._nn.linear( - x_473, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_473 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_475 = torch.nn.functional.dropout(x_474, 0.0, False, False) - x_474 = None - x_476 = mul_84 + x_475 - mul_84 = x_475 = None - view_37 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_85 = x_476 * view_37 - view_37 = None - layer_norm_71 = torch.nn.functional.layer_norm( - x_476, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_476 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_477 = torch._C._nn.linear( - layer_norm_71, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_71 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_48 = torch.nn.functional.relu(x_477, inplace=False) - x_477 = None - pow_49 = relu_48**2 - relu_48 = None - mul_86 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_49 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_49 - ) = None - x_478 = ( - mul_86 - + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_86 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_479 = torch.nn.functional.dropout(x_478, 0.0, False, False) - x_478 = None - x_480 = torch._C._nn.linear( - x_479, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_479 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_481 = torch.nn.functional.dropout(x_480, 0.0, False, False) - x_480 = None - x_482 = mul_85 + x_481 - mul_85 = x_481 = None - view_38 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_87 = x_482 * view_38 - view_38 = None - layer_norm_72 = torch.nn.functional.layer_norm( - x_482, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_482 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_76 = torch._C._nn.linear( - layer_norm_72, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_72 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_41 = linear_76.reshape(1, pow_53, 3, 24, 32) - linear_76 = None - qkv_19 = reshape_41.permute(2, 0, 3, 1, 4) - reshape_41 = None - unbind_19 = qkv_19.unbind(0) - qkv_19 = None - q_19 = unbind_19[0] - k_19 = unbind_19[1] - v_19 = unbind_19[2] - unbind_19 = None - x_483 = torch._C._nn.scaled_dot_product_attention( - q_19, k_19, v_19, dropout_p=0.0 - ) - q_19 = k_19 = v_19 = None - transpose_22 = x_483.transpose(1, 2) - x_483 = None - x_484 = transpose_22.reshape(1, pow_53, 768) - transpose_22 = None - x_485 = torch._C._nn.linear( - x_484, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_484 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_486 = torch.nn.functional.dropout(x_485, 0.0, False, False) - x_485 = None - x_487 = mul_87 + x_486 - mul_87 = x_486 = None - view_39 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_88 = x_487 * view_39 - view_39 = None - layer_norm_73 = torch.nn.functional.layer_norm( - x_487, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_487 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_488 = torch._C._nn.linear( - layer_norm_73, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_73 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_49 = torch.nn.functional.relu(x_488, inplace=False) - x_488 = None - pow_50 = relu_49**2 - relu_49 = None - mul_89 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_50 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_50 - ) = None - x_489 = ( - mul_89 - + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_89 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_490 = torch.nn.functional.dropout(x_489, 0.0, False, False) - x_489 = None - x_491 = torch._C._nn.linear( - x_490, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_490 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_492 = torch.nn.functional.dropout(x_491, 0.0, False, False) - x_491 = None - x_493 = mul_88 + x_492 - mul_88 = x_492 = None - view_40 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_90 = x_493 * view_40 - view_40 = None - layer_norm_74 = torch.nn.functional.layer_norm( - x_493, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_493 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_80 = torch._C._nn.linear( - layer_norm_74, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_74 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_43 = linear_80.reshape(1, pow_53, 3, 24, 32) - linear_80 = None - qkv_20 = reshape_43.permute(2, 0, 3, 1, 4) - reshape_43 = None - unbind_20 = qkv_20.unbind(0) - qkv_20 = None - q_20 = unbind_20[0] - k_20 = unbind_20[1] - v_20 = unbind_20[2] - unbind_20 = None - x_494 = torch._C._nn.scaled_dot_product_attention( - q_20, k_20, v_20, dropout_p=0.0 - ) - q_20 = k_20 = v_20 = None - transpose_23 = x_494.transpose(1, 2) - x_494 = None - x_495 = transpose_23.reshape(1, pow_53, 768) - transpose_23 = pow_53 = None - x_496 = torch._C._nn.linear( - x_495, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_495 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_497 = torch.nn.functional.dropout(x_496, 0.0, False, False) - x_496 = None - x_498 = mul_90 + x_497 - mul_90 = x_497 = None - view_41 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (768,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_91 = x_498 * view_41 - view_41 = None - layer_norm_75 = torch.nn.functional.layer_norm( - x_498, - (768,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_498 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_499 = torch._C._nn.linear( - layer_norm_75, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_75 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_50 = torch.nn.functional.relu(x_499, inplace=False) - x_499 = None - pow_51 = relu_50**2 - relu_50 = None - mul_92 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_51 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_51 - ) = None - x_500 = ( - mul_92 - + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_92 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_501 = torch.nn.functional.dropout(x_500, 0.0, False, False) - x_500 = None - x_502 = torch._C._nn.linear( - x_501, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_501 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_503 = torch.nn.functional.dropout(x_502, 0.0, False, False) - x_502 = None - x_504 = mul_91 + x_503 - mul_91 = x_503 = None - transpose_24 = x_504.transpose(1, 2) - x_504 = None - x_505 = transpose_24.reshape(1, 768, sym_sum_4, sym_sum_4) - transpose_24 = sym_sum_4 = None - x_506 = torch.nn.functional.adaptive_avg_pool2d(x_505, 1) - x_505 = None - x_507 = x_506.permute(0, 2, 3, 1) - x_506 = None - x_508 = torch.nn.functional.layer_norm( - x_507, - (768,), - l_self_modules_head_modules_norm_parameters_weight_, - l_self_modules_head_modules_norm_parameters_bias_, - 1e-06, - ) - x_507 = ( - l_self_modules_head_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_norm_parameters_bias_ = None - x_509 = x_508.permute(0, 3, 1, 2) - x_508 = None - x_510 = x_509.flatten(1, -1) - x_509 = None - x_511 = torch.nn.functional.dropout(x_510, 0.0, False, False) - x_510 = None - x_512 = torch._C._nn.linear( - x_511, - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, - ) - x_511 = ( - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None - relu_51 = torch.nn.functional.relu(x_512, inplace=False) - x_512 = None - x_513 = torch.square(relu_51) - relu_51 = None - x_514 = torch.nn.functional.layer_norm( - x_513, - (3072,), - l_self_modules_head_modules_fc_modules_norm_parameters_weight_, - l_self_modules_head_modules_fc_modules_norm_parameters_bias_, - 1e-06, - ) - x_513 = ( - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None - x_515 = torch.nn.functional.dropout(x_514, 0.0, False, False) - x_514 = None - x_516 = torch._C._nn.linear( - x_515, - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, - ) - x_515 = ( - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None - return (x_516,) diff --git a/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/weight_meta.py b/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/weight_meta.py deleted file mode 100644 index 6cf290fca4..0000000000 --- a/samples/timm/caformer_b36.sail_in22k_ft_in1k_384/weight_meta.py +++ /dev/null @@ -1,3968 +0,0 @@ -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: - name = "L_self_modules_stem_modules_conv_parameters_weight_" - shape = [128, 3, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: - name = "L_self_modules_stem_modules_conv_parameters_bias_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_s1: - name = "s1" - shape = [] - dtype = "torch.int64" - device = "cpu" - mean = None - std = None - data = [4] - - -class Program_weight_tensor_meta_L_x_: - name = "L_x_" - shape = [1, 3, 224, 224] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.226 - std = 1.284 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: - name = "L_self_modules_stem_modules_norm_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [256, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [256, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [128, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [512, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [128, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [256, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [256, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [128, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [512, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [128, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [256, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [256, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [128, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [512, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [128, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" - shape = [256, 128, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [512, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [512, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [256, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" - shape = [1024, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" - shape = [256, 1024, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" - shape = [256] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" - shape = [512, 256, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" - shape = [768, 512, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [2304, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [768, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [3072, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [768, 3072] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [2304, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [768, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [3072, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [768, 3072] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [2304, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [768, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [3072, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [768, 3072] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_norm_parameters_weight_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_norm_parameters_bias_" - shape = [768] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" - shape = [3072, 768] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" - shape = [1000, 3072] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" - shape = [1000] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None diff --git a/samples/timm/caformer_m36.sail_in1k/graph_net.json b/samples/timm/caformer_m36.sail_in1k/graph_net.json deleted file mode 100644 index 1373fe3b58..0000000000 --- a/samples/timm/caformer_m36.sail_in1k/graph_net.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "framework": "torch", - "num_devices_required": 1, - "num_nodes_required": 1 -} \ No newline at end of file diff --git a/samples/timm/caformer_m36.sail_in1k/input_meta.py b/samples/timm/caformer_m36.sail_in1k/input_meta.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_m36.sail_in1k/input_tensor_constraints.py b/samples/timm/caformer_m36.sail_in1k/input_tensor_constraints.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_m36.sail_in1k/model.py b/samples/timm/caformer_m36.sail_in1k/model.py deleted file mode 100644 index dcce286c5c..0000000000 --- a/samples/timm/caformer_m36.sail_in1k/model.py +++ /dev/null @@ -1,4754 +0,0 @@ -import torch - - -class GraphModule(torch.nn.Module): - def forward( - self, - L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - s1: torch.SymInt, - L_x_: torch.Tensor, - L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - ): - l_self_modules_stem_modules_conv_parameters_weight_ = ( - L_self_modules_stem_modules_conv_parameters_weight_ - ) - l_self_modules_stem_modules_conv_parameters_bias_ = ( - L_self_modules_stem_modules_conv_parameters_bias_ - ) - l_x_ = L_x_ - l_self_modules_stem_modules_norm_parameters_weight_ = ( - L_self_modules_stem_modules_norm_parameters_weight_ - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_head_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ - ) - x = torch.conv2d( - l_x_, - l_self_modules_stem_modules_conv_parameters_weight_, - l_self_modules_stem_modules_conv_parameters_bias_, - (4, 4), - (2, 2), - (1, 1), - 1, - ) - l_x_ = ( - l_self_modules_stem_modules_conv_parameters_weight_ - ) = l_self_modules_stem_modules_conv_parameters_bias_ = None - x_1 = x.permute(0, 2, 3, 1) - x = None - x_2 = torch.nn.functional.layer_norm( - x_1, (96,), l_self_modules_stem_modules_norm_parameters_weight_, None, 1e-06 - ) - x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None - x_3 = x_2.permute(0, 3, 1, 2) - x_2 = None - sym_sum = torch.sym_sum([-3, s1]) - s1 = None - floordiv = sym_sum // 4 - sym_sum_1 = torch.sym_sum([1, floordiv]) - floordiv = sym_sum_1 = None - x_4 = x_3.permute(0, 2, 3, 1) - x_5 = torch.nn.functional.layer_norm( - x_4, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_6 = x_5.permute(0, 3, 1, 2) - x_5 = None - x_7 = torch.conv2d( - x_6, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu = torch.nn.functional.relu(x_7, inplace=False) - x_7 = None - pow_1 = relu**2 - relu = None - mul = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_1 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_1 - ) = None - x_8 = ( - mul - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_9 = torch.conv2d( - x_8, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 192, - ) - x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_10 = torch.conv2d( - x_9, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_11 = x_3 + x_10 - x_3 = x_10 = None - x_12 = x_11.permute(0, 2, 3, 1) - x_13 = torch.nn.functional.layer_norm( - x_12, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_14 = x_13.permute(0, 3, 1, 2) - x_13 = None - x_15 = torch.conv2d( - x_14, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_1 = torch.nn.functional.relu(x_15, inplace=False) - x_15 = None - pow_2 = relu_1**2 - relu_1 = None - mul_1 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_2 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_2 - ) = None - x_16 = ( - mul_1 - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) - x_16 = None - x_18 = torch.conv2d( - x_17, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) - x_18 = None - x_20 = x_11 + x_19 - x_11 = x_19 = None - x_21 = x_20.permute(0, 2, 3, 1) - x_22 = torch.nn.functional.layer_norm( - x_21, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_23 = x_22.permute(0, 3, 1, 2) - x_22 = None - x_24 = torch.conv2d( - x_23, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_2 = torch.nn.functional.relu(x_24, inplace=False) - x_24 = None - pow_3 = relu_2**2 - relu_2 = None - mul_2 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_3 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_3 - ) = None - x_25 = ( - mul_2 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_26 = torch.conv2d( - x_25, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 192, - ) - x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_27 = torch.conv2d( - x_26, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_28 = x_20 + x_27 - x_20 = x_27 = None - x_29 = x_28.permute(0, 2, 3, 1) - x_30 = torch.nn.functional.layer_norm( - x_29, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_31 = x_30.permute(0, 3, 1, 2) - x_30 = None - x_32 = torch.conv2d( - x_31, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_3 = torch.nn.functional.relu(x_32, inplace=False) - x_32 = None - pow_4 = relu_3**2 - relu_3 = None - mul_3 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_4 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_4 - ) = None - x_33 = ( - mul_3 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) - x_33 = None - x_35 = torch.conv2d( - x_34, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) - x_35 = None - x_37 = x_28 + x_36 - x_28 = x_36 = None - x_38 = x_37.permute(0, 2, 3, 1) - x_39 = torch.nn.functional.layer_norm( - x_38, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_40 = x_39.permute(0, 3, 1, 2) - x_39 = None - x_41 = torch.conv2d( - x_40, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_4 = torch.nn.functional.relu(x_41, inplace=False) - x_41 = None - pow_5 = relu_4**2 - relu_4 = None - mul_4 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_5 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_5 - ) = None - x_42 = ( - mul_4 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_43 = torch.conv2d( - x_42, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 192, - ) - x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_44 = torch.conv2d( - x_43, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_45 = x_37 + x_44 - x_37 = x_44 = None - x_46 = x_45.permute(0, 2, 3, 1) - x_47 = torch.nn.functional.layer_norm( - x_46, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_48 = x_47.permute(0, 3, 1, 2) - x_47 = None - x_49 = torch.conv2d( - x_48, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_5 = torch.nn.functional.relu(x_49, inplace=False) - x_49 = None - pow_6 = relu_5**2 - relu_5 = None - mul_5 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_6 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_6 - ) = None - x_50 = ( - mul_5 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) - x_50 = None - x_52 = torch.conv2d( - x_51, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) - x_52 = None - x_54 = x_45 + x_53 - x_45 = x_53 = None - x_55 = x_54.permute(0, 2, 3, 1) - x_54 = None - x_56 = torch.nn.functional.layer_norm( - x_55, - (96,), - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) - x_57 = x_56.permute(0, 3, 1, 2) - x_56 = None - x_58 = torch.conv2d( - x_57, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_1 = sym_sum // 8 - sym_sum_2 = torch.sym_sum([1, floordiv_1]) - floordiv_1 = sym_sum_2 = None - x_59 = x_58.permute(0, 2, 3, 1) - x_60 = torch.nn.functional.layer_norm( - x_59, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_61 = x_60.permute(0, 3, 1, 2) - x_60 = None - x_62 = torch.conv2d( - x_61, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_6 = torch.nn.functional.relu(x_62, inplace=False) - x_62 = None - pow_7 = relu_6**2 - relu_6 = None - mul_6 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_7 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_7 - ) = None - x_63 = ( - mul_6 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_64 = torch.conv2d( - x_63, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_65 = torch.conv2d( - x_64, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_66 = x_58 + x_65 - x_58 = x_65 = None - x_67 = x_66.permute(0, 2, 3, 1) - x_68 = torch.nn.functional.layer_norm( - x_67, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_69 = x_68.permute(0, 3, 1, 2) - x_68 = None - x_70 = torch.conv2d( - x_69, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_7 = torch.nn.functional.relu(x_70, inplace=False) - x_70 = None - pow_8 = relu_7**2 - relu_7 = None - mul_7 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_8 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_8 - ) = None - x_71 = ( - mul_7 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) - x_71 = None - x_73 = torch.conv2d( - x_72, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) - x_73 = None - x_75 = x_66 + x_74 - x_66 = x_74 = None - x_76 = x_75.permute(0, 2, 3, 1) - x_77 = torch.nn.functional.layer_norm( - x_76, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_78 = x_77.permute(0, 3, 1, 2) - x_77 = None - x_79 = torch.conv2d( - x_78, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_8 = torch.nn.functional.relu(x_79, inplace=False) - x_79 = None - pow_9 = relu_8**2 - relu_8 = None - mul_8 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_9 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_9 - ) = None - x_80 = ( - mul_8 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_81 = torch.conv2d( - x_80, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_82 = torch.conv2d( - x_81, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_83 = x_75 + x_82 - x_75 = x_82 = None - x_84 = x_83.permute(0, 2, 3, 1) - x_85 = torch.nn.functional.layer_norm( - x_84, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_86 = x_85.permute(0, 3, 1, 2) - x_85 = None - x_87 = torch.conv2d( - x_86, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_9 = torch.nn.functional.relu(x_87, inplace=False) - x_87 = None - pow_10 = relu_9**2 - relu_9 = None - mul_9 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_10 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_10 - ) = None - x_88 = ( - mul_9 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) - x_88 = None - x_90 = torch.conv2d( - x_89, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) - x_90 = None - x_92 = x_83 + x_91 - x_83 = x_91 = None - x_93 = x_92.permute(0, 2, 3, 1) - x_94 = torch.nn.functional.layer_norm( - x_93, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_95 = x_94.permute(0, 3, 1, 2) - x_94 = None - x_96 = torch.conv2d( - x_95, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_10 = torch.nn.functional.relu(x_96, inplace=False) - x_96 = None - pow_11 = relu_10**2 - relu_10 = None - mul_10 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_11 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_11 - ) = None - x_97 = ( - mul_10 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_98 = torch.conv2d( - x_97, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_99 = torch.conv2d( - x_98, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_100 = x_92 + x_99 - x_92 = x_99 = None - x_101 = x_100.permute(0, 2, 3, 1) - x_102 = torch.nn.functional.layer_norm( - x_101, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_103 = x_102.permute(0, 3, 1, 2) - x_102 = None - x_104 = torch.conv2d( - x_103, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_11 = torch.nn.functional.relu(x_104, inplace=False) - x_104 = None - pow_12 = relu_11**2 - relu_11 = None - mul_11 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_12 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_12 - ) = None - x_105 = ( - mul_11 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) - x_105 = None - x_107 = torch.conv2d( - x_106, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) - x_107 = None - x_109 = x_100 + x_108 - x_100 = x_108 = None - x_110 = x_109.permute(0, 2, 3, 1) - x_111 = torch.nn.functional.layer_norm( - x_110, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_110 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) - x_112 = x_111.permute(0, 3, 1, 2) - x_111 = None - x_113 = torch.conv2d( - x_112, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_112 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_12 = torch.nn.functional.relu(x_113, inplace=False) - x_113 = None - pow_13 = relu_12**2 - relu_12 = None - mul_12 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ - * pow_13 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_13 - ) = None - x_114 = ( - mul_12 - + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_12 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_115 = torch.conv2d( - x_114, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_114 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_116 = torch.conv2d( - x_115, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_115 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_117 = x_109 + x_116 - x_109 = x_116 = None - x_118 = x_117.permute(0, 2, 3, 1) - x_119 = torch.nn.functional.layer_norm( - x_118, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_118 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) - x_120 = x_119.permute(0, 3, 1, 2) - x_119 = None - x_121 = torch.conv2d( - x_120, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_120 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_13 = torch.nn.functional.relu(x_121, inplace=False) - x_121 = None - pow_14 = relu_13**2 - relu_13 = None - mul_13 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - * pow_14 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( - pow_14 - ) = None - x_122 = ( - mul_13 - + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - ) - mul_13 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) - x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) - x_122 = None - x_124 = torch.conv2d( - x_123, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_123 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) - x_124 = None - x_126 = x_117 + x_125 - x_117 = x_125 = None - x_127 = x_126.permute(0, 2, 3, 1) - x_128 = torch.nn.functional.layer_norm( - x_127, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_127 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) - x_129 = x_128.permute(0, 3, 1, 2) - x_128 = None - x_130 = torch.conv2d( - x_129, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_129 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_14 = torch.nn.functional.relu(x_130, inplace=False) - x_130 = None - pow_15 = relu_14**2 - relu_14 = None - mul_14 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ - * pow_15 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_15 - ) = None - x_131 = ( - mul_14 - + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_14 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_132 = torch.conv2d( - x_131, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_131 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_133 = torch.conv2d( - x_132, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_132 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_134 = x_126 + x_133 - x_126 = x_133 = None - x_135 = x_134.permute(0, 2, 3, 1) - x_136 = torch.nn.functional.layer_norm( - x_135, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_135 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) - x_137 = x_136.permute(0, 3, 1, 2) - x_136 = None - x_138 = torch.conv2d( - x_137, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_137 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_15 = torch.nn.functional.relu(x_138, inplace=False) - x_138 = None - pow_16 = relu_15**2 - relu_15 = None - mul_15 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - * pow_16 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( - pow_16 - ) = None - x_139 = ( - mul_15 - + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - ) - mul_15 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) - x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) - x_139 = None - x_141 = torch.conv2d( - x_140, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_140 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) - x_141 = None - x_143 = x_134 + x_142 - x_134 = x_142 = None - x_144 = x_143.permute(0, 2, 3, 1) - x_145 = torch.nn.functional.layer_norm( - x_144, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_144 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) - x_146 = x_145.permute(0, 3, 1, 2) - x_145 = None - x_147 = torch.conv2d( - x_146, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_146 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_16 = torch.nn.functional.relu(x_147, inplace=False) - x_147 = None - pow_17 = relu_16**2 - relu_16 = None - mul_16 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ - * pow_17 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_17 - ) = None - x_148 = ( - mul_16 - + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_16 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_149 = torch.conv2d( - x_148, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_148 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_150 = torch.conv2d( - x_149, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_149 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_151 = x_143 + x_150 - x_143 = x_150 = None - x_152 = x_151.permute(0, 2, 3, 1) - x_153 = torch.nn.functional.layer_norm( - x_152, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_152 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) - x_154 = x_153.permute(0, 3, 1, 2) - x_153 = None - x_155 = torch.conv2d( - x_154, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_154 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_17 = torch.nn.functional.relu(x_155, inplace=False) - x_155 = None - pow_18 = relu_17**2 - relu_17 = None - mul_17 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - * pow_18 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( - pow_18 - ) = None - x_156 = ( - mul_17 - + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - ) - mul_17 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) - x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) - x_156 = None - x_158 = torch.conv2d( - x_157, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_157 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) - x_158 = None - x_160 = x_151 + x_159 - x_151 = x_159 = None - x_161 = x_160.permute(0, 2, 3, 1) - x_162 = torch.nn.functional.layer_norm( - x_161, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_161 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) - x_163 = x_162.permute(0, 3, 1, 2) - x_162 = None - x_164 = torch.conv2d( - x_163, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_163 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_18 = torch.nn.functional.relu(x_164, inplace=False) - x_164 = None - pow_19 = relu_18**2 - relu_18 = None - mul_18 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ - * pow_19 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_19 - ) = None - x_165 = ( - mul_18 - + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_18 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_166 = torch.conv2d( - x_165, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_165 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_167 = torch.conv2d( - x_166, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_166 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_168 = x_160 + x_167 - x_160 = x_167 = None - x_169 = x_168.permute(0, 2, 3, 1) - x_170 = torch.nn.functional.layer_norm( - x_169, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_169 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) - x_171 = x_170.permute(0, 3, 1, 2) - x_170 = None - x_172 = torch.conv2d( - x_171, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_171 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_19 = torch.nn.functional.relu(x_172, inplace=False) - x_172 = None - pow_20 = relu_19**2 - relu_19 = None - mul_19 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - * pow_20 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( - pow_20 - ) = None - x_173 = ( - mul_19 - + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - ) - mul_19 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) - x_174 = torch.nn.functional.dropout(x_173, 0.0, False, False) - x_173 = None - x_175 = torch.conv2d( - x_174, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_174 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) - x_175 = None - x_177 = x_168 + x_176 - x_168 = x_176 = None - x_178 = x_177.permute(0, 2, 3, 1) - x_179 = torch.nn.functional.layer_norm( - x_178, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_178 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) - x_180 = x_179.permute(0, 3, 1, 2) - x_179 = None - x_181 = torch.conv2d( - x_180, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_180 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_20 = torch.nn.functional.relu(x_181, inplace=False) - x_181 = None - pow_21 = relu_20**2 - relu_20 = None - mul_20 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ - * pow_21 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_21 - ) = None - x_182 = ( - mul_20 - + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_20 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_183 = torch.conv2d( - x_182, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_182 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_184 = torch.conv2d( - x_183, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_183 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_185 = x_177 + x_184 - x_177 = x_184 = None - x_186 = x_185.permute(0, 2, 3, 1) - x_187 = torch.nn.functional.layer_norm( - x_186, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_186 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) - x_188 = x_187.permute(0, 3, 1, 2) - x_187 = None - x_189 = torch.conv2d( - x_188, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_188 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_21 = torch.nn.functional.relu(x_189, inplace=False) - x_189 = None - pow_22 = relu_21**2 - relu_21 = None - mul_21 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - * pow_22 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( - pow_22 - ) = None - x_190 = ( - mul_21 - + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - ) - mul_21 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) - x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) - x_190 = None - x_192 = torch.conv2d( - x_191, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_191 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) - x_192 = None - x_194 = x_185 + x_193 - x_185 = x_193 = None - x_195 = x_194.permute(0, 2, 3, 1) - x_196 = torch.nn.functional.layer_norm( - x_195, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_195 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) - x_197 = x_196.permute(0, 3, 1, 2) - x_196 = None - x_198 = torch.conv2d( - x_197, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_197 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_22 = torch.nn.functional.relu(x_198, inplace=False) - x_198 = None - pow_23 = relu_22**2 - relu_22 = None - mul_22 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ - * pow_23 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_23 - ) = None - x_199 = ( - mul_22 - + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_22 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_200 = torch.conv2d( - x_199, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_199 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_201 = torch.conv2d( - x_200, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_200 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_202 = x_194 + x_201 - x_194 = x_201 = None - x_203 = x_202.permute(0, 2, 3, 1) - x_204 = torch.nn.functional.layer_norm( - x_203, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_203 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) - x_205 = x_204.permute(0, 3, 1, 2) - x_204 = None - x_206 = torch.conv2d( - x_205, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_205 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_23 = torch.nn.functional.relu(x_206, inplace=False) - x_206 = None - pow_24 = relu_23**2 - relu_23 = None - mul_23 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - * pow_24 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( - pow_24 - ) = None - x_207 = ( - mul_23 - + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - ) - mul_23 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) - x_208 = torch.nn.functional.dropout(x_207, 0.0, False, False) - x_207 = None - x_209 = torch.conv2d( - x_208, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_208 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) - x_209 = None - x_211 = x_202 + x_210 - x_202 = x_210 = None - x_212 = x_211.permute(0, 2, 3, 1) - x_213 = torch.nn.functional.layer_norm( - x_212, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_212 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) - x_214 = x_213.permute(0, 3, 1, 2) - x_213 = None - x_215 = torch.conv2d( - x_214, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_214 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_24 = torch.nn.functional.relu(x_215, inplace=False) - x_215 = None - pow_25 = relu_24**2 - relu_24 = None - mul_24 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ - * pow_25 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_25 - ) = None - x_216 = ( - mul_24 - + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_24 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_217 = torch.conv2d( - x_216, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_216 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_218 = torch.conv2d( - x_217, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_217 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_219 = x_211 + x_218 - x_211 = x_218 = None - x_220 = x_219.permute(0, 2, 3, 1) - x_221 = torch.nn.functional.layer_norm( - x_220, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_220 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) - x_222 = x_221.permute(0, 3, 1, 2) - x_221 = None - x_223 = torch.conv2d( - x_222, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_222 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_25 = torch.nn.functional.relu(x_223, inplace=False) - x_223 = None - pow_26 = relu_25**2 - relu_25 = None - mul_25 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - * pow_26 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( - pow_26 - ) = None - x_224 = ( - mul_25 - + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - ) - mul_25 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) - x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) - x_224 = None - x_226 = torch.conv2d( - x_225, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_225 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) - x_226 = None - x_228 = x_219 + x_227 - x_219 = x_227 = None - x_229 = x_228.permute(0, 2, 3, 1) - x_230 = torch.nn.functional.layer_norm( - x_229, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_229 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) - x_231 = x_230.permute(0, 3, 1, 2) - x_230 = None - x_232 = torch.conv2d( - x_231, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_231 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_26 = torch.nn.functional.relu(x_232, inplace=False) - x_232 = None - pow_27 = relu_26**2 - relu_26 = None - mul_26 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ - * pow_27 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_27 - ) = None - x_233 = ( - mul_26 - + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_26 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_234 = torch.conv2d( - x_233, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_233 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_235 = torch.conv2d( - x_234, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_234 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_236 = x_228 + x_235 - x_228 = x_235 = None - x_237 = x_236.permute(0, 2, 3, 1) - x_238 = torch.nn.functional.layer_norm( - x_237, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_237 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) - x_239 = x_238.permute(0, 3, 1, 2) - x_238 = None - x_240 = torch.conv2d( - x_239, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_239 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_27 = torch.nn.functional.relu(x_240, inplace=False) - x_240 = None - pow_28 = relu_27**2 - relu_27 = None - mul_27 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - * pow_28 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( - pow_28 - ) = None - x_241 = ( - mul_27 - + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - ) - mul_27 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) - x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) - x_241 = None - x_243 = torch.conv2d( - x_242, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_242 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_244 = torch.nn.functional.dropout(x_243, 0.0, False, False) - x_243 = None - x_245 = x_236 + x_244 - x_236 = x_244 = None - x_246 = x_245.permute(0, 2, 3, 1) - x_247 = torch.nn.functional.layer_norm( - x_246, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_246 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) - x_248 = x_247.permute(0, 3, 1, 2) - x_247 = None - x_249 = torch.conv2d( - x_248, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_248 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_28 = torch.nn.functional.relu(x_249, inplace=False) - x_249 = None - pow_29 = relu_28**2 - relu_28 = None - mul_28 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ - * pow_29 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_29 - ) = None - x_250 = ( - mul_28 - + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_28 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_251 = torch.conv2d( - x_250, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_250 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_252 = torch.conv2d( - x_251, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_251 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_253 = x_245 + x_252 - x_245 = x_252 = None - x_254 = x_253.permute(0, 2, 3, 1) - x_255 = torch.nn.functional.layer_norm( - x_254, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_254 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) - x_256 = x_255.permute(0, 3, 1, 2) - x_255 = None - x_257 = torch.conv2d( - x_256, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_256 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_29 = torch.nn.functional.relu(x_257, inplace=False) - x_257 = None - pow_30 = relu_29**2 - relu_29 = None - mul_29 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - * pow_30 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( - pow_30 - ) = None - x_258 = ( - mul_29 - + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - ) - mul_29 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) - x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) - x_258 = None - x_260 = torch.conv2d( - x_259, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_259 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_261 = torch.nn.functional.dropout(x_260, 0.0, False, False) - x_260 = None - x_262 = x_253 + x_261 - x_253 = x_261 = None - x_263 = x_262.permute(0, 2, 3, 1) - x_262 = None - x_264 = torch.nn.functional.layer_norm( - x_263, - (192,), - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_263 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) - x_265 = x_264.permute(0, 3, 1, 2) - x_264 = None - x_266 = torch.conv2d( - x_265, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_265 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_2 = sym_sum // 16 - sym_sum_3 = torch.sym_sum([1, floordiv_2]) - floordiv_2 = None - reshape = x_266.reshape(1, 384, -1) - x_266 = None - x_267 = reshape.transpose(1, 2) - reshape = None - view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_30 = x_267 * view - view = None - layer_norm_33 = torch.nn.functional.layer_norm( - x_267, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_52 = sym_sum_3**2 - linear = torch._C._nn.linear( - layer_norm_33, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_33 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_1 = linear.reshape(1, pow_52, 3, 12, 32) - linear = None - qkv = reshape_1.permute(2, 0, 3, 1, 4) - reshape_1 = None - unbind = qkv.unbind(0) - qkv = None - q = unbind[0] - k = unbind[1] - v = unbind[2] - unbind = None - x_268 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) - q = k = v = None - transpose_1 = x_268.transpose(1, 2) - x_268 = None - x_269 = transpose_1.reshape(1, pow_52, 384) - transpose_1 = None - x_270 = torch._C._nn.linear( - x_269, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_269 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_271 = torch.nn.functional.dropout(x_270, 0.0, False, False) - x_270 = None - x_272 = mul_30 + x_271 - mul_30 = x_271 = None - view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_31 = x_272 * view_1 - view_1 = None - layer_norm_34 = torch.nn.functional.layer_norm( - x_272, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_272 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_273 = torch._C._nn.linear( - layer_norm_34, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_34 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_30 = torch.nn.functional.relu(x_273, inplace=False) - x_273 = None - pow_31 = relu_30**2 - relu_30 = None - mul_32 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_31 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_31 - ) = None - x_274 = ( - mul_32 - + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_275 = torch.nn.functional.dropout(x_274, 0.0, False, False) - x_274 = None - x_276 = torch._C._nn.linear( - x_275, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_277 = torch.nn.functional.dropout(x_276, 0.0, False, False) - x_276 = None - x_278 = mul_31 + x_277 - mul_31 = x_277 = None - view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_33 = x_278 * view_2 - view_2 = None - layer_norm_35 = torch.nn.functional.layer_norm( - x_278, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_4 = torch._C._nn.linear( - layer_norm_35, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_3 = linear_4.reshape(1, pow_52, 3, 12, 32) - linear_4 = None - qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) - reshape_3 = None - unbind_1 = qkv_1.unbind(0) - qkv_1 = None - q_1 = unbind_1[0] - k_1 = unbind_1[1] - v_1 = unbind_1[2] - unbind_1 = None - x_279 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) - q_1 = k_1 = v_1 = None - transpose_2 = x_279.transpose(1, 2) - x_279 = None - x_280 = transpose_2.reshape(1, pow_52, 384) - transpose_2 = None - x_281 = torch._C._nn.linear( - x_280, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_280 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) - x_281 = None - x_283 = mul_33 + x_282 - mul_33 = x_282 = None - view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_34 = x_283 * view_3 - view_3 = None - layer_norm_36 = torch.nn.functional.layer_norm( - x_283, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_283 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_284 = torch._C._nn.linear( - layer_norm_36, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_36 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_31 = torch.nn.functional.relu(x_284, inplace=False) - x_284 = None - pow_32 = relu_31**2 - relu_31 = None - mul_35 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_32 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_32 - ) = None - x_285 = ( - mul_35 - + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_286 = torch.nn.functional.dropout(x_285, 0.0, False, False) - x_285 = None - x_287 = torch._C._nn.linear( - x_286, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) - x_287 = None - x_289 = mul_34 + x_288 - mul_34 = x_288 = None - view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_36 = x_289 * view_4 - view_4 = None - layer_norm_37 = torch.nn.functional.layer_norm( - x_289, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_8 = torch._C._nn.linear( - layer_norm_37, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_37 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_5 = linear_8.reshape(1, pow_52, 3, 12, 32) - linear_8 = None - qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) - reshape_5 = None - unbind_2 = qkv_2.unbind(0) - qkv_2 = None - q_2 = unbind_2[0] - k_2 = unbind_2[1] - v_2 = unbind_2[2] - unbind_2 = None - x_290 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) - q_2 = k_2 = v_2 = None - transpose_3 = x_290.transpose(1, 2) - x_290 = None - x_291 = transpose_3.reshape(1, pow_52, 384) - transpose_3 = None - x_292 = torch._C._nn.linear( - x_291, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_291 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_293 = torch.nn.functional.dropout(x_292, 0.0, False, False) - x_292 = None - x_294 = mul_36 + x_293 - mul_36 = x_293 = None - view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_37 = x_294 * view_5 - view_5 = None - layer_norm_38 = torch.nn.functional.layer_norm( - x_294, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_294 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_295 = torch._C._nn.linear( - layer_norm_38, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_32 = torch.nn.functional.relu(x_295, inplace=False) - x_295 = None - pow_33 = relu_32**2 - relu_32 = None - mul_38 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_33 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_33 - ) = None - x_296 = ( - mul_38 - + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_297 = torch.nn.functional.dropout(x_296, 0.0, False, False) - x_296 = None - x_298 = torch._C._nn.linear( - x_297, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_299 = torch.nn.functional.dropout(x_298, 0.0, False, False) - x_298 = None - x_300 = mul_37 + x_299 - mul_37 = x_299 = None - view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_39 = x_300 * view_6 - view_6 = None - layer_norm_39 = torch.nn.functional.layer_norm( - x_300, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) - linear_12 = torch._C._nn.linear( - layer_norm_39, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_39 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_7 = linear_12.reshape(1, pow_52, 3, 12, 32) - linear_12 = None - qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) - reshape_7 = None - unbind_3 = qkv_3.unbind(0) - qkv_3 = None - q_3 = unbind_3[0] - k_3 = unbind_3[1] - v_3 = unbind_3[2] - unbind_3 = None - x_301 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) - q_3 = k_3 = v_3 = None - transpose_4 = x_301.transpose(1, 2) - x_301 = None - x_302 = transpose_4.reshape(1, pow_52, 384) - transpose_4 = None - x_303 = torch._C._nn.linear( - x_302, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_302 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_304 = torch.nn.functional.dropout(x_303, 0.0, False, False) - x_303 = None - x_305 = mul_39 + x_304 - mul_39 = x_304 = None - view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_40 = x_305 * view_7 - view_7 = None - layer_norm_40 = torch.nn.functional.layer_norm( - x_305, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_305 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) - x_306 = torch._C._nn.linear( - layer_norm_40, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_40 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_33 = torch.nn.functional.relu(x_306, inplace=False) - x_306 = None - pow_34 = relu_33**2 - relu_33 = None - mul_41 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - * pow_34 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( - pow_34 - ) = None - x_307 = ( - mul_41 - + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - ) - mul_41 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) - x_308 = torch.nn.functional.dropout(x_307, 0.0, False, False) - x_307 = None - x_309 = torch._C._nn.linear( - x_308, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) - x_309 = None - x_311 = mul_40 + x_310 - mul_40 = x_310 = None - view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_42 = x_311 * view_8 - view_8 = None - layer_norm_41 = torch.nn.functional.layer_norm( - x_311, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) - linear_16 = torch._C._nn.linear( - layer_norm_41, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_41 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_9 = linear_16.reshape(1, pow_52, 3, 12, 32) - linear_16 = None - qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) - reshape_9 = None - unbind_4 = qkv_4.unbind(0) - qkv_4 = None - q_4 = unbind_4[0] - k_4 = unbind_4[1] - v_4 = unbind_4[2] - unbind_4 = None - x_312 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) - q_4 = k_4 = v_4 = None - transpose_5 = x_312.transpose(1, 2) - x_312 = None - x_313 = transpose_5.reshape(1, pow_52, 384) - transpose_5 = None - x_314 = torch._C._nn.linear( - x_313, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_313 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) - x_314 = None - x_316 = mul_42 + x_315 - mul_42 = x_315 = None - view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_43 = x_316 * view_9 - view_9 = None - layer_norm_42 = torch.nn.functional.layer_norm( - x_316, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_316 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) - x_317 = torch._C._nn.linear( - layer_norm_42, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_42 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_34 = torch.nn.functional.relu(x_317, inplace=False) - x_317 = None - pow_35 = relu_34**2 - relu_34 = None - mul_44 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - * pow_35 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( - pow_35 - ) = None - x_318 = ( - mul_44 - + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - ) - mul_44 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) - x_319 = torch.nn.functional.dropout(x_318, 0.0, False, False) - x_318 = None - x_320 = torch._C._nn.linear( - x_319, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_321 = torch.nn.functional.dropout(x_320, 0.0, False, False) - x_320 = None - x_322 = mul_43 + x_321 - mul_43 = x_321 = None - view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_45 = x_322 * view_10 - view_10 = None - layer_norm_43 = torch.nn.functional.layer_norm( - x_322, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) - linear_20 = torch._C._nn.linear( - layer_norm_43, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_43 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_11 = linear_20.reshape(1, pow_52, 3, 12, 32) - linear_20 = None - qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) - reshape_11 = None - unbind_5 = qkv_5.unbind(0) - qkv_5 = None - q_5 = unbind_5[0] - k_5 = unbind_5[1] - v_5 = unbind_5[2] - unbind_5 = None - x_323 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) - q_5 = k_5 = v_5 = None - transpose_6 = x_323.transpose(1, 2) - x_323 = None - x_324 = transpose_6.reshape(1, pow_52, 384) - transpose_6 = None - x_325 = torch._C._nn.linear( - x_324, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_324 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_326 = torch.nn.functional.dropout(x_325, 0.0, False, False) - x_325 = None - x_327 = mul_45 + x_326 - mul_45 = x_326 = None - view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_46 = x_327 * view_11 - view_11 = None - layer_norm_44 = torch.nn.functional.layer_norm( - x_327, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_327 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) - x_328 = torch._C._nn.linear( - layer_norm_44, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_44 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_35 = torch.nn.functional.relu(x_328, inplace=False) - x_328 = None - pow_36 = relu_35**2 - relu_35 = None - mul_47 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - * pow_36 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( - pow_36 - ) = None - x_329 = ( - mul_47 - + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - ) - mul_47 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) - x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) - x_329 = None - x_331 = torch._C._nn.linear( - x_330, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_332 = torch.nn.functional.dropout(x_331, 0.0, False, False) - x_331 = None - x_333 = mul_46 + x_332 - mul_46 = x_332 = None - view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_48 = x_333 * view_12 - view_12 = None - layer_norm_45 = torch.nn.functional.layer_norm( - x_333, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) - linear_24 = torch._C._nn.linear( - layer_norm_45, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_45 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_13 = linear_24.reshape(1, pow_52, 3, 12, 32) - linear_24 = None - qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) - reshape_13 = None - unbind_6 = qkv_6.unbind(0) - qkv_6 = None - q_6 = unbind_6[0] - k_6 = unbind_6[1] - v_6 = unbind_6[2] - unbind_6 = None - x_334 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) - q_6 = k_6 = v_6 = None - transpose_7 = x_334.transpose(1, 2) - x_334 = None - x_335 = transpose_7.reshape(1, pow_52, 384) - transpose_7 = None - x_336 = torch._C._nn.linear( - x_335, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_335 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_337 = torch.nn.functional.dropout(x_336, 0.0, False, False) - x_336 = None - x_338 = mul_48 + x_337 - mul_48 = x_337 = None - view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_49 = x_338 * view_13 - view_13 = None - layer_norm_46 = torch.nn.functional.layer_norm( - x_338, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_338 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) - x_339 = torch._C._nn.linear( - layer_norm_46, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_46 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_36 = torch.nn.functional.relu(x_339, inplace=False) - x_339 = None - pow_37 = relu_36**2 - relu_36 = None - mul_50 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - * pow_37 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( - pow_37 - ) = None - x_340 = ( - mul_50 - + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - ) - mul_50 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) - x_341 = torch.nn.functional.dropout(x_340, 0.0, False, False) - x_340 = None - x_342 = torch._C._nn.linear( - x_341, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_343 = torch.nn.functional.dropout(x_342, 0.0, False, False) - x_342 = None - x_344 = mul_49 + x_343 - mul_49 = x_343 = None - view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_51 = x_344 * view_14 - view_14 = None - layer_norm_47 = torch.nn.functional.layer_norm( - x_344, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) - linear_28 = torch._C._nn.linear( - layer_norm_47, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_47 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_15 = linear_28.reshape(1, pow_52, 3, 12, 32) - linear_28 = None - qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) - reshape_15 = None - unbind_7 = qkv_7.unbind(0) - qkv_7 = None - q_7 = unbind_7[0] - k_7 = unbind_7[1] - v_7 = unbind_7[2] - unbind_7 = None - x_345 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) - q_7 = k_7 = v_7 = None - transpose_8 = x_345.transpose(1, 2) - x_345 = None - x_346 = transpose_8.reshape(1, pow_52, 384) - transpose_8 = None - x_347 = torch._C._nn.linear( - x_346, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_346 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) - x_347 = None - x_349 = mul_51 + x_348 - mul_51 = x_348 = None - view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_52 = x_349 * view_15 - view_15 = None - layer_norm_48 = torch.nn.functional.layer_norm( - x_349, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_349 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) - x_350 = torch._C._nn.linear( - layer_norm_48, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_48 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_37 = torch.nn.functional.relu(x_350, inplace=False) - x_350 = None - pow_38 = relu_37**2 - relu_37 = None - mul_53 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - * pow_38 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( - pow_38 - ) = None - x_351 = ( - mul_53 - + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - ) - mul_53 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) - x_352 = torch.nn.functional.dropout(x_351, 0.0, False, False) - x_351 = None - x_353 = torch._C._nn.linear( - x_352, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_354 = torch.nn.functional.dropout(x_353, 0.0, False, False) - x_353 = None - x_355 = mul_52 + x_354 - mul_52 = x_354 = None - view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_54 = x_355 * view_16 - view_16 = None - layer_norm_49 = torch.nn.functional.layer_norm( - x_355, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) - linear_32 = torch._C._nn.linear( - layer_norm_49, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_49 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_17 = linear_32.reshape(1, pow_52, 3, 12, 32) - linear_32 = None - qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) - reshape_17 = None - unbind_8 = qkv_8.unbind(0) - qkv_8 = None - q_8 = unbind_8[0] - k_8 = unbind_8[1] - v_8 = unbind_8[2] - unbind_8 = None - x_356 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) - q_8 = k_8 = v_8 = None - transpose_9 = x_356.transpose(1, 2) - x_356 = None - x_357 = transpose_9.reshape(1, pow_52, 384) - transpose_9 = None - x_358 = torch._C._nn.linear( - x_357, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_357 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_359 = torch.nn.functional.dropout(x_358, 0.0, False, False) - x_358 = None - x_360 = mul_54 + x_359 - mul_54 = x_359 = None - view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_55 = x_360 * view_17 - view_17 = None - layer_norm_50 = torch.nn.functional.layer_norm( - x_360, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_360 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) - x_361 = torch._C._nn.linear( - layer_norm_50, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_50 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_38 = torch.nn.functional.relu(x_361, inplace=False) - x_361 = None - pow_39 = relu_38**2 - relu_38 = None - mul_56 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - * pow_39 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( - pow_39 - ) = None - x_362 = ( - mul_56 - + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - ) - mul_56 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) - x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) - x_362 = None - x_364 = torch._C._nn.linear( - x_363, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_365 = torch.nn.functional.dropout(x_364, 0.0, False, False) - x_364 = None - x_366 = mul_55 + x_365 - mul_55 = x_365 = None - view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_57 = x_366 * view_18 - view_18 = None - layer_norm_51 = torch.nn.functional.layer_norm( - x_366, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) - linear_36 = torch._C._nn.linear( - layer_norm_51, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_51 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_19 = linear_36.reshape(1, pow_52, 3, 12, 32) - linear_36 = None - qkv_9 = reshape_19.permute(2, 0, 3, 1, 4) - reshape_19 = None - unbind_9 = qkv_9.unbind(0) - qkv_9 = None - q_9 = unbind_9[0] - k_9 = unbind_9[1] - v_9 = unbind_9[2] - unbind_9 = None - x_367 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) - q_9 = k_9 = v_9 = None - transpose_10 = x_367.transpose(1, 2) - x_367 = None - x_368 = transpose_10.reshape(1, pow_52, 384) - transpose_10 = None - x_369 = torch._C._nn.linear( - x_368, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_368 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_370 = torch.nn.functional.dropout(x_369, 0.0, False, False) - x_369 = None - x_371 = mul_57 + x_370 - mul_57 = x_370 = None - view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_58 = x_371 * view_19 - view_19 = None - layer_norm_52 = torch.nn.functional.layer_norm( - x_371, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_371 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) - x_372 = torch._C._nn.linear( - layer_norm_52, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_52 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_39 = torch.nn.functional.relu(x_372, inplace=False) - x_372 = None - pow_40 = relu_39**2 - relu_39 = None - mul_59 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - * pow_40 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( - pow_40 - ) = None - x_373 = ( - mul_59 - + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - ) - mul_59 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) - x_374 = torch.nn.functional.dropout(x_373, 0.0, False, False) - x_373 = None - x_375 = torch._C._nn.linear( - x_374, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_374 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_376 = torch.nn.functional.dropout(x_375, 0.0, False, False) - x_375 = None - x_377 = mul_58 + x_376 - mul_58 = x_376 = None - view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_60 = x_377 * view_20 - view_20 = None - layer_norm_53 = torch.nn.functional.layer_norm( - x_377, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_377 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) - linear_40 = torch._C._nn.linear( - layer_norm_53, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_53 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_21 = linear_40.reshape(1, pow_52, 3, 12, 32) - linear_40 = None - qkv_10 = reshape_21.permute(2, 0, 3, 1, 4) - reshape_21 = None - unbind_10 = qkv_10.unbind(0) - qkv_10 = None - q_10 = unbind_10[0] - k_10 = unbind_10[1] - v_10 = unbind_10[2] - unbind_10 = None - x_378 = torch._C._nn.scaled_dot_product_attention( - q_10, k_10, v_10, dropout_p=0.0 - ) - q_10 = k_10 = v_10 = None - transpose_11 = x_378.transpose(1, 2) - x_378 = None - x_379 = transpose_11.reshape(1, pow_52, 384) - transpose_11 = None - x_380 = torch._C._nn.linear( - x_379, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_379 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_381 = torch.nn.functional.dropout(x_380, 0.0, False, False) - x_380 = None - x_382 = mul_60 + x_381 - mul_60 = x_381 = None - view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_61 = x_382 * view_21 - view_21 = None - layer_norm_54 = torch.nn.functional.layer_norm( - x_382, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_382 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) - x_383 = torch._C._nn.linear( - layer_norm_54, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_54 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_40 = torch.nn.functional.relu(x_383, inplace=False) - x_383 = None - pow_41 = relu_40**2 - relu_40 = None - mul_62 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - * pow_41 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( - pow_41 - ) = None - x_384 = ( - mul_62 - + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - ) - mul_62 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) - x_385 = torch.nn.functional.dropout(x_384, 0.0, False, False) - x_384 = None - x_386 = torch._C._nn.linear( - x_385, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_385 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_387 = torch.nn.functional.dropout(x_386, 0.0, False, False) - x_386 = None - x_388 = mul_61 + x_387 - mul_61 = x_387 = None - view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_63 = x_388 * view_22 - view_22 = None - layer_norm_55 = torch.nn.functional.layer_norm( - x_388, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_388 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) - linear_44 = torch._C._nn.linear( - layer_norm_55, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_55 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_23 = linear_44.reshape(1, pow_52, 3, 12, 32) - linear_44 = None - qkv_11 = reshape_23.permute(2, 0, 3, 1, 4) - reshape_23 = None - unbind_11 = qkv_11.unbind(0) - qkv_11 = None - q_11 = unbind_11[0] - k_11 = unbind_11[1] - v_11 = unbind_11[2] - unbind_11 = None - x_389 = torch._C._nn.scaled_dot_product_attention( - q_11, k_11, v_11, dropout_p=0.0 - ) - q_11 = k_11 = v_11 = None - transpose_12 = x_389.transpose(1, 2) - x_389 = None - x_390 = transpose_12.reshape(1, pow_52, 384) - transpose_12 = None - x_391 = torch._C._nn.linear( - x_390, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_390 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_392 = torch.nn.functional.dropout(x_391, 0.0, False, False) - x_391 = None - x_393 = mul_63 + x_392 - mul_63 = x_392 = None - view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_64 = x_393 * view_23 - view_23 = None - layer_norm_56 = torch.nn.functional.layer_norm( - x_393, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_393 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) - x_394 = torch._C._nn.linear( - layer_norm_56, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_56 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_41 = torch.nn.functional.relu(x_394, inplace=False) - x_394 = None - pow_42 = relu_41**2 - relu_41 = None - mul_65 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - * pow_42 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( - pow_42 - ) = None - x_395 = ( - mul_65 - + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - ) - mul_65 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) - x_396 = torch.nn.functional.dropout(x_395, 0.0, False, False) - x_395 = None - x_397 = torch._C._nn.linear( - x_396, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_396 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_398 = torch.nn.functional.dropout(x_397, 0.0, False, False) - x_397 = None - x_399 = mul_64 + x_398 - mul_64 = x_398 = None - view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_66 = x_399 * view_24 - view_24 = None - layer_norm_57 = torch.nn.functional.layer_norm( - x_399, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_399 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = (None) - linear_48 = torch._C._nn.linear( - layer_norm_57, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_57 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_25 = linear_48.reshape(1, pow_52, 3, 12, 32) - linear_48 = None - qkv_12 = reshape_25.permute(2, 0, 3, 1, 4) - reshape_25 = None - unbind_12 = qkv_12.unbind(0) - qkv_12 = None - q_12 = unbind_12[0] - k_12 = unbind_12[1] - v_12 = unbind_12[2] - unbind_12 = None - x_400 = torch._C._nn.scaled_dot_product_attention( - q_12, k_12, v_12, dropout_p=0.0 - ) - q_12 = k_12 = v_12 = None - transpose_13 = x_400.transpose(1, 2) - x_400 = None - x_401 = transpose_13.reshape(1, pow_52, 384) - transpose_13 = None - x_402 = torch._C._nn.linear( - x_401, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_401 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_403 = torch.nn.functional.dropout(x_402, 0.0, False, False) - x_402 = None - x_404 = mul_66 + x_403 - mul_66 = x_403 = None - view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_67 = x_404 * view_25 - view_25 = None - layer_norm_58 = torch.nn.functional.layer_norm( - x_404, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_404 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = (None) - x_405 = torch._C._nn.linear( - layer_norm_58, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_58 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_42 = torch.nn.functional.relu(x_405, inplace=False) - x_405 = None - pow_43 = relu_42**2 - relu_42 = None - mul_68 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ - * pow_43 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = ( - pow_43 - ) = None - x_406 = ( - mul_68 - + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ - ) - mul_68 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = (None) - x_407 = torch.nn.functional.dropout(x_406, 0.0, False, False) - x_406 = None - x_408 = torch._C._nn.linear( - x_407, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_407 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_409 = torch.nn.functional.dropout(x_408, 0.0, False, False) - x_408 = None - x_410 = mul_67 + x_409 - mul_67 = x_409 = None - view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_69 = x_410 * view_26 - view_26 = None - layer_norm_59 = torch.nn.functional.layer_norm( - x_410, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_410 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = (None) - linear_52 = torch._C._nn.linear( - layer_norm_59, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_59 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_27 = linear_52.reshape(1, pow_52, 3, 12, 32) - linear_52 = None - qkv_13 = reshape_27.permute(2, 0, 3, 1, 4) - reshape_27 = None - unbind_13 = qkv_13.unbind(0) - qkv_13 = None - q_13 = unbind_13[0] - k_13 = unbind_13[1] - v_13 = unbind_13[2] - unbind_13 = None - x_411 = torch._C._nn.scaled_dot_product_attention( - q_13, k_13, v_13, dropout_p=0.0 - ) - q_13 = k_13 = v_13 = None - transpose_14 = x_411.transpose(1, 2) - x_411 = None - x_412 = transpose_14.reshape(1, pow_52, 384) - transpose_14 = None - x_413 = torch._C._nn.linear( - x_412, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_412 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_414 = torch.nn.functional.dropout(x_413, 0.0, False, False) - x_413 = None - x_415 = mul_69 + x_414 - mul_69 = x_414 = None - view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_70 = x_415 * view_27 - view_27 = None - layer_norm_60 = torch.nn.functional.layer_norm( - x_415, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_415 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = (None) - x_416 = torch._C._nn.linear( - layer_norm_60, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_60 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_43 = torch.nn.functional.relu(x_416, inplace=False) - x_416 = None - pow_44 = relu_43**2 - relu_43 = None - mul_71 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ - * pow_44 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = ( - pow_44 - ) = None - x_417 = ( - mul_71 - + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ - ) - mul_71 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = (None) - x_418 = torch.nn.functional.dropout(x_417, 0.0, False, False) - x_417 = None - x_419 = torch._C._nn.linear( - x_418, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_418 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_420 = torch.nn.functional.dropout(x_419, 0.0, False, False) - x_419 = None - x_421 = mul_70 + x_420 - mul_70 = x_420 = None - view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_72 = x_421 * view_28 - view_28 = None - layer_norm_61 = torch.nn.functional.layer_norm( - x_421, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_421 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = (None) - linear_56 = torch._C._nn.linear( - layer_norm_61, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_61 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_29 = linear_56.reshape(1, pow_52, 3, 12, 32) - linear_56 = None - qkv_14 = reshape_29.permute(2, 0, 3, 1, 4) - reshape_29 = None - unbind_14 = qkv_14.unbind(0) - qkv_14 = None - q_14 = unbind_14[0] - k_14 = unbind_14[1] - v_14 = unbind_14[2] - unbind_14 = None - x_422 = torch._C._nn.scaled_dot_product_attention( - q_14, k_14, v_14, dropout_p=0.0 - ) - q_14 = k_14 = v_14 = None - transpose_15 = x_422.transpose(1, 2) - x_422 = None - x_423 = transpose_15.reshape(1, pow_52, 384) - transpose_15 = None - x_424 = torch._C._nn.linear( - x_423, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_423 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_425 = torch.nn.functional.dropout(x_424, 0.0, False, False) - x_424 = None - x_426 = mul_72 + x_425 - mul_72 = x_425 = None - view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_73 = x_426 * view_29 - view_29 = None - layer_norm_62 = torch.nn.functional.layer_norm( - x_426, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_426 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = (None) - x_427 = torch._C._nn.linear( - layer_norm_62, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_62 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_44 = torch.nn.functional.relu(x_427, inplace=False) - x_427 = None - pow_45 = relu_44**2 - relu_44 = None - mul_74 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ - * pow_45 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = ( - pow_45 - ) = None - x_428 = ( - mul_74 - + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ - ) - mul_74 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = (None) - x_429 = torch.nn.functional.dropout(x_428, 0.0, False, False) - x_428 = None - x_430 = torch._C._nn.linear( - x_429, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_429 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_431 = torch.nn.functional.dropout(x_430, 0.0, False, False) - x_430 = None - x_432 = mul_73 + x_431 - mul_73 = x_431 = None - view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_75 = x_432 * view_30 - view_30 = None - layer_norm_63 = torch.nn.functional.layer_norm( - x_432, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_432 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = (None) - linear_60 = torch._C._nn.linear( - layer_norm_63, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_63 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_31 = linear_60.reshape(1, pow_52, 3, 12, 32) - linear_60 = None - qkv_15 = reshape_31.permute(2, 0, 3, 1, 4) - reshape_31 = None - unbind_15 = qkv_15.unbind(0) - qkv_15 = None - q_15 = unbind_15[0] - k_15 = unbind_15[1] - v_15 = unbind_15[2] - unbind_15 = None - x_433 = torch._C._nn.scaled_dot_product_attention( - q_15, k_15, v_15, dropout_p=0.0 - ) - q_15 = k_15 = v_15 = None - transpose_16 = x_433.transpose(1, 2) - x_433 = None - x_434 = transpose_16.reshape(1, pow_52, 384) - transpose_16 = None - x_435 = torch._C._nn.linear( - x_434, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_434 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_436 = torch.nn.functional.dropout(x_435, 0.0, False, False) - x_435 = None - x_437 = mul_75 + x_436 - mul_75 = x_436 = None - view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_76 = x_437 * view_31 - view_31 = None - layer_norm_64 = torch.nn.functional.layer_norm( - x_437, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_437 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = (None) - x_438 = torch._C._nn.linear( - layer_norm_64, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_64 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_45 = torch.nn.functional.relu(x_438, inplace=False) - x_438 = None - pow_46 = relu_45**2 - relu_45 = None - mul_77 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ - * pow_46 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = ( - pow_46 - ) = None - x_439 = ( - mul_77 - + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ - ) - mul_77 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = (None) - x_440 = torch.nn.functional.dropout(x_439, 0.0, False, False) - x_439 = None - x_441 = torch._C._nn.linear( - x_440, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_440 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_442 = torch.nn.functional.dropout(x_441, 0.0, False, False) - x_441 = None - x_443 = mul_76 + x_442 - mul_76 = x_442 = None - view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_78 = x_443 * view_32 - view_32 = None - layer_norm_65 = torch.nn.functional.layer_norm( - x_443, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_443 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = (None) - linear_64 = torch._C._nn.linear( - layer_norm_65, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_65 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_33 = linear_64.reshape(1, pow_52, 3, 12, 32) - linear_64 = None - qkv_16 = reshape_33.permute(2, 0, 3, 1, 4) - reshape_33 = None - unbind_16 = qkv_16.unbind(0) - qkv_16 = None - q_16 = unbind_16[0] - k_16 = unbind_16[1] - v_16 = unbind_16[2] - unbind_16 = None - x_444 = torch._C._nn.scaled_dot_product_attention( - q_16, k_16, v_16, dropout_p=0.0 - ) - q_16 = k_16 = v_16 = None - transpose_17 = x_444.transpose(1, 2) - x_444 = None - x_445 = transpose_17.reshape(1, pow_52, 384) - transpose_17 = None - x_446 = torch._C._nn.linear( - x_445, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_445 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_447 = torch.nn.functional.dropout(x_446, 0.0, False, False) - x_446 = None - x_448 = mul_78 + x_447 - mul_78 = x_447 = None - view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_79 = x_448 * view_33 - view_33 = None - layer_norm_66 = torch.nn.functional.layer_norm( - x_448, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_448 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = (None) - x_449 = torch._C._nn.linear( - layer_norm_66, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_66 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_46 = torch.nn.functional.relu(x_449, inplace=False) - x_449 = None - pow_47 = relu_46**2 - relu_46 = None - mul_80 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ - * pow_47 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = ( - pow_47 - ) = None - x_450 = ( - mul_80 - + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ - ) - mul_80 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = (None) - x_451 = torch.nn.functional.dropout(x_450, 0.0, False, False) - x_450 = None - x_452 = torch._C._nn.linear( - x_451, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_451 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_453 = torch.nn.functional.dropout(x_452, 0.0, False, False) - x_452 = None - x_454 = mul_79 + x_453 - mul_79 = x_453 = None - view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_81 = x_454 * view_34 - view_34 = None - layer_norm_67 = torch.nn.functional.layer_norm( - x_454, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_454 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = (None) - linear_68 = torch._C._nn.linear( - layer_norm_67, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_67 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_35 = linear_68.reshape(1, pow_52, 3, 12, 32) - linear_68 = None - qkv_17 = reshape_35.permute(2, 0, 3, 1, 4) - reshape_35 = None - unbind_17 = qkv_17.unbind(0) - qkv_17 = None - q_17 = unbind_17[0] - k_17 = unbind_17[1] - v_17 = unbind_17[2] - unbind_17 = None - x_455 = torch._C._nn.scaled_dot_product_attention( - q_17, k_17, v_17, dropout_p=0.0 - ) - q_17 = k_17 = v_17 = None - transpose_18 = x_455.transpose(1, 2) - x_455 = None - x_456 = transpose_18.reshape(1, pow_52, 384) - transpose_18 = pow_52 = None - x_457 = torch._C._nn.linear( - x_456, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_456 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_458 = torch.nn.functional.dropout(x_457, 0.0, False, False) - x_457 = None - x_459 = mul_81 + x_458 - mul_81 = x_458 = None - view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_82 = x_459 * view_35 - view_35 = None - layer_norm_68 = torch.nn.functional.layer_norm( - x_459, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_459 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = (None) - x_460 = torch._C._nn.linear( - layer_norm_68, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_68 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_47 = torch.nn.functional.relu(x_460, inplace=False) - x_460 = None - pow_48 = relu_47**2 - relu_47 = None - mul_83 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ - * pow_48 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = ( - pow_48 - ) = None - x_461 = ( - mul_83 - + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ - ) - mul_83 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = (None) - x_462 = torch.nn.functional.dropout(x_461, 0.0, False, False) - x_461 = None - x_463 = torch._C._nn.linear( - x_462, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_462 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_464 = torch.nn.functional.dropout(x_463, 0.0, False, False) - x_463 = None - x_465 = mul_82 + x_464 - mul_82 = x_464 = None - transpose_19 = x_465.transpose(1, 2) - x_465 = None - x_466 = transpose_19.reshape(1, 384, sym_sum_3, sym_sum_3) - transpose_19 = sym_sum_3 = None - x_467 = x_466.permute(0, 2, 3, 1) - x_466 = None - x_468 = torch.nn.functional.layer_norm( - x_467, - (384,), - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_467 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) - x_469 = x_468.permute(0, 3, 1, 2) - x_468 = None - x_470 = torch.conv2d( - x_469, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_469 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_3 = sym_sum // 32 - sym_sum = None - sym_sum_4 = torch.sym_sum([1, floordiv_3]) - floordiv_3 = None - reshape_38 = x_470.reshape(1, 576, -1) - x_470 = None - x_471 = reshape_38.transpose(1, 2) - reshape_38 = None - view_36 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_84 = x_471 * view_36 - view_36 = None - layer_norm_70 = torch.nn.functional.layer_norm( - x_471, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_471 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_53 = sym_sum_4**2 - linear_72 = torch._C._nn.linear( - layer_norm_70, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_70 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_39 = linear_72.reshape(1, pow_53, 3, 18, 32) - linear_72 = None - qkv_18 = reshape_39.permute(2, 0, 3, 1, 4) - reshape_39 = None - unbind_18 = qkv_18.unbind(0) - qkv_18 = None - q_18 = unbind_18[0] - k_18 = unbind_18[1] - v_18 = unbind_18[2] - unbind_18 = None - x_472 = torch._C._nn.scaled_dot_product_attention( - q_18, k_18, v_18, dropout_p=0.0 - ) - q_18 = k_18 = v_18 = None - transpose_21 = x_472.transpose(1, 2) - x_472 = None - x_473 = transpose_21.reshape(1, pow_53, 576) - transpose_21 = None - x_474 = torch._C._nn.linear( - x_473, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_473 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_475 = torch.nn.functional.dropout(x_474, 0.0, False, False) - x_474 = None - x_476 = mul_84 + x_475 - mul_84 = x_475 = None - view_37 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_85 = x_476 * view_37 - view_37 = None - layer_norm_71 = torch.nn.functional.layer_norm( - x_476, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_476 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_477 = torch._C._nn.linear( - layer_norm_71, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_71 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_48 = torch.nn.functional.relu(x_477, inplace=False) - x_477 = None - pow_49 = relu_48**2 - relu_48 = None - mul_86 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_49 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_49 - ) = None - x_478 = ( - mul_86 - + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_86 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_479 = torch.nn.functional.dropout(x_478, 0.0, False, False) - x_478 = None - x_480 = torch._C._nn.linear( - x_479, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_479 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_481 = torch.nn.functional.dropout(x_480, 0.0, False, False) - x_480 = None - x_482 = mul_85 + x_481 - mul_85 = x_481 = None - view_38 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_87 = x_482 * view_38 - view_38 = None - layer_norm_72 = torch.nn.functional.layer_norm( - x_482, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_482 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_76 = torch._C._nn.linear( - layer_norm_72, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_72 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_41 = linear_76.reshape(1, pow_53, 3, 18, 32) - linear_76 = None - qkv_19 = reshape_41.permute(2, 0, 3, 1, 4) - reshape_41 = None - unbind_19 = qkv_19.unbind(0) - qkv_19 = None - q_19 = unbind_19[0] - k_19 = unbind_19[1] - v_19 = unbind_19[2] - unbind_19 = None - x_483 = torch._C._nn.scaled_dot_product_attention( - q_19, k_19, v_19, dropout_p=0.0 - ) - q_19 = k_19 = v_19 = None - transpose_22 = x_483.transpose(1, 2) - x_483 = None - x_484 = transpose_22.reshape(1, pow_53, 576) - transpose_22 = None - x_485 = torch._C._nn.linear( - x_484, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_484 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_486 = torch.nn.functional.dropout(x_485, 0.0, False, False) - x_485 = None - x_487 = mul_87 + x_486 - mul_87 = x_486 = None - view_39 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_88 = x_487 * view_39 - view_39 = None - layer_norm_73 = torch.nn.functional.layer_norm( - x_487, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_487 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_488 = torch._C._nn.linear( - layer_norm_73, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_73 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_49 = torch.nn.functional.relu(x_488, inplace=False) - x_488 = None - pow_50 = relu_49**2 - relu_49 = None - mul_89 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_50 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_50 - ) = None - x_489 = ( - mul_89 - + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_89 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_490 = torch.nn.functional.dropout(x_489, 0.0, False, False) - x_489 = None - x_491 = torch._C._nn.linear( - x_490, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_490 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_492 = torch.nn.functional.dropout(x_491, 0.0, False, False) - x_491 = None - x_493 = mul_88 + x_492 - mul_88 = x_492 = None - view_40 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_90 = x_493 * view_40 - view_40 = None - layer_norm_74 = torch.nn.functional.layer_norm( - x_493, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_493 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_80 = torch._C._nn.linear( - layer_norm_74, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_74 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_43 = linear_80.reshape(1, pow_53, 3, 18, 32) - linear_80 = None - qkv_20 = reshape_43.permute(2, 0, 3, 1, 4) - reshape_43 = None - unbind_20 = qkv_20.unbind(0) - qkv_20 = None - q_20 = unbind_20[0] - k_20 = unbind_20[1] - v_20 = unbind_20[2] - unbind_20 = None - x_494 = torch._C._nn.scaled_dot_product_attention( - q_20, k_20, v_20, dropout_p=0.0 - ) - q_20 = k_20 = v_20 = None - transpose_23 = x_494.transpose(1, 2) - x_494 = None - x_495 = transpose_23.reshape(1, pow_53, 576) - transpose_23 = pow_53 = None - x_496 = torch._C._nn.linear( - x_495, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_495 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_497 = torch.nn.functional.dropout(x_496, 0.0, False, False) - x_496 = None - x_498 = mul_90 + x_497 - mul_90 = x_497 = None - view_41 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_91 = x_498 * view_41 - view_41 = None - layer_norm_75 = torch.nn.functional.layer_norm( - x_498, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_498 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_499 = torch._C._nn.linear( - layer_norm_75, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_75 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_50 = torch.nn.functional.relu(x_499, inplace=False) - x_499 = None - pow_51 = relu_50**2 - relu_50 = None - mul_92 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_51 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_51 - ) = None - x_500 = ( - mul_92 - + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_92 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_501 = torch.nn.functional.dropout(x_500, 0.0, False, False) - x_500 = None - x_502 = torch._C._nn.linear( - x_501, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_501 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_503 = torch.nn.functional.dropout(x_502, 0.0, False, False) - x_502 = None - x_504 = mul_91 + x_503 - mul_91 = x_503 = None - transpose_24 = x_504.transpose(1, 2) - x_504 = None - x_505 = transpose_24.reshape(1, 576, sym_sum_4, sym_sum_4) - transpose_24 = sym_sum_4 = None - x_506 = torch.nn.functional.adaptive_avg_pool2d(x_505, 1) - x_505 = None - x_507 = x_506.permute(0, 2, 3, 1) - x_506 = None - x_508 = torch.nn.functional.layer_norm( - x_507, - (576,), - l_self_modules_head_modules_norm_parameters_weight_, - l_self_modules_head_modules_norm_parameters_bias_, - 1e-06, - ) - x_507 = ( - l_self_modules_head_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_norm_parameters_bias_ = None - x_509 = x_508.permute(0, 3, 1, 2) - x_508 = None - x_510 = x_509.flatten(1, -1) - x_509 = None - x_511 = torch.nn.functional.dropout(x_510, 0.0, False, False) - x_510 = None - x_512 = torch._C._nn.linear( - x_511, - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, - ) - x_511 = ( - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None - relu_51 = torch.nn.functional.relu(x_512, inplace=False) - x_512 = None - x_513 = torch.square(relu_51) - relu_51 = None - x_514 = torch.nn.functional.layer_norm( - x_513, - (2304,), - l_self_modules_head_modules_fc_modules_norm_parameters_weight_, - l_self_modules_head_modules_fc_modules_norm_parameters_bias_, - 1e-06, - ) - x_513 = ( - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None - x_515 = torch.nn.functional.dropout(x_514, 0.0, False, False) - x_514 = None - x_516 = torch._C._nn.linear( - x_515, - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, - ) - x_515 = ( - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None - return (x_516,) diff --git a/samples/timm/caformer_m36.sail_in1k/weight_meta.py b/samples/timm/caformer_m36.sail_in1k/weight_meta.py deleted file mode 100644 index a2c0914908..0000000000 --- a/samples/timm/caformer_m36.sail_in1k/weight_meta.py +++ /dev/null @@ -1,3968 +0,0 @@ -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: - name = "L_self_modules_stem_modules_conv_parameters_weight_" - shape = [96, 3, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: - name = "L_self_modules_stem_modules_conv_parameters_bias_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_s1: - name = "s1" - shape = [] - dtype = "torch.int64" - device = "cpu" - mean = None - std = None - data = [4] - - -class Program_weight_tensor_meta_L_x_: - name = "L_x_" - shape = [1, 3, 224, 224] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.223 - std = 1.285 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: - name = "L_self_modules_stem_modules_norm_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [192, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [192, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [96, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [384, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [96, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [192, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [192, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [96, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [384, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [96, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [192, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [192, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [96, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [384, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [96, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" - shape = [192, 96, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.021 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" - shape = [384, 192, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" - shape = [576, 384, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1728, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [576, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [2304, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [576, 2304] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1728, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [576, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [2304, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [576, 2304] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1728, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [576, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [2304, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [576, 2304] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_norm_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_norm_parameters_bias_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" - shape = [2304, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" - shape = [2304] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" - shape = [2304] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" - shape = [2304] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" - shape = [1000, 2304] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" - shape = [1000] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None diff --git a/samples/timm/caformer_m36.sail_in1k_384/graph_net.json b/samples/timm/caformer_m36.sail_in1k_384/graph_net.json deleted file mode 100644 index 1373fe3b58..0000000000 --- a/samples/timm/caformer_m36.sail_in1k_384/graph_net.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "framework": "torch", - "num_devices_required": 1, - "num_nodes_required": 1 -} \ No newline at end of file diff --git a/samples/timm/caformer_m36.sail_in1k_384/input_meta.py b/samples/timm/caformer_m36.sail_in1k_384/input_meta.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_m36.sail_in1k_384/input_tensor_constraints.py b/samples/timm/caformer_m36.sail_in1k_384/input_tensor_constraints.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_m36.sail_in1k_384/model.py b/samples/timm/caformer_m36.sail_in1k_384/model.py deleted file mode 100644 index dcce286c5c..0000000000 --- a/samples/timm/caformer_m36.sail_in1k_384/model.py +++ /dev/null @@ -1,4754 +0,0 @@ -import torch - - -class GraphModule(torch.nn.Module): - def forward( - self, - L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - s1: torch.SymInt, - L_x_: torch.Tensor, - L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - ): - l_self_modules_stem_modules_conv_parameters_weight_ = ( - L_self_modules_stem_modules_conv_parameters_weight_ - ) - l_self_modules_stem_modules_conv_parameters_bias_ = ( - L_self_modules_stem_modules_conv_parameters_bias_ - ) - l_x_ = L_x_ - l_self_modules_stem_modules_norm_parameters_weight_ = ( - L_self_modules_stem_modules_norm_parameters_weight_ - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_head_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ - ) - x = torch.conv2d( - l_x_, - l_self_modules_stem_modules_conv_parameters_weight_, - l_self_modules_stem_modules_conv_parameters_bias_, - (4, 4), - (2, 2), - (1, 1), - 1, - ) - l_x_ = ( - l_self_modules_stem_modules_conv_parameters_weight_ - ) = l_self_modules_stem_modules_conv_parameters_bias_ = None - x_1 = x.permute(0, 2, 3, 1) - x = None - x_2 = torch.nn.functional.layer_norm( - x_1, (96,), l_self_modules_stem_modules_norm_parameters_weight_, None, 1e-06 - ) - x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None - x_3 = x_2.permute(0, 3, 1, 2) - x_2 = None - sym_sum = torch.sym_sum([-3, s1]) - s1 = None - floordiv = sym_sum // 4 - sym_sum_1 = torch.sym_sum([1, floordiv]) - floordiv = sym_sum_1 = None - x_4 = x_3.permute(0, 2, 3, 1) - x_5 = torch.nn.functional.layer_norm( - x_4, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_6 = x_5.permute(0, 3, 1, 2) - x_5 = None - x_7 = torch.conv2d( - x_6, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu = torch.nn.functional.relu(x_7, inplace=False) - x_7 = None - pow_1 = relu**2 - relu = None - mul = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_1 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_1 - ) = None - x_8 = ( - mul - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_9 = torch.conv2d( - x_8, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 192, - ) - x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_10 = torch.conv2d( - x_9, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_11 = x_3 + x_10 - x_3 = x_10 = None - x_12 = x_11.permute(0, 2, 3, 1) - x_13 = torch.nn.functional.layer_norm( - x_12, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_14 = x_13.permute(0, 3, 1, 2) - x_13 = None - x_15 = torch.conv2d( - x_14, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_1 = torch.nn.functional.relu(x_15, inplace=False) - x_15 = None - pow_2 = relu_1**2 - relu_1 = None - mul_1 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_2 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_2 - ) = None - x_16 = ( - mul_1 - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) - x_16 = None - x_18 = torch.conv2d( - x_17, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) - x_18 = None - x_20 = x_11 + x_19 - x_11 = x_19 = None - x_21 = x_20.permute(0, 2, 3, 1) - x_22 = torch.nn.functional.layer_norm( - x_21, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_23 = x_22.permute(0, 3, 1, 2) - x_22 = None - x_24 = torch.conv2d( - x_23, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_2 = torch.nn.functional.relu(x_24, inplace=False) - x_24 = None - pow_3 = relu_2**2 - relu_2 = None - mul_2 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_3 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_3 - ) = None - x_25 = ( - mul_2 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_26 = torch.conv2d( - x_25, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 192, - ) - x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_27 = torch.conv2d( - x_26, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_28 = x_20 + x_27 - x_20 = x_27 = None - x_29 = x_28.permute(0, 2, 3, 1) - x_30 = torch.nn.functional.layer_norm( - x_29, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_31 = x_30.permute(0, 3, 1, 2) - x_30 = None - x_32 = torch.conv2d( - x_31, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_3 = torch.nn.functional.relu(x_32, inplace=False) - x_32 = None - pow_4 = relu_3**2 - relu_3 = None - mul_3 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_4 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_4 - ) = None - x_33 = ( - mul_3 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) - x_33 = None - x_35 = torch.conv2d( - x_34, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) - x_35 = None - x_37 = x_28 + x_36 - x_28 = x_36 = None - x_38 = x_37.permute(0, 2, 3, 1) - x_39 = torch.nn.functional.layer_norm( - x_38, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_40 = x_39.permute(0, 3, 1, 2) - x_39 = None - x_41 = torch.conv2d( - x_40, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_4 = torch.nn.functional.relu(x_41, inplace=False) - x_41 = None - pow_5 = relu_4**2 - relu_4 = None - mul_4 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_5 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_5 - ) = None - x_42 = ( - mul_4 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_43 = torch.conv2d( - x_42, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 192, - ) - x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_44 = torch.conv2d( - x_43, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_45 = x_37 + x_44 - x_37 = x_44 = None - x_46 = x_45.permute(0, 2, 3, 1) - x_47 = torch.nn.functional.layer_norm( - x_46, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_48 = x_47.permute(0, 3, 1, 2) - x_47 = None - x_49 = torch.conv2d( - x_48, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_5 = torch.nn.functional.relu(x_49, inplace=False) - x_49 = None - pow_6 = relu_5**2 - relu_5 = None - mul_5 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_6 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_6 - ) = None - x_50 = ( - mul_5 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) - x_50 = None - x_52 = torch.conv2d( - x_51, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) - x_52 = None - x_54 = x_45 + x_53 - x_45 = x_53 = None - x_55 = x_54.permute(0, 2, 3, 1) - x_54 = None - x_56 = torch.nn.functional.layer_norm( - x_55, - (96,), - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) - x_57 = x_56.permute(0, 3, 1, 2) - x_56 = None - x_58 = torch.conv2d( - x_57, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_1 = sym_sum // 8 - sym_sum_2 = torch.sym_sum([1, floordiv_1]) - floordiv_1 = sym_sum_2 = None - x_59 = x_58.permute(0, 2, 3, 1) - x_60 = torch.nn.functional.layer_norm( - x_59, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_61 = x_60.permute(0, 3, 1, 2) - x_60 = None - x_62 = torch.conv2d( - x_61, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_6 = torch.nn.functional.relu(x_62, inplace=False) - x_62 = None - pow_7 = relu_6**2 - relu_6 = None - mul_6 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_7 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_7 - ) = None - x_63 = ( - mul_6 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_64 = torch.conv2d( - x_63, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_65 = torch.conv2d( - x_64, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_66 = x_58 + x_65 - x_58 = x_65 = None - x_67 = x_66.permute(0, 2, 3, 1) - x_68 = torch.nn.functional.layer_norm( - x_67, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_69 = x_68.permute(0, 3, 1, 2) - x_68 = None - x_70 = torch.conv2d( - x_69, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_7 = torch.nn.functional.relu(x_70, inplace=False) - x_70 = None - pow_8 = relu_7**2 - relu_7 = None - mul_7 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_8 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_8 - ) = None - x_71 = ( - mul_7 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) - x_71 = None - x_73 = torch.conv2d( - x_72, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) - x_73 = None - x_75 = x_66 + x_74 - x_66 = x_74 = None - x_76 = x_75.permute(0, 2, 3, 1) - x_77 = torch.nn.functional.layer_norm( - x_76, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_78 = x_77.permute(0, 3, 1, 2) - x_77 = None - x_79 = torch.conv2d( - x_78, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_8 = torch.nn.functional.relu(x_79, inplace=False) - x_79 = None - pow_9 = relu_8**2 - relu_8 = None - mul_8 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_9 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_9 - ) = None - x_80 = ( - mul_8 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_81 = torch.conv2d( - x_80, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_82 = torch.conv2d( - x_81, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_83 = x_75 + x_82 - x_75 = x_82 = None - x_84 = x_83.permute(0, 2, 3, 1) - x_85 = torch.nn.functional.layer_norm( - x_84, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_86 = x_85.permute(0, 3, 1, 2) - x_85 = None - x_87 = torch.conv2d( - x_86, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_9 = torch.nn.functional.relu(x_87, inplace=False) - x_87 = None - pow_10 = relu_9**2 - relu_9 = None - mul_9 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_10 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_10 - ) = None - x_88 = ( - mul_9 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) - x_88 = None - x_90 = torch.conv2d( - x_89, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) - x_90 = None - x_92 = x_83 + x_91 - x_83 = x_91 = None - x_93 = x_92.permute(0, 2, 3, 1) - x_94 = torch.nn.functional.layer_norm( - x_93, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_95 = x_94.permute(0, 3, 1, 2) - x_94 = None - x_96 = torch.conv2d( - x_95, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_10 = torch.nn.functional.relu(x_96, inplace=False) - x_96 = None - pow_11 = relu_10**2 - relu_10 = None - mul_10 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_11 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_11 - ) = None - x_97 = ( - mul_10 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_98 = torch.conv2d( - x_97, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_99 = torch.conv2d( - x_98, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_100 = x_92 + x_99 - x_92 = x_99 = None - x_101 = x_100.permute(0, 2, 3, 1) - x_102 = torch.nn.functional.layer_norm( - x_101, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_103 = x_102.permute(0, 3, 1, 2) - x_102 = None - x_104 = torch.conv2d( - x_103, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_11 = torch.nn.functional.relu(x_104, inplace=False) - x_104 = None - pow_12 = relu_11**2 - relu_11 = None - mul_11 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_12 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_12 - ) = None - x_105 = ( - mul_11 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) - x_105 = None - x_107 = torch.conv2d( - x_106, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) - x_107 = None - x_109 = x_100 + x_108 - x_100 = x_108 = None - x_110 = x_109.permute(0, 2, 3, 1) - x_111 = torch.nn.functional.layer_norm( - x_110, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_110 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) - x_112 = x_111.permute(0, 3, 1, 2) - x_111 = None - x_113 = torch.conv2d( - x_112, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_112 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_12 = torch.nn.functional.relu(x_113, inplace=False) - x_113 = None - pow_13 = relu_12**2 - relu_12 = None - mul_12 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ - * pow_13 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_13 - ) = None - x_114 = ( - mul_12 - + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_12 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_115 = torch.conv2d( - x_114, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_114 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_116 = torch.conv2d( - x_115, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_115 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_117 = x_109 + x_116 - x_109 = x_116 = None - x_118 = x_117.permute(0, 2, 3, 1) - x_119 = torch.nn.functional.layer_norm( - x_118, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_118 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) - x_120 = x_119.permute(0, 3, 1, 2) - x_119 = None - x_121 = torch.conv2d( - x_120, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_120 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_13 = torch.nn.functional.relu(x_121, inplace=False) - x_121 = None - pow_14 = relu_13**2 - relu_13 = None - mul_13 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - * pow_14 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( - pow_14 - ) = None - x_122 = ( - mul_13 - + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - ) - mul_13 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) - x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) - x_122 = None - x_124 = torch.conv2d( - x_123, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_123 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) - x_124 = None - x_126 = x_117 + x_125 - x_117 = x_125 = None - x_127 = x_126.permute(0, 2, 3, 1) - x_128 = torch.nn.functional.layer_norm( - x_127, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_127 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) - x_129 = x_128.permute(0, 3, 1, 2) - x_128 = None - x_130 = torch.conv2d( - x_129, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_129 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_14 = torch.nn.functional.relu(x_130, inplace=False) - x_130 = None - pow_15 = relu_14**2 - relu_14 = None - mul_14 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ - * pow_15 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_15 - ) = None - x_131 = ( - mul_14 - + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_14 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_132 = torch.conv2d( - x_131, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_131 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_133 = torch.conv2d( - x_132, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_132 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_134 = x_126 + x_133 - x_126 = x_133 = None - x_135 = x_134.permute(0, 2, 3, 1) - x_136 = torch.nn.functional.layer_norm( - x_135, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_135 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) - x_137 = x_136.permute(0, 3, 1, 2) - x_136 = None - x_138 = torch.conv2d( - x_137, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_137 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_15 = torch.nn.functional.relu(x_138, inplace=False) - x_138 = None - pow_16 = relu_15**2 - relu_15 = None - mul_15 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - * pow_16 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( - pow_16 - ) = None - x_139 = ( - mul_15 - + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - ) - mul_15 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) - x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) - x_139 = None - x_141 = torch.conv2d( - x_140, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_140 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) - x_141 = None - x_143 = x_134 + x_142 - x_134 = x_142 = None - x_144 = x_143.permute(0, 2, 3, 1) - x_145 = torch.nn.functional.layer_norm( - x_144, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_144 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) - x_146 = x_145.permute(0, 3, 1, 2) - x_145 = None - x_147 = torch.conv2d( - x_146, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_146 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_16 = torch.nn.functional.relu(x_147, inplace=False) - x_147 = None - pow_17 = relu_16**2 - relu_16 = None - mul_16 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ - * pow_17 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_17 - ) = None - x_148 = ( - mul_16 - + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_16 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_149 = torch.conv2d( - x_148, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_148 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_150 = torch.conv2d( - x_149, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_149 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_151 = x_143 + x_150 - x_143 = x_150 = None - x_152 = x_151.permute(0, 2, 3, 1) - x_153 = torch.nn.functional.layer_norm( - x_152, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_152 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) - x_154 = x_153.permute(0, 3, 1, 2) - x_153 = None - x_155 = torch.conv2d( - x_154, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_154 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_17 = torch.nn.functional.relu(x_155, inplace=False) - x_155 = None - pow_18 = relu_17**2 - relu_17 = None - mul_17 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - * pow_18 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( - pow_18 - ) = None - x_156 = ( - mul_17 - + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - ) - mul_17 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) - x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) - x_156 = None - x_158 = torch.conv2d( - x_157, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_157 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) - x_158 = None - x_160 = x_151 + x_159 - x_151 = x_159 = None - x_161 = x_160.permute(0, 2, 3, 1) - x_162 = torch.nn.functional.layer_norm( - x_161, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_161 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) - x_163 = x_162.permute(0, 3, 1, 2) - x_162 = None - x_164 = torch.conv2d( - x_163, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_163 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_18 = torch.nn.functional.relu(x_164, inplace=False) - x_164 = None - pow_19 = relu_18**2 - relu_18 = None - mul_18 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ - * pow_19 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_19 - ) = None - x_165 = ( - mul_18 - + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_18 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_166 = torch.conv2d( - x_165, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_165 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_167 = torch.conv2d( - x_166, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_166 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_168 = x_160 + x_167 - x_160 = x_167 = None - x_169 = x_168.permute(0, 2, 3, 1) - x_170 = torch.nn.functional.layer_norm( - x_169, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_169 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) - x_171 = x_170.permute(0, 3, 1, 2) - x_170 = None - x_172 = torch.conv2d( - x_171, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_171 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_19 = torch.nn.functional.relu(x_172, inplace=False) - x_172 = None - pow_20 = relu_19**2 - relu_19 = None - mul_19 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - * pow_20 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( - pow_20 - ) = None - x_173 = ( - mul_19 - + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - ) - mul_19 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) - x_174 = torch.nn.functional.dropout(x_173, 0.0, False, False) - x_173 = None - x_175 = torch.conv2d( - x_174, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_174 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) - x_175 = None - x_177 = x_168 + x_176 - x_168 = x_176 = None - x_178 = x_177.permute(0, 2, 3, 1) - x_179 = torch.nn.functional.layer_norm( - x_178, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_178 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) - x_180 = x_179.permute(0, 3, 1, 2) - x_179 = None - x_181 = torch.conv2d( - x_180, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_180 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_20 = torch.nn.functional.relu(x_181, inplace=False) - x_181 = None - pow_21 = relu_20**2 - relu_20 = None - mul_20 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ - * pow_21 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_21 - ) = None - x_182 = ( - mul_20 - + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_20 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_183 = torch.conv2d( - x_182, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_182 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_184 = torch.conv2d( - x_183, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_183 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_185 = x_177 + x_184 - x_177 = x_184 = None - x_186 = x_185.permute(0, 2, 3, 1) - x_187 = torch.nn.functional.layer_norm( - x_186, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_186 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) - x_188 = x_187.permute(0, 3, 1, 2) - x_187 = None - x_189 = torch.conv2d( - x_188, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_188 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_21 = torch.nn.functional.relu(x_189, inplace=False) - x_189 = None - pow_22 = relu_21**2 - relu_21 = None - mul_21 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - * pow_22 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( - pow_22 - ) = None - x_190 = ( - mul_21 - + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - ) - mul_21 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) - x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) - x_190 = None - x_192 = torch.conv2d( - x_191, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_191 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) - x_192 = None - x_194 = x_185 + x_193 - x_185 = x_193 = None - x_195 = x_194.permute(0, 2, 3, 1) - x_196 = torch.nn.functional.layer_norm( - x_195, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_195 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) - x_197 = x_196.permute(0, 3, 1, 2) - x_196 = None - x_198 = torch.conv2d( - x_197, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_197 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_22 = torch.nn.functional.relu(x_198, inplace=False) - x_198 = None - pow_23 = relu_22**2 - relu_22 = None - mul_22 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ - * pow_23 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_23 - ) = None - x_199 = ( - mul_22 - + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_22 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_200 = torch.conv2d( - x_199, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_199 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_201 = torch.conv2d( - x_200, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_200 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_202 = x_194 + x_201 - x_194 = x_201 = None - x_203 = x_202.permute(0, 2, 3, 1) - x_204 = torch.nn.functional.layer_norm( - x_203, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_203 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) - x_205 = x_204.permute(0, 3, 1, 2) - x_204 = None - x_206 = torch.conv2d( - x_205, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_205 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_23 = torch.nn.functional.relu(x_206, inplace=False) - x_206 = None - pow_24 = relu_23**2 - relu_23 = None - mul_23 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - * pow_24 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( - pow_24 - ) = None - x_207 = ( - mul_23 - + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - ) - mul_23 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) - x_208 = torch.nn.functional.dropout(x_207, 0.0, False, False) - x_207 = None - x_209 = torch.conv2d( - x_208, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_208 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) - x_209 = None - x_211 = x_202 + x_210 - x_202 = x_210 = None - x_212 = x_211.permute(0, 2, 3, 1) - x_213 = torch.nn.functional.layer_norm( - x_212, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_212 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) - x_214 = x_213.permute(0, 3, 1, 2) - x_213 = None - x_215 = torch.conv2d( - x_214, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_214 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_24 = torch.nn.functional.relu(x_215, inplace=False) - x_215 = None - pow_25 = relu_24**2 - relu_24 = None - mul_24 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ - * pow_25 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_25 - ) = None - x_216 = ( - mul_24 - + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_24 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_217 = torch.conv2d( - x_216, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_216 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_218 = torch.conv2d( - x_217, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_217 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_219 = x_211 + x_218 - x_211 = x_218 = None - x_220 = x_219.permute(0, 2, 3, 1) - x_221 = torch.nn.functional.layer_norm( - x_220, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_220 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) - x_222 = x_221.permute(0, 3, 1, 2) - x_221 = None - x_223 = torch.conv2d( - x_222, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_222 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_25 = torch.nn.functional.relu(x_223, inplace=False) - x_223 = None - pow_26 = relu_25**2 - relu_25 = None - mul_25 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - * pow_26 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( - pow_26 - ) = None - x_224 = ( - mul_25 - + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - ) - mul_25 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) - x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) - x_224 = None - x_226 = torch.conv2d( - x_225, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_225 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) - x_226 = None - x_228 = x_219 + x_227 - x_219 = x_227 = None - x_229 = x_228.permute(0, 2, 3, 1) - x_230 = torch.nn.functional.layer_norm( - x_229, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_229 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) - x_231 = x_230.permute(0, 3, 1, 2) - x_230 = None - x_232 = torch.conv2d( - x_231, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_231 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_26 = torch.nn.functional.relu(x_232, inplace=False) - x_232 = None - pow_27 = relu_26**2 - relu_26 = None - mul_26 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ - * pow_27 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_27 - ) = None - x_233 = ( - mul_26 - + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_26 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_234 = torch.conv2d( - x_233, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_233 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_235 = torch.conv2d( - x_234, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_234 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_236 = x_228 + x_235 - x_228 = x_235 = None - x_237 = x_236.permute(0, 2, 3, 1) - x_238 = torch.nn.functional.layer_norm( - x_237, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_237 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) - x_239 = x_238.permute(0, 3, 1, 2) - x_238 = None - x_240 = torch.conv2d( - x_239, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_239 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_27 = torch.nn.functional.relu(x_240, inplace=False) - x_240 = None - pow_28 = relu_27**2 - relu_27 = None - mul_27 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - * pow_28 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( - pow_28 - ) = None - x_241 = ( - mul_27 - + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - ) - mul_27 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) - x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) - x_241 = None - x_243 = torch.conv2d( - x_242, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_242 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_244 = torch.nn.functional.dropout(x_243, 0.0, False, False) - x_243 = None - x_245 = x_236 + x_244 - x_236 = x_244 = None - x_246 = x_245.permute(0, 2, 3, 1) - x_247 = torch.nn.functional.layer_norm( - x_246, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_246 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) - x_248 = x_247.permute(0, 3, 1, 2) - x_247 = None - x_249 = torch.conv2d( - x_248, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_248 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_28 = torch.nn.functional.relu(x_249, inplace=False) - x_249 = None - pow_29 = relu_28**2 - relu_28 = None - mul_28 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ - * pow_29 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_29 - ) = None - x_250 = ( - mul_28 - + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_28 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_251 = torch.conv2d( - x_250, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_250 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_252 = torch.conv2d( - x_251, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_251 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_253 = x_245 + x_252 - x_245 = x_252 = None - x_254 = x_253.permute(0, 2, 3, 1) - x_255 = torch.nn.functional.layer_norm( - x_254, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_254 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) - x_256 = x_255.permute(0, 3, 1, 2) - x_255 = None - x_257 = torch.conv2d( - x_256, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_256 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_29 = torch.nn.functional.relu(x_257, inplace=False) - x_257 = None - pow_30 = relu_29**2 - relu_29 = None - mul_29 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - * pow_30 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( - pow_30 - ) = None - x_258 = ( - mul_29 - + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - ) - mul_29 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) - x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) - x_258 = None - x_260 = torch.conv2d( - x_259, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_259 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_261 = torch.nn.functional.dropout(x_260, 0.0, False, False) - x_260 = None - x_262 = x_253 + x_261 - x_253 = x_261 = None - x_263 = x_262.permute(0, 2, 3, 1) - x_262 = None - x_264 = torch.nn.functional.layer_norm( - x_263, - (192,), - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_263 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) - x_265 = x_264.permute(0, 3, 1, 2) - x_264 = None - x_266 = torch.conv2d( - x_265, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_265 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_2 = sym_sum // 16 - sym_sum_3 = torch.sym_sum([1, floordiv_2]) - floordiv_2 = None - reshape = x_266.reshape(1, 384, -1) - x_266 = None - x_267 = reshape.transpose(1, 2) - reshape = None - view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_30 = x_267 * view - view = None - layer_norm_33 = torch.nn.functional.layer_norm( - x_267, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_52 = sym_sum_3**2 - linear = torch._C._nn.linear( - layer_norm_33, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_33 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_1 = linear.reshape(1, pow_52, 3, 12, 32) - linear = None - qkv = reshape_1.permute(2, 0, 3, 1, 4) - reshape_1 = None - unbind = qkv.unbind(0) - qkv = None - q = unbind[0] - k = unbind[1] - v = unbind[2] - unbind = None - x_268 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) - q = k = v = None - transpose_1 = x_268.transpose(1, 2) - x_268 = None - x_269 = transpose_1.reshape(1, pow_52, 384) - transpose_1 = None - x_270 = torch._C._nn.linear( - x_269, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_269 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_271 = torch.nn.functional.dropout(x_270, 0.0, False, False) - x_270 = None - x_272 = mul_30 + x_271 - mul_30 = x_271 = None - view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_31 = x_272 * view_1 - view_1 = None - layer_norm_34 = torch.nn.functional.layer_norm( - x_272, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_272 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_273 = torch._C._nn.linear( - layer_norm_34, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_34 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_30 = torch.nn.functional.relu(x_273, inplace=False) - x_273 = None - pow_31 = relu_30**2 - relu_30 = None - mul_32 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_31 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_31 - ) = None - x_274 = ( - mul_32 - + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_275 = torch.nn.functional.dropout(x_274, 0.0, False, False) - x_274 = None - x_276 = torch._C._nn.linear( - x_275, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_277 = torch.nn.functional.dropout(x_276, 0.0, False, False) - x_276 = None - x_278 = mul_31 + x_277 - mul_31 = x_277 = None - view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_33 = x_278 * view_2 - view_2 = None - layer_norm_35 = torch.nn.functional.layer_norm( - x_278, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_4 = torch._C._nn.linear( - layer_norm_35, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_3 = linear_4.reshape(1, pow_52, 3, 12, 32) - linear_4 = None - qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) - reshape_3 = None - unbind_1 = qkv_1.unbind(0) - qkv_1 = None - q_1 = unbind_1[0] - k_1 = unbind_1[1] - v_1 = unbind_1[2] - unbind_1 = None - x_279 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) - q_1 = k_1 = v_1 = None - transpose_2 = x_279.transpose(1, 2) - x_279 = None - x_280 = transpose_2.reshape(1, pow_52, 384) - transpose_2 = None - x_281 = torch._C._nn.linear( - x_280, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_280 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) - x_281 = None - x_283 = mul_33 + x_282 - mul_33 = x_282 = None - view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_34 = x_283 * view_3 - view_3 = None - layer_norm_36 = torch.nn.functional.layer_norm( - x_283, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_283 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_284 = torch._C._nn.linear( - layer_norm_36, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_36 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_31 = torch.nn.functional.relu(x_284, inplace=False) - x_284 = None - pow_32 = relu_31**2 - relu_31 = None - mul_35 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_32 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_32 - ) = None - x_285 = ( - mul_35 - + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_286 = torch.nn.functional.dropout(x_285, 0.0, False, False) - x_285 = None - x_287 = torch._C._nn.linear( - x_286, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) - x_287 = None - x_289 = mul_34 + x_288 - mul_34 = x_288 = None - view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_36 = x_289 * view_4 - view_4 = None - layer_norm_37 = torch.nn.functional.layer_norm( - x_289, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_8 = torch._C._nn.linear( - layer_norm_37, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_37 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_5 = linear_8.reshape(1, pow_52, 3, 12, 32) - linear_8 = None - qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) - reshape_5 = None - unbind_2 = qkv_2.unbind(0) - qkv_2 = None - q_2 = unbind_2[0] - k_2 = unbind_2[1] - v_2 = unbind_2[2] - unbind_2 = None - x_290 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) - q_2 = k_2 = v_2 = None - transpose_3 = x_290.transpose(1, 2) - x_290 = None - x_291 = transpose_3.reshape(1, pow_52, 384) - transpose_3 = None - x_292 = torch._C._nn.linear( - x_291, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_291 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_293 = torch.nn.functional.dropout(x_292, 0.0, False, False) - x_292 = None - x_294 = mul_36 + x_293 - mul_36 = x_293 = None - view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_37 = x_294 * view_5 - view_5 = None - layer_norm_38 = torch.nn.functional.layer_norm( - x_294, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_294 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_295 = torch._C._nn.linear( - layer_norm_38, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_32 = torch.nn.functional.relu(x_295, inplace=False) - x_295 = None - pow_33 = relu_32**2 - relu_32 = None - mul_38 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_33 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_33 - ) = None - x_296 = ( - mul_38 - + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_297 = torch.nn.functional.dropout(x_296, 0.0, False, False) - x_296 = None - x_298 = torch._C._nn.linear( - x_297, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_299 = torch.nn.functional.dropout(x_298, 0.0, False, False) - x_298 = None - x_300 = mul_37 + x_299 - mul_37 = x_299 = None - view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_39 = x_300 * view_6 - view_6 = None - layer_norm_39 = torch.nn.functional.layer_norm( - x_300, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) - linear_12 = torch._C._nn.linear( - layer_norm_39, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_39 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_7 = linear_12.reshape(1, pow_52, 3, 12, 32) - linear_12 = None - qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) - reshape_7 = None - unbind_3 = qkv_3.unbind(0) - qkv_3 = None - q_3 = unbind_3[0] - k_3 = unbind_3[1] - v_3 = unbind_3[2] - unbind_3 = None - x_301 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) - q_3 = k_3 = v_3 = None - transpose_4 = x_301.transpose(1, 2) - x_301 = None - x_302 = transpose_4.reshape(1, pow_52, 384) - transpose_4 = None - x_303 = torch._C._nn.linear( - x_302, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_302 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_304 = torch.nn.functional.dropout(x_303, 0.0, False, False) - x_303 = None - x_305 = mul_39 + x_304 - mul_39 = x_304 = None - view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_40 = x_305 * view_7 - view_7 = None - layer_norm_40 = torch.nn.functional.layer_norm( - x_305, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_305 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) - x_306 = torch._C._nn.linear( - layer_norm_40, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_40 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_33 = torch.nn.functional.relu(x_306, inplace=False) - x_306 = None - pow_34 = relu_33**2 - relu_33 = None - mul_41 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - * pow_34 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( - pow_34 - ) = None - x_307 = ( - mul_41 - + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - ) - mul_41 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) - x_308 = torch.nn.functional.dropout(x_307, 0.0, False, False) - x_307 = None - x_309 = torch._C._nn.linear( - x_308, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) - x_309 = None - x_311 = mul_40 + x_310 - mul_40 = x_310 = None - view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_42 = x_311 * view_8 - view_8 = None - layer_norm_41 = torch.nn.functional.layer_norm( - x_311, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) - linear_16 = torch._C._nn.linear( - layer_norm_41, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_41 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_9 = linear_16.reshape(1, pow_52, 3, 12, 32) - linear_16 = None - qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) - reshape_9 = None - unbind_4 = qkv_4.unbind(0) - qkv_4 = None - q_4 = unbind_4[0] - k_4 = unbind_4[1] - v_4 = unbind_4[2] - unbind_4 = None - x_312 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) - q_4 = k_4 = v_4 = None - transpose_5 = x_312.transpose(1, 2) - x_312 = None - x_313 = transpose_5.reshape(1, pow_52, 384) - transpose_5 = None - x_314 = torch._C._nn.linear( - x_313, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_313 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) - x_314 = None - x_316 = mul_42 + x_315 - mul_42 = x_315 = None - view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_43 = x_316 * view_9 - view_9 = None - layer_norm_42 = torch.nn.functional.layer_norm( - x_316, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_316 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) - x_317 = torch._C._nn.linear( - layer_norm_42, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_42 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_34 = torch.nn.functional.relu(x_317, inplace=False) - x_317 = None - pow_35 = relu_34**2 - relu_34 = None - mul_44 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - * pow_35 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( - pow_35 - ) = None - x_318 = ( - mul_44 - + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - ) - mul_44 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) - x_319 = torch.nn.functional.dropout(x_318, 0.0, False, False) - x_318 = None - x_320 = torch._C._nn.linear( - x_319, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_321 = torch.nn.functional.dropout(x_320, 0.0, False, False) - x_320 = None - x_322 = mul_43 + x_321 - mul_43 = x_321 = None - view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_45 = x_322 * view_10 - view_10 = None - layer_norm_43 = torch.nn.functional.layer_norm( - x_322, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) - linear_20 = torch._C._nn.linear( - layer_norm_43, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_43 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_11 = linear_20.reshape(1, pow_52, 3, 12, 32) - linear_20 = None - qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) - reshape_11 = None - unbind_5 = qkv_5.unbind(0) - qkv_5 = None - q_5 = unbind_5[0] - k_5 = unbind_5[1] - v_5 = unbind_5[2] - unbind_5 = None - x_323 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) - q_5 = k_5 = v_5 = None - transpose_6 = x_323.transpose(1, 2) - x_323 = None - x_324 = transpose_6.reshape(1, pow_52, 384) - transpose_6 = None - x_325 = torch._C._nn.linear( - x_324, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_324 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_326 = torch.nn.functional.dropout(x_325, 0.0, False, False) - x_325 = None - x_327 = mul_45 + x_326 - mul_45 = x_326 = None - view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_46 = x_327 * view_11 - view_11 = None - layer_norm_44 = torch.nn.functional.layer_norm( - x_327, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_327 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) - x_328 = torch._C._nn.linear( - layer_norm_44, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_44 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_35 = torch.nn.functional.relu(x_328, inplace=False) - x_328 = None - pow_36 = relu_35**2 - relu_35 = None - mul_47 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - * pow_36 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( - pow_36 - ) = None - x_329 = ( - mul_47 - + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - ) - mul_47 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) - x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) - x_329 = None - x_331 = torch._C._nn.linear( - x_330, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_332 = torch.nn.functional.dropout(x_331, 0.0, False, False) - x_331 = None - x_333 = mul_46 + x_332 - mul_46 = x_332 = None - view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_48 = x_333 * view_12 - view_12 = None - layer_norm_45 = torch.nn.functional.layer_norm( - x_333, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) - linear_24 = torch._C._nn.linear( - layer_norm_45, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_45 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_13 = linear_24.reshape(1, pow_52, 3, 12, 32) - linear_24 = None - qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) - reshape_13 = None - unbind_6 = qkv_6.unbind(0) - qkv_6 = None - q_6 = unbind_6[0] - k_6 = unbind_6[1] - v_6 = unbind_6[2] - unbind_6 = None - x_334 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) - q_6 = k_6 = v_6 = None - transpose_7 = x_334.transpose(1, 2) - x_334 = None - x_335 = transpose_7.reshape(1, pow_52, 384) - transpose_7 = None - x_336 = torch._C._nn.linear( - x_335, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_335 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_337 = torch.nn.functional.dropout(x_336, 0.0, False, False) - x_336 = None - x_338 = mul_48 + x_337 - mul_48 = x_337 = None - view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_49 = x_338 * view_13 - view_13 = None - layer_norm_46 = torch.nn.functional.layer_norm( - x_338, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_338 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) - x_339 = torch._C._nn.linear( - layer_norm_46, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_46 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_36 = torch.nn.functional.relu(x_339, inplace=False) - x_339 = None - pow_37 = relu_36**2 - relu_36 = None - mul_50 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - * pow_37 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( - pow_37 - ) = None - x_340 = ( - mul_50 - + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - ) - mul_50 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) - x_341 = torch.nn.functional.dropout(x_340, 0.0, False, False) - x_340 = None - x_342 = torch._C._nn.linear( - x_341, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_343 = torch.nn.functional.dropout(x_342, 0.0, False, False) - x_342 = None - x_344 = mul_49 + x_343 - mul_49 = x_343 = None - view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_51 = x_344 * view_14 - view_14 = None - layer_norm_47 = torch.nn.functional.layer_norm( - x_344, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) - linear_28 = torch._C._nn.linear( - layer_norm_47, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_47 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_15 = linear_28.reshape(1, pow_52, 3, 12, 32) - linear_28 = None - qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) - reshape_15 = None - unbind_7 = qkv_7.unbind(0) - qkv_7 = None - q_7 = unbind_7[0] - k_7 = unbind_7[1] - v_7 = unbind_7[2] - unbind_7 = None - x_345 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) - q_7 = k_7 = v_7 = None - transpose_8 = x_345.transpose(1, 2) - x_345 = None - x_346 = transpose_8.reshape(1, pow_52, 384) - transpose_8 = None - x_347 = torch._C._nn.linear( - x_346, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_346 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) - x_347 = None - x_349 = mul_51 + x_348 - mul_51 = x_348 = None - view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_52 = x_349 * view_15 - view_15 = None - layer_norm_48 = torch.nn.functional.layer_norm( - x_349, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_349 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) - x_350 = torch._C._nn.linear( - layer_norm_48, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_48 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_37 = torch.nn.functional.relu(x_350, inplace=False) - x_350 = None - pow_38 = relu_37**2 - relu_37 = None - mul_53 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - * pow_38 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( - pow_38 - ) = None - x_351 = ( - mul_53 - + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - ) - mul_53 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) - x_352 = torch.nn.functional.dropout(x_351, 0.0, False, False) - x_351 = None - x_353 = torch._C._nn.linear( - x_352, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_354 = torch.nn.functional.dropout(x_353, 0.0, False, False) - x_353 = None - x_355 = mul_52 + x_354 - mul_52 = x_354 = None - view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_54 = x_355 * view_16 - view_16 = None - layer_norm_49 = torch.nn.functional.layer_norm( - x_355, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) - linear_32 = torch._C._nn.linear( - layer_norm_49, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_49 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_17 = linear_32.reshape(1, pow_52, 3, 12, 32) - linear_32 = None - qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) - reshape_17 = None - unbind_8 = qkv_8.unbind(0) - qkv_8 = None - q_8 = unbind_8[0] - k_8 = unbind_8[1] - v_8 = unbind_8[2] - unbind_8 = None - x_356 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) - q_8 = k_8 = v_8 = None - transpose_9 = x_356.transpose(1, 2) - x_356 = None - x_357 = transpose_9.reshape(1, pow_52, 384) - transpose_9 = None - x_358 = torch._C._nn.linear( - x_357, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_357 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_359 = torch.nn.functional.dropout(x_358, 0.0, False, False) - x_358 = None - x_360 = mul_54 + x_359 - mul_54 = x_359 = None - view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_55 = x_360 * view_17 - view_17 = None - layer_norm_50 = torch.nn.functional.layer_norm( - x_360, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_360 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) - x_361 = torch._C._nn.linear( - layer_norm_50, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_50 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_38 = torch.nn.functional.relu(x_361, inplace=False) - x_361 = None - pow_39 = relu_38**2 - relu_38 = None - mul_56 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - * pow_39 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( - pow_39 - ) = None - x_362 = ( - mul_56 - + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - ) - mul_56 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) - x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) - x_362 = None - x_364 = torch._C._nn.linear( - x_363, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_365 = torch.nn.functional.dropout(x_364, 0.0, False, False) - x_364 = None - x_366 = mul_55 + x_365 - mul_55 = x_365 = None - view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_57 = x_366 * view_18 - view_18 = None - layer_norm_51 = torch.nn.functional.layer_norm( - x_366, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) - linear_36 = torch._C._nn.linear( - layer_norm_51, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_51 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_19 = linear_36.reshape(1, pow_52, 3, 12, 32) - linear_36 = None - qkv_9 = reshape_19.permute(2, 0, 3, 1, 4) - reshape_19 = None - unbind_9 = qkv_9.unbind(0) - qkv_9 = None - q_9 = unbind_9[0] - k_9 = unbind_9[1] - v_9 = unbind_9[2] - unbind_9 = None - x_367 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) - q_9 = k_9 = v_9 = None - transpose_10 = x_367.transpose(1, 2) - x_367 = None - x_368 = transpose_10.reshape(1, pow_52, 384) - transpose_10 = None - x_369 = torch._C._nn.linear( - x_368, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_368 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_370 = torch.nn.functional.dropout(x_369, 0.0, False, False) - x_369 = None - x_371 = mul_57 + x_370 - mul_57 = x_370 = None - view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_58 = x_371 * view_19 - view_19 = None - layer_norm_52 = torch.nn.functional.layer_norm( - x_371, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_371 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) - x_372 = torch._C._nn.linear( - layer_norm_52, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_52 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_39 = torch.nn.functional.relu(x_372, inplace=False) - x_372 = None - pow_40 = relu_39**2 - relu_39 = None - mul_59 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - * pow_40 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( - pow_40 - ) = None - x_373 = ( - mul_59 - + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - ) - mul_59 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) - x_374 = torch.nn.functional.dropout(x_373, 0.0, False, False) - x_373 = None - x_375 = torch._C._nn.linear( - x_374, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_374 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_376 = torch.nn.functional.dropout(x_375, 0.0, False, False) - x_375 = None - x_377 = mul_58 + x_376 - mul_58 = x_376 = None - view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_60 = x_377 * view_20 - view_20 = None - layer_norm_53 = torch.nn.functional.layer_norm( - x_377, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_377 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) - linear_40 = torch._C._nn.linear( - layer_norm_53, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_53 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_21 = linear_40.reshape(1, pow_52, 3, 12, 32) - linear_40 = None - qkv_10 = reshape_21.permute(2, 0, 3, 1, 4) - reshape_21 = None - unbind_10 = qkv_10.unbind(0) - qkv_10 = None - q_10 = unbind_10[0] - k_10 = unbind_10[1] - v_10 = unbind_10[2] - unbind_10 = None - x_378 = torch._C._nn.scaled_dot_product_attention( - q_10, k_10, v_10, dropout_p=0.0 - ) - q_10 = k_10 = v_10 = None - transpose_11 = x_378.transpose(1, 2) - x_378 = None - x_379 = transpose_11.reshape(1, pow_52, 384) - transpose_11 = None - x_380 = torch._C._nn.linear( - x_379, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_379 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_381 = torch.nn.functional.dropout(x_380, 0.0, False, False) - x_380 = None - x_382 = mul_60 + x_381 - mul_60 = x_381 = None - view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_61 = x_382 * view_21 - view_21 = None - layer_norm_54 = torch.nn.functional.layer_norm( - x_382, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_382 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) - x_383 = torch._C._nn.linear( - layer_norm_54, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_54 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_40 = torch.nn.functional.relu(x_383, inplace=False) - x_383 = None - pow_41 = relu_40**2 - relu_40 = None - mul_62 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - * pow_41 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( - pow_41 - ) = None - x_384 = ( - mul_62 - + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - ) - mul_62 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) - x_385 = torch.nn.functional.dropout(x_384, 0.0, False, False) - x_384 = None - x_386 = torch._C._nn.linear( - x_385, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_385 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_387 = torch.nn.functional.dropout(x_386, 0.0, False, False) - x_386 = None - x_388 = mul_61 + x_387 - mul_61 = x_387 = None - view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_63 = x_388 * view_22 - view_22 = None - layer_norm_55 = torch.nn.functional.layer_norm( - x_388, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_388 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) - linear_44 = torch._C._nn.linear( - layer_norm_55, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_55 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_23 = linear_44.reshape(1, pow_52, 3, 12, 32) - linear_44 = None - qkv_11 = reshape_23.permute(2, 0, 3, 1, 4) - reshape_23 = None - unbind_11 = qkv_11.unbind(0) - qkv_11 = None - q_11 = unbind_11[0] - k_11 = unbind_11[1] - v_11 = unbind_11[2] - unbind_11 = None - x_389 = torch._C._nn.scaled_dot_product_attention( - q_11, k_11, v_11, dropout_p=0.0 - ) - q_11 = k_11 = v_11 = None - transpose_12 = x_389.transpose(1, 2) - x_389 = None - x_390 = transpose_12.reshape(1, pow_52, 384) - transpose_12 = None - x_391 = torch._C._nn.linear( - x_390, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_390 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_392 = torch.nn.functional.dropout(x_391, 0.0, False, False) - x_391 = None - x_393 = mul_63 + x_392 - mul_63 = x_392 = None - view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_64 = x_393 * view_23 - view_23 = None - layer_norm_56 = torch.nn.functional.layer_norm( - x_393, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_393 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) - x_394 = torch._C._nn.linear( - layer_norm_56, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_56 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_41 = torch.nn.functional.relu(x_394, inplace=False) - x_394 = None - pow_42 = relu_41**2 - relu_41 = None - mul_65 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - * pow_42 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( - pow_42 - ) = None - x_395 = ( - mul_65 - + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - ) - mul_65 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) - x_396 = torch.nn.functional.dropout(x_395, 0.0, False, False) - x_395 = None - x_397 = torch._C._nn.linear( - x_396, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_396 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_398 = torch.nn.functional.dropout(x_397, 0.0, False, False) - x_397 = None - x_399 = mul_64 + x_398 - mul_64 = x_398 = None - view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_66 = x_399 * view_24 - view_24 = None - layer_norm_57 = torch.nn.functional.layer_norm( - x_399, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_399 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = (None) - linear_48 = torch._C._nn.linear( - layer_norm_57, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_57 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_25 = linear_48.reshape(1, pow_52, 3, 12, 32) - linear_48 = None - qkv_12 = reshape_25.permute(2, 0, 3, 1, 4) - reshape_25 = None - unbind_12 = qkv_12.unbind(0) - qkv_12 = None - q_12 = unbind_12[0] - k_12 = unbind_12[1] - v_12 = unbind_12[2] - unbind_12 = None - x_400 = torch._C._nn.scaled_dot_product_attention( - q_12, k_12, v_12, dropout_p=0.0 - ) - q_12 = k_12 = v_12 = None - transpose_13 = x_400.transpose(1, 2) - x_400 = None - x_401 = transpose_13.reshape(1, pow_52, 384) - transpose_13 = None - x_402 = torch._C._nn.linear( - x_401, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_401 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_403 = torch.nn.functional.dropout(x_402, 0.0, False, False) - x_402 = None - x_404 = mul_66 + x_403 - mul_66 = x_403 = None - view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_67 = x_404 * view_25 - view_25 = None - layer_norm_58 = torch.nn.functional.layer_norm( - x_404, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_404 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = (None) - x_405 = torch._C._nn.linear( - layer_norm_58, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_58 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_42 = torch.nn.functional.relu(x_405, inplace=False) - x_405 = None - pow_43 = relu_42**2 - relu_42 = None - mul_68 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ - * pow_43 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = ( - pow_43 - ) = None - x_406 = ( - mul_68 - + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ - ) - mul_68 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = (None) - x_407 = torch.nn.functional.dropout(x_406, 0.0, False, False) - x_406 = None - x_408 = torch._C._nn.linear( - x_407, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_407 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_409 = torch.nn.functional.dropout(x_408, 0.0, False, False) - x_408 = None - x_410 = mul_67 + x_409 - mul_67 = x_409 = None - view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_69 = x_410 * view_26 - view_26 = None - layer_norm_59 = torch.nn.functional.layer_norm( - x_410, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_410 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = (None) - linear_52 = torch._C._nn.linear( - layer_norm_59, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_59 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_27 = linear_52.reshape(1, pow_52, 3, 12, 32) - linear_52 = None - qkv_13 = reshape_27.permute(2, 0, 3, 1, 4) - reshape_27 = None - unbind_13 = qkv_13.unbind(0) - qkv_13 = None - q_13 = unbind_13[0] - k_13 = unbind_13[1] - v_13 = unbind_13[2] - unbind_13 = None - x_411 = torch._C._nn.scaled_dot_product_attention( - q_13, k_13, v_13, dropout_p=0.0 - ) - q_13 = k_13 = v_13 = None - transpose_14 = x_411.transpose(1, 2) - x_411 = None - x_412 = transpose_14.reshape(1, pow_52, 384) - transpose_14 = None - x_413 = torch._C._nn.linear( - x_412, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_412 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_414 = torch.nn.functional.dropout(x_413, 0.0, False, False) - x_413 = None - x_415 = mul_69 + x_414 - mul_69 = x_414 = None - view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_70 = x_415 * view_27 - view_27 = None - layer_norm_60 = torch.nn.functional.layer_norm( - x_415, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_415 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = (None) - x_416 = torch._C._nn.linear( - layer_norm_60, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_60 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_43 = torch.nn.functional.relu(x_416, inplace=False) - x_416 = None - pow_44 = relu_43**2 - relu_43 = None - mul_71 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ - * pow_44 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = ( - pow_44 - ) = None - x_417 = ( - mul_71 - + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ - ) - mul_71 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = (None) - x_418 = torch.nn.functional.dropout(x_417, 0.0, False, False) - x_417 = None - x_419 = torch._C._nn.linear( - x_418, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_418 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_420 = torch.nn.functional.dropout(x_419, 0.0, False, False) - x_419 = None - x_421 = mul_70 + x_420 - mul_70 = x_420 = None - view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_72 = x_421 * view_28 - view_28 = None - layer_norm_61 = torch.nn.functional.layer_norm( - x_421, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_421 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = (None) - linear_56 = torch._C._nn.linear( - layer_norm_61, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_61 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_29 = linear_56.reshape(1, pow_52, 3, 12, 32) - linear_56 = None - qkv_14 = reshape_29.permute(2, 0, 3, 1, 4) - reshape_29 = None - unbind_14 = qkv_14.unbind(0) - qkv_14 = None - q_14 = unbind_14[0] - k_14 = unbind_14[1] - v_14 = unbind_14[2] - unbind_14 = None - x_422 = torch._C._nn.scaled_dot_product_attention( - q_14, k_14, v_14, dropout_p=0.0 - ) - q_14 = k_14 = v_14 = None - transpose_15 = x_422.transpose(1, 2) - x_422 = None - x_423 = transpose_15.reshape(1, pow_52, 384) - transpose_15 = None - x_424 = torch._C._nn.linear( - x_423, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_423 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_425 = torch.nn.functional.dropout(x_424, 0.0, False, False) - x_424 = None - x_426 = mul_72 + x_425 - mul_72 = x_425 = None - view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_73 = x_426 * view_29 - view_29 = None - layer_norm_62 = torch.nn.functional.layer_norm( - x_426, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_426 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = (None) - x_427 = torch._C._nn.linear( - layer_norm_62, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_62 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_44 = torch.nn.functional.relu(x_427, inplace=False) - x_427 = None - pow_45 = relu_44**2 - relu_44 = None - mul_74 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ - * pow_45 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = ( - pow_45 - ) = None - x_428 = ( - mul_74 - + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ - ) - mul_74 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = (None) - x_429 = torch.nn.functional.dropout(x_428, 0.0, False, False) - x_428 = None - x_430 = torch._C._nn.linear( - x_429, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_429 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_431 = torch.nn.functional.dropout(x_430, 0.0, False, False) - x_430 = None - x_432 = mul_73 + x_431 - mul_73 = x_431 = None - view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_75 = x_432 * view_30 - view_30 = None - layer_norm_63 = torch.nn.functional.layer_norm( - x_432, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_432 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = (None) - linear_60 = torch._C._nn.linear( - layer_norm_63, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_63 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_31 = linear_60.reshape(1, pow_52, 3, 12, 32) - linear_60 = None - qkv_15 = reshape_31.permute(2, 0, 3, 1, 4) - reshape_31 = None - unbind_15 = qkv_15.unbind(0) - qkv_15 = None - q_15 = unbind_15[0] - k_15 = unbind_15[1] - v_15 = unbind_15[2] - unbind_15 = None - x_433 = torch._C._nn.scaled_dot_product_attention( - q_15, k_15, v_15, dropout_p=0.0 - ) - q_15 = k_15 = v_15 = None - transpose_16 = x_433.transpose(1, 2) - x_433 = None - x_434 = transpose_16.reshape(1, pow_52, 384) - transpose_16 = None - x_435 = torch._C._nn.linear( - x_434, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_434 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_436 = torch.nn.functional.dropout(x_435, 0.0, False, False) - x_435 = None - x_437 = mul_75 + x_436 - mul_75 = x_436 = None - view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_76 = x_437 * view_31 - view_31 = None - layer_norm_64 = torch.nn.functional.layer_norm( - x_437, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_437 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = (None) - x_438 = torch._C._nn.linear( - layer_norm_64, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_64 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_45 = torch.nn.functional.relu(x_438, inplace=False) - x_438 = None - pow_46 = relu_45**2 - relu_45 = None - mul_77 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ - * pow_46 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = ( - pow_46 - ) = None - x_439 = ( - mul_77 - + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ - ) - mul_77 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = (None) - x_440 = torch.nn.functional.dropout(x_439, 0.0, False, False) - x_439 = None - x_441 = torch._C._nn.linear( - x_440, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_440 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_442 = torch.nn.functional.dropout(x_441, 0.0, False, False) - x_441 = None - x_443 = mul_76 + x_442 - mul_76 = x_442 = None - view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_78 = x_443 * view_32 - view_32 = None - layer_norm_65 = torch.nn.functional.layer_norm( - x_443, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_443 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = (None) - linear_64 = torch._C._nn.linear( - layer_norm_65, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_65 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_33 = linear_64.reshape(1, pow_52, 3, 12, 32) - linear_64 = None - qkv_16 = reshape_33.permute(2, 0, 3, 1, 4) - reshape_33 = None - unbind_16 = qkv_16.unbind(0) - qkv_16 = None - q_16 = unbind_16[0] - k_16 = unbind_16[1] - v_16 = unbind_16[2] - unbind_16 = None - x_444 = torch._C._nn.scaled_dot_product_attention( - q_16, k_16, v_16, dropout_p=0.0 - ) - q_16 = k_16 = v_16 = None - transpose_17 = x_444.transpose(1, 2) - x_444 = None - x_445 = transpose_17.reshape(1, pow_52, 384) - transpose_17 = None - x_446 = torch._C._nn.linear( - x_445, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_445 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_447 = torch.nn.functional.dropout(x_446, 0.0, False, False) - x_446 = None - x_448 = mul_78 + x_447 - mul_78 = x_447 = None - view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_79 = x_448 * view_33 - view_33 = None - layer_norm_66 = torch.nn.functional.layer_norm( - x_448, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_448 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = (None) - x_449 = torch._C._nn.linear( - layer_norm_66, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_66 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_46 = torch.nn.functional.relu(x_449, inplace=False) - x_449 = None - pow_47 = relu_46**2 - relu_46 = None - mul_80 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ - * pow_47 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = ( - pow_47 - ) = None - x_450 = ( - mul_80 - + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ - ) - mul_80 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = (None) - x_451 = torch.nn.functional.dropout(x_450, 0.0, False, False) - x_450 = None - x_452 = torch._C._nn.linear( - x_451, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_451 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_453 = torch.nn.functional.dropout(x_452, 0.0, False, False) - x_452 = None - x_454 = mul_79 + x_453 - mul_79 = x_453 = None - view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_81 = x_454 * view_34 - view_34 = None - layer_norm_67 = torch.nn.functional.layer_norm( - x_454, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_454 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = (None) - linear_68 = torch._C._nn.linear( - layer_norm_67, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_67 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_35 = linear_68.reshape(1, pow_52, 3, 12, 32) - linear_68 = None - qkv_17 = reshape_35.permute(2, 0, 3, 1, 4) - reshape_35 = None - unbind_17 = qkv_17.unbind(0) - qkv_17 = None - q_17 = unbind_17[0] - k_17 = unbind_17[1] - v_17 = unbind_17[2] - unbind_17 = None - x_455 = torch._C._nn.scaled_dot_product_attention( - q_17, k_17, v_17, dropout_p=0.0 - ) - q_17 = k_17 = v_17 = None - transpose_18 = x_455.transpose(1, 2) - x_455 = None - x_456 = transpose_18.reshape(1, pow_52, 384) - transpose_18 = pow_52 = None - x_457 = torch._C._nn.linear( - x_456, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_456 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_458 = torch.nn.functional.dropout(x_457, 0.0, False, False) - x_457 = None - x_459 = mul_81 + x_458 - mul_81 = x_458 = None - view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_82 = x_459 * view_35 - view_35 = None - layer_norm_68 = torch.nn.functional.layer_norm( - x_459, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_459 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = (None) - x_460 = torch._C._nn.linear( - layer_norm_68, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_68 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_47 = torch.nn.functional.relu(x_460, inplace=False) - x_460 = None - pow_48 = relu_47**2 - relu_47 = None - mul_83 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ - * pow_48 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = ( - pow_48 - ) = None - x_461 = ( - mul_83 - + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ - ) - mul_83 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = (None) - x_462 = torch.nn.functional.dropout(x_461, 0.0, False, False) - x_461 = None - x_463 = torch._C._nn.linear( - x_462, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_462 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_464 = torch.nn.functional.dropout(x_463, 0.0, False, False) - x_463 = None - x_465 = mul_82 + x_464 - mul_82 = x_464 = None - transpose_19 = x_465.transpose(1, 2) - x_465 = None - x_466 = transpose_19.reshape(1, 384, sym_sum_3, sym_sum_3) - transpose_19 = sym_sum_3 = None - x_467 = x_466.permute(0, 2, 3, 1) - x_466 = None - x_468 = torch.nn.functional.layer_norm( - x_467, - (384,), - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_467 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) - x_469 = x_468.permute(0, 3, 1, 2) - x_468 = None - x_470 = torch.conv2d( - x_469, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_469 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_3 = sym_sum // 32 - sym_sum = None - sym_sum_4 = torch.sym_sum([1, floordiv_3]) - floordiv_3 = None - reshape_38 = x_470.reshape(1, 576, -1) - x_470 = None - x_471 = reshape_38.transpose(1, 2) - reshape_38 = None - view_36 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_84 = x_471 * view_36 - view_36 = None - layer_norm_70 = torch.nn.functional.layer_norm( - x_471, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_471 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_53 = sym_sum_4**2 - linear_72 = torch._C._nn.linear( - layer_norm_70, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_70 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_39 = linear_72.reshape(1, pow_53, 3, 18, 32) - linear_72 = None - qkv_18 = reshape_39.permute(2, 0, 3, 1, 4) - reshape_39 = None - unbind_18 = qkv_18.unbind(0) - qkv_18 = None - q_18 = unbind_18[0] - k_18 = unbind_18[1] - v_18 = unbind_18[2] - unbind_18 = None - x_472 = torch._C._nn.scaled_dot_product_attention( - q_18, k_18, v_18, dropout_p=0.0 - ) - q_18 = k_18 = v_18 = None - transpose_21 = x_472.transpose(1, 2) - x_472 = None - x_473 = transpose_21.reshape(1, pow_53, 576) - transpose_21 = None - x_474 = torch._C._nn.linear( - x_473, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_473 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_475 = torch.nn.functional.dropout(x_474, 0.0, False, False) - x_474 = None - x_476 = mul_84 + x_475 - mul_84 = x_475 = None - view_37 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_85 = x_476 * view_37 - view_37 = None - layer_norm_71 = torch.nn.functional.layer_norm( - x_476, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_476 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_477 = torch._C._nn.linear( - layer_norm_71, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_71 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_48 = torch.nn.functional.relu(x_477, inplace=False) - x_477 = None - pow_49 = relu_48**2 - relu_48 = None - mul_86 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_49 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_49 - ) = None - x_478 = ( - mul_86 - + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_86 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_479 = torch.nn.functional.dropout(x_478, 0.0, False, False) - x_478 = None - x_480 = torch._C._nn.linear( - x_479, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_479 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_481 = torch.nn.functional.dropout(x_480, 0.0, False, False) - x_480 = None - x_482 = mul_85 + x_481 - mul_85 = x_481 = None - view_38 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_87 = x_482 * view_38 - view_38 = None - layer_norm_72 = torch.nn.functional.layer_norm( - x_482, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_482 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_76 = torch._C._nn.linear( - layer_norm_72, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_72 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_41 = linear_76.reshape(1, pow_53, 3, 18, 32) - linear_76 = None - qkv_19 = reshape_41.permute(2, 0, 3, 1, 4) - reshape_41 = None - unbind_19 = qkv_19.unbind(0) - qkv_19 = None - q_19 = unbind_19[0] - k_19 = unbind_19[1] - v_19 = unbind_19[2] - unbind_19 = None - x_483 = torch._C._nn.scaled_dot_product_attention( - q_19, k_19, v_19, dropout_p=0.0 - ) - q_19 = k_19 = v_19 = None - transpose_22 = x_483.transpose(1, 2) - x_483 = None - x_484 = transpose_22.reshape(1, pow_53, 576) - transpose_22 = None - x_485 = torch._C._nn.linear( - x_484, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_484 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_486 = torch.nn.functional.dropout(x_485, 0.0, False, False) - x_485 = None - x_487 = mul_87 + x_486 - mul_87 = x_486 = None - view_39 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_88 = x_487 * view_39 - view_39 = None - layer_norm_73 = torch.nn.functional.layer_norm( - x_487, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_487 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_488 = torch._C._nn.linear( - layer_norm_73, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_73 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_49 = torch.nn.functional.relu(x_488, inplace=False) - x_488 = None - pow_50 = relu_49**2 - relu_49 = None - mul_89 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_50 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_50 - ) = None - x_489 = ( - mul_89 - + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_89 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_490 = torch.nn.functional.dropout(x_489, 0.0, False, False) - x_489 = None - x_491 = torch._C._nn.linear( - x_490, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_490 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_492 = torch.nn.functional.dropout(x_491, 0.0, False, False) - x_491 = None - x_493 = mul_88 + x_492 - mul_88 = x_492 = None - view_40 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_90 = x_493 * view_40 - view_40 = None - layer_norm_74 = torch.nn.functional.layer_norm( - x_493, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_493 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_80 = torch._C._nn.linear( - layer_norm_74, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_74 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_43 = linear_80.reshape(1, pow_53, 3, 18, 32) - linear_80 = None - qkv_20 = reshape_43.permute(2, 0, 3, 1, 4) - reshape_43 = None - unbind_20 = qkv_20.unbind(0) - qkv_20 = None - q_20 = unbind_20[0] - k_20 = unbind_20[1] - v_20 = unbind_20[2] - unbind_20 = None - x_494 = torch._C._nn.scaled_dot_product_attention( - q_20, k_20, v_20, dropout_p=0.0 - ) - q_20 = k_20 = v_20 = None - transpose_23 = x_494.transpose(1, 2) - x_494 = None - x_495 = transpose_23.reshape(1, pow_53, 576) - transpose_23 = pow_53 = None - x_496 = torch._C._nn.linear( - x_495, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_495 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_497 = torch.nn.functional.dropout(x_496, 0.0, False, False) - x_496 = None - x_498 = mul_90 + x_497 - mul_90 = x_497 = None - view_41 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_91 = x_498 * view_41 - view_41 = None - layer_norm_75 = torch.nn.functional.layer_norm( - x_498, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_498 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_499 = torch._C._nn.linear( - layer_norm_75, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_75 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_50 = torch.nn.functional.relu(x_499, inplace=False) - x_499 = None - pow_51 = relu_50**2 - relu_50 = None - mul_92 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_51 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_51 - ) = None - x_500 = ( - mul_92 - + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_92 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_501 = torch.nn.functional.dropout(x_500, 0.0, False, False) - x_500 = None - x_502 = torch._C._nn.linear( - x_501, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_501 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_503 = torch.nn.functional.dropout(x_502, 0.0, False, False) - x_502 = None - x_504 = mul_91 + x_503 - mul_91 = x_503 = None - transpose_24 = x_504.transpose(1, 2) - x_504 = None - x_505 = transpose_24.reshape(1, 576, sym_sum_4, sym_sum_4) - transpose_24 = sym_sum_4 = None - x_506 = torch.nn.functional.adaptive_avg_pool2d(x_505, 1) - x_505 = None - x_507 = x_506.permute(0, 2, 3, 1) - x_506 = None - x_508 = torch.nn.functional.layer_norm( - x_507, - (576,), - l_self_modules_head_modules_norm_parameters_weight_, - l_self_modules_head_modules_norm_parameters_bias_, - 1e-06, - ) - x_507 = ( - l_self_modules_head_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_norm_parameters_bias_ = None - x_509 = x_508.permute(0, 3, 1, 2) - x_508 = None - x_510 = x_509.flatten(1, -1) - x_509 = None - x_511 = torch.nn.functional.dropout(x_510, 0.0, False, False) - x_510 = None - x_512 = torch._C._nn.linear( - x_511, - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, - ) - x_511 = ( - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None - relu_51 = torch.nn.functional.relu(x_512, inplace=False) - x_512 = None - x_513 = torch.square(relu_51) - relu_51 = None - x_514 = torch.nn.functional.layer_norm( - x_513, - (2304,), - l_self_modules_head_modules_fc_modules_norm_parameters_weight_, - l_self_modules_head_modules_fc_modules_norm_parameters_bias_, - 1e-06, - ) - x_513 = ( - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None - x_515 = torch.nn.functional.dropout(x_514, 0.0, False, False) - x_514 = None - x_516 = torch._C._nn.linear( - x_515, - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, - ) - x_515 = ( - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None - return (x_516,) diff --git a/samples/timm/caformer_m36.sail_in1k_384/weight_meta.py b/samples/timm/caformer_m36.sail_in1k_384/weight_meta.py deleted file mode 100644 index 1a948dec5d..0000000000 --- a/samples/timm/caformer_m36.sail_in1k_384/weight_meta.py +++ /dev/null @@ -1,3968 +0,0 @@ -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: - name = "L_self_modules_stem_modules_conv_parameters_weight_" - shape = [96, 3, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: - name = "L_self_modules_stem_modules_conv_parameters_bias_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_s1: - name = "s1" - shape = [] - dtype = "torch.int64" - device = "cpu" - mean = None - std = None - data = [4] - - -class Program_weight_tensor_meta_L_x_: - name = "L_x_" - shape = [1, 3, 224, 224] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.229 - std = 1.285 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: - name = "L_self_modules_stem_modules_norm_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [192, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [192, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [96, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [384, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [96, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [192, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [192, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [96, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [384, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [96, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [192, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [192, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [96, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [384, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [96, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" - shape = [192, 96, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.021 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" - shape = [384, 192, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" - shape = [576, 384, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1728, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [576, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [2304, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [576, 2304] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1728, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [576, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [2304, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [576, 2304] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1728, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [576, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [2304, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [576, 2304] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_norm_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_norm_parameters_bias_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" - shape = [2304, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" - shape = [2304] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" - shape = [2304] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" - shape = [2304] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" - shape = [1000, 2304] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" - shape = [1000] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None diff --git a/samples/timm/caformer_m36.sail_in22k/graph_net.json b/samples/timm/caformer_m36.sail_in22k/graph_net.json deleted file mode 100644 index 1373fe3b58..0000000000 --- a/samples/timm/caformer_m36.sail_in22k/graph_net.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "framework": "torch", - "num_devices_required": 1, - "num_nodes_required": 1 -} \ No newline at end of file diff --git a/samples/timm/caformer_m36.sail_in22k/input_meta.py b/samples/timm/caformer_m36.sail_in22k/input_meta.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_m36.sail_in22k/input_tensor_constraints.py b/samples/timm/caformer_m36.sail_in22k/input_tensor_constraints.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_m36.sail_in22k/model.py b/samples/timm/caformer_m36.sail_in22k/model.py deleted file mode 100644 index dcce286c5c..0000000000 --- a/samples/timm/caformer_m36.sail_in22k/model.py +++ /dev/null @@ -1,4754 +0,0 @@ -import torch - - -class GraphModule(torch.nn.Module): - def forward( - self, - L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - s1: torch.SymInt, - L_x_: torch.Tensor, - L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - ): - l_self_modules_stem_modules_conv_parameters_weight_ = ( - L_self_modules_stem_modules_conv_parameters_weight_ - ) - l_self_modules_stem_modules_conv_parameters_bias_ = ( - L_self_modules_stem_modules_conv_parameters_bias_ - ) - l_x_ = L_x_ - l_self_modules_stem_modules_norm_parameters_weight_ = ( - L_self_modules_stem_modules_norm_parameters_weight_ - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_head_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ - ) - x = torch.conv2d( - l_x_, - l_self_modules_stem_modules_conv_parameters_weight_, - l_self_modules_stem_modules_conv_parameters_bias_, - (4, 4), - (2, 2), - (1, 1), - 1, - ) - l_x_ = ( - l_self_modules_stem_modules_conv_parameters_weight_ - ) = l_self_modules_stem_modules_conv_parameters_bias_ = None - x_1 = x.permute(0, 2, 3, 1) - x = None - x_2 = torch.nn.functional.layer_norm( - x_1, (96,), l_self_modules_stem_modules_norm_parameters_weight_, None, 1e-06 - ) - x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None - x_3 = x_2.permute(0, 3, 1, 2) - x_2 = None - sym_sum = torch.sym_sum([-3, s1]) - s1 = None - floordiv = sym_sum // 4 - sym_sum_1 = torch.sym_sum([1, floordiv]) - floordiv = sym_sum_1 = None - x_4 = x_3.permute(0, 2, 3, 1) - x_5 = torch.nn.functional.layer_norm( - x_4, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_6 = x_5.permute(0, 3, 1, 2) - x_5 = None - x_7 = torch.conv2d( - x_6, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu = torch.nn.functional.relu(x_7, inplace=False) - x_7 = None - pow_1 = relu**2 - relu = None - mul = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_1 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_1 - ) = None - x_8 = ( - mul - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_9 = torch.conv2d( - x_8, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 192, - ) - x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_10 = torch.conv2d( - x_9, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_11 = x_3 + x_10 - x_3 = x_10 = None - x_12 = x_11.permute(0, 2, 3, 1) - x_13 = torch.nn.functional.layer_norm( - x_12, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_14 = x_13.permute(0, 3, 1, 2) - x_13 = None - x_15 = torch.conv2d( - x_14, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_1 = torch.nn.functional.relu(x_15, inplace=False) - x_15 = None - pow_2 = relu_1**2 - relu_1 = None - mul_1 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_2 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_2 - ) = None - x_16 = ( - mul_1 - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) - x_16 = None - x_18 = torch.conv2d( - x_17, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) - x_18 = None - x_20 = x_11 + x_19 - x_11 = x_19 = None - x_21 = x_20.permute(0, 2, 3, 1) - x_22 = torch.nn.functional.layer_norm( - x_21, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_23 = x_22.permute(0, 3, 1, 2) - x_22 = None - x_24 = torch.conv2d( - x_23, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_2 = torch.nn.functional.relu(x_24, inplace=False) - x_24 = None - pow_3 = relu_2**2 - relu_2 = None - mul_2 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_3 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_3 - ) = None - x_25 = ( - mul_2 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_26 = torch.conv2d( - x_25, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 192, - ) - x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_27 = torch.conv2d( - x_26, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_28 = x_20 + x_27 - x_20 = x_27 = None - x_29 = x_28.permute(0, 2, 3, 1) - x_30 = torch.nn.functional.layer_norm( - x_29, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_31 = x_30.permute(0, 3, 1, 2) - x_30 = None - x_32 = torch.conv2d( - x_31, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_3 = torch.nn.functional.relu(x_32, inplace=False) - x_32 = None - pow_4 = relu_3**2 - relu_3 = None - mul_3 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_4 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_4 - ) = None - x_33 = ( - mul_3 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) - x_33 = None - x_35 = torch.conv2d( - x_34, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) - x_35 = None - x_37 = x_28 + x_36 - x_28 = x_36 = None - x_38 = x_37.permute(0, 2, 3, 1) - x_39 = torch.nn.functional.layer_norm( - x_38, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_40 = x_39.permute(0, 3, 1, 2) - x_39 = None - x_41 = torch.conv2d( - x_40, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_4 = torch.nn.functional.relu(x_41, inplace=False) - x_41 = None - pow_5 = relu_4**2 - relu_4 = None - mul_4 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_5 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_5 - ) = None - x_42 = ( - mul_4 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_43 = torch.conv2d( - x_42, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 192, - ) - x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_44 = torch.conv2d( - x_43, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_45 = x_37 + x_44 - x_37 = x_44 = None - x_46 = x_45.permute(0, 2, 3, 1) - x_47 = torch.nn.functional.layer_norm( - x_46, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_48 = x_47.permute(0, 3, 1, 2) - x_47 = None - x_49 = torch.conv2d( - x_48, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_5 = torch.nn.functional.relu(x_49, inplace=False) - x_49 = None - pow_6 = relu_5**2 - relu_5 = None - mul_5 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_6 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_6 - ) = None - x_50 = ( - mul_5 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) - x_50 = None - x_52 = torch.conv2d( - x_51, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) - x_52 = None - x_54 = x_45 + x_53 - x_45 = x_53 = None - x_55 = x_54.permute(0, 2, 3, 1) - x_54 = None - x_56 = torch.nn.functional.layer_norm( - x_55, - (96,), - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) - x_57 = x_56.permute(0, 3, 1, 2) - x_56 = None - x_58 = torch.conv2d( - x_57, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_1 = sym_sum // 8 - sym_sum_2 = torch.sym_sum([1, floordiv_1]) - floordiv_1 = sym_sum_2 = None - x_59 = x_58.permute(0, 2, 3, 1) - x_60 = torch.nn.functional.layer_norm( - x_59, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_61 = x_60.permute(0, 3, 1, 2) - x_60 = None - x_62 = torch.conv2d( - x_61, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_6 = torch.nn.functional.relu(x_62, inplace=False) - x_62 = None - pow_7 = relu_6**2 - relu_6 = None - mul_6 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_7 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_7 - ) = None - x_63 = ( - mul_6 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_64 = torch.conv2d( - x_63, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_65 = torch.conv2d( - x_64, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_66 = x_58 + x_65 - x_58 = x_65 = None - x_67 = x_66.permute(0, 2, 3, 1) - x_68 = torch.nn.functional.layer_norm( - x_67, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_69 = x_68.permute(0, 3, 1, 2) - x_68 = None - x_70 = torch.conv2d( - x_69, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_7 = torch.nn.functional.relu(x_70, inplace=False) - x_70 = None - pow_8 = relu_7**2 - relu_7 = None - mul_7 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_8 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_8 - ) = None - x_71 = ( - mul_7 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) - x_71 = None - x_73 = torch.conv2d( - x_72, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) - x_73 = None - x_75 = x_66 + x_74 - x_66 = x_74 = None - x_76 = x_75.permute(0, 2, 3, 1) - x_77 = torch.nn.functional.layer_norm( - x_76, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_78 = x_77.permute(0, 3, 1, 2) - x_77 = None - x_79 = torch.conv2d( - x_78, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_8 = torch.nn.functional.relu(x_79, inplace=False) - x_79 = None - pow_9 = relu_8**2 - relu_8 = None - mul_8 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_9 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_9 - ) = None - x_80 = ( - mul_8 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_81 = torch.conv2d( - x_80, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_82 = torch.conv2d( - x_81, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_83 = x_75 + x_82 - x_75 = x_82 = None - x_84 = x_83.permute(0, 2, 3, 1) - x_85 = torch.nn.functional.layer_norm( - x_84, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_86 = x_85.permute(0, 3, 1, 2) - x_85 = None - x_87 = torch.conv2d( - x_86, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_9 = torch.nn.functional.relu(x_87, inplace=False) - x_87 = None - pow_10 = relu_9**2 - relu_9 = None - mul_9 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_10 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_10 - ) = None - x_88 = ( - mul_9 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) - x_88 = None - x_90 = torch.conv2d( - x_89, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) - x_90 = None - x_92 = x_83 + x_91 - x_83 = x_91 = None - x_93 = x_92.permute(0, 2, 3, 1) - x_94 = torch.nn.functional.layer_norm( - x_93, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_95 = x_94.permute(0, 3, 1, 2) - x_94 = None - x_96 = torch.conv2d( - x_95, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_10 = torch.nn.functional.relu(x_96, inplace=False) - x_96 = None - pow_11 = relu_10**2 - relu_10 = None - mul_10 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_11 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_11 - ) = None - x_97 = ( - mul_10 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_98 = torch.conv2d( - x_97, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_99 = torch.conv2d( - x_98, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_100 = x_92 + x_99 - x_92 = x_99 = None - x_101 = x_100.permute(0, 2, 3, 1) - x_102 = torch.nn.functional.layer_norm( - x_101, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_103 = x_102.permute(0, 3, 1, 2) - x_102 = None - x_104 = torch.conv2d( - x_103, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_11 = torch.nn.functional.relu(x_104, inplace=False) - x_104 = None - pow_12 = relu_11**2 - relu_11 = None - mul_11 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_12 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_12 - ) = None - x_105 = ( - mul_11 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) - x_105 = None - x_107 = torch.conv2d( - x_106, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) - x_107 = None - x_109 = x_100 + x_108 - x_100 = x_108 = None - x_110 = x_109.permute(0, 2, 3, 1) - x_111 = torch.nn.functional.layer_norm( - x_110, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_110 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) - x_112 = x_111.permute(0, 3, 1, 2) - x_111 = None - x_113 = torch.conv2d( - x_112, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_112 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_12 = torch.nn.functional.relu(x_113, inplace=False) - x_113 = None - pow_13 = relu_12**2 - relu_12 = None - mul_12 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ - * pow_13 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_13 - ) = None - x_114 = ( - mul_12 - + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_12 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_115 = torch.conv2d( - x_114, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_114 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_116 = torch.conv2d( - x_115, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_115 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_117 = x_109 + x_116 - x_109 = x_116 = None - x_118 = x_117.permute(0, 2, 3, 1) - x_119 = torch.nn.functional.layer_norm( - x_118, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_118 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) - x_120 = x_119.permute(0, 3, 1, 2) - x_119 = None - x_121 = torch.conv2d( - x_120, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_120 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_13 = torch.nn.functional.relu(x_121, inplace=False) - x_121 = None - pow_14 = relu_13**2 - relu_13 = None - mul_13 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - * pow_14 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( - pow_14 - ) = None - x_122 = ( - mul_13 - + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - ) - mul_13 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) - x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) - x_122 = None - x_124 = torch.conv2d( - x_123, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_123 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) - x_124 = None - x_126 = x_117 + x_125 - x_117 = x_125 = None - x_127 = x_126.permute(0, 2, 3, 1) - x_128 = torch.nn.functional.layer_norm( - x_127, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_127 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) - x_129 = x_128.permute(0, 3, 1, 2) - x_128 = None - x_130 = torch.conv2d( - x_129, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_129 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_14 = torch.nn.functional.relu(x_130, inplace=False) - x_130 = None - pow_15 = relu_14**2 - relu_14 = None - mul_14 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ - * pow_15 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_15 - ) = None - x_131 = ( - mul_14 - + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_14 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_132 = torch.conv2d( - x_131, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_131 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_133 = torch.conv2d( - x_132, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_132 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_134 = x_126 + x_133 - x_126 = x_133 = None - x_135 = x_134.permute(0, 2, 3, 1) - x_136 = torch.nn.functional.layer_norm( - x_135, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_135 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) - x_137 = x_136.permute(0, 3, 1, 2) - x_136 = None - x_138 = torch.conv2d( - x_137, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_137 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_15 = torch.nn.functional.relu(x_138, inplace=False) - x_138 = None - pow_16 = relu_15**2 - relu_15 = None - mul_15 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - * pow_16 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( - pow_16 - ) = None - x_139 = ( - mul_15 - + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - ) - mul_15 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) - x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) - x_139 = None - x_141 = torch.conv2d( - x_140, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_140 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) - x_141 = None - x_143 = x_134 + x_142 - x_134 = x_142 = None - x_144 = x_143.permute(0, 2, 3, 1) - x_145 = torch.nn.functional.layer_norm( - x_144, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_144 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) - x_146 = x_145.permute(0, 3, 1, 2) - x_145 = None - x_147 = torch.conv2d( - x_146, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_146 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_16 = torch.nn.functional.relu(x_147, inplace=False) - x_147 = None - pow_17 = relu_16**2 - relu_16 = None - mul_16 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ - * pow_17 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_17 - ) = None - x_148 = ( - mul_16 - + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_16 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_149 = torch.conv2d( - x_148, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_148 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_150 = torch.conv2d( - x_149, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_149 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_151 = x_143 + x_150 - x_143 = x_150 = None - x_152 = x_151.permute(0, 2, 3, 1) - x_153 = torch.nn.functional.layer_norm( - x_152, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_152 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) - x_154 = x_153.permute(0, 3, 1, 2) - x_153 = None - x_155 = torch.conv2d( - x_154, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_154 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_17 = torch.nn.functional.relu(x_155, inplace=False) - x_155 = None - pow_18 = relu_17**2 - relu_17 = None - mul_17 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - * pow_18 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( - pow_18 - ) = None - x_156 = ( - mul_17 - + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - ) - mul_17 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) - x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) - x_156 = None - x_158 = torch.conv2d( - x_157, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_157 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) - x_158 = None - x_160 = x_151 + x_159 - x_151 = x_159 = None - x_161 = x_160.permute(0, 2, 3, 1) - x_162 = torch.nn.functional.layer_norm( - x_161, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_161 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) - x_163 = x_162.permute(0, 3, 1, 2) - x_162 = None - x_164 = torch.conv2d( - x_163, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_163 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_18 = torch.nn.functional.relu(x_164, inplace=False) - x_164 = None - pow_19 = relu_18**2 - relu_18 = None - mul_18 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ - * pow_19 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_19 - ) = None - x_165 = ( - mul_18 - + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_18 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_166 = torch.conv2d( - x_165, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_165 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_167 = torch.conv2d( - x_166, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_166 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_168 = x_160 + x_167 - x_160 = x_167 = None - x_169 = x_168.permute(0, 2, 3, 1) - x_170 = torch.nn.functional.layer_norm( - x_169, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_169 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) - x_171 = x_170.permute(0, 3, 1, 2) - x_170 = None - x_172 = torch.conv2d( - x_171, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_171 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_19 = torch.nn.functional.relu(x_172, inplace=False) - x_172 = None - pow_20 = relu_19**2 - relu_19 = None - mul_19 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - * pow_20 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( - pow_20 - ) = None - x_173 = ( - mul_19 - + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - ) - mul_19 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) - x_174 = torch.nn.functional.dropout(x_173, 0.0, False, False) - x_173 = None - x_175 = torch.conv2d( - x_174, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_174 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) - x_175 = None - x_177 = x_168 + x_176 - x_168 = x_176 = None - x_178 = x_177.permute(0, 2, 3, 1) - x_179 = torch.nn.functional.layer_norm( - x_178, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_178 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) - x_180 = x_179.permute(0, 3, 1, 2) - x_179 = None - x_181 = torch.conv2d( - x_180, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_180 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_20 = torch.nn.functional.relu(x_181, inplace=False) - x_181 = None - pow_21 = relu_20**2 - relu_20 = None - mul_20 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ - * pow_21 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_21 - ) = None - x_182 = ( - mul_20 - + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_20 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_183 = torch.conv2d( - x_182, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_182 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_184 = torch.conv2d( - x_183, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_183 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_185 = x_177 + x_184 - x_177 = x_184 = None - x_186 = x_185.permute(0, 2, 3, 1) - x_187 = torch.nn.functional.layer_norm( - x_186, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_186 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) - x_188 = x_187.permute(0, 3, 1, 2) - x_187 = None - x_189 = torch.conv2d( - x_188, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_188 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_21 = torch.nn.functional.relu(x_189, inplace=False) - x_189 = None - pow_22 = relu_21**2 - relu_21 = None - mul_21 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - * pow_22 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( - pow_22 - ) = None - x_190 = ( - mul_21 - + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - ) - mul_21 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) - x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) - x_190 = None - x_192 = torch.conv2d( - x_191, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_191 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) - x_192 = None - x_194 = x_185 + x_193 - x_185 = x_193 = None - x_195 = x_194.permute(0, 2, 3, 1) - x_196 = torch.nn.functional.layer_norm( - x_195, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_195 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) - x_197 = x_196.permute(0, 3, 1, 2) - x_196 = None - x_198 = torch.conv2d( - x_197, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_197 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_22 = torch.nn.functional.relu(x_198, inplace=False) - x_198 = None - pow_23 = relu_22**2 - relu_22 = None - mul_22 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ - * pow_23 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_23 - ) = None - x_199 = ( - mul_22 - + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_22 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_200 = torch.conv2d( - x_199, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_199 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_201 = torch.conv2d( - x_200, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_200 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_202 = x_194 + x_201 - x_194 = x_201 = None - x_203 = x_202.permute(0, 2, 3, 1) - x_204 = torch.nn.functional.layer_norm( - x_203, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_203 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) - x_205 = x_204.permute(0, 3, 1, 2) - x_204 = None - x_206 = torch.conv2d( - x_205, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_205 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_23 = torch.nn.functional.relu(x_206, inplace=False) - x_206 = None - pow_24 = relu_23**2 - relu_23 = None - mul_23 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - * pow_24 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( - pow_24 - ) = None - x_207 = ( - mul_23 - + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - ) - mul_23 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) - x_208 = torch.nn.functional.dropout(x_207, 0.0, False, False) - x_207 = None - x_209 = torch.conv2d( - x_208, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_208 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) - x_209 = None - x_211 = x_202 + x_210 - x_202 = x_210 = None - x_212 = x_211.permute(0, 2, 3, 1) - x_213 = torch.nn.functional.layer_norm( - x_212, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_212 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) - x_214 = x_213.permute(0, 3, 1, 2) - x_213 = None - x_215 = torch.conv2d( - x_214, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_214 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_24 = torch.nn.functional.relu(x_215, inplace=False) - x_215 = None - pow_25 = relu_24**2 - relu_24 = None - mul_24 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ - * pow_25 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_25 - ) = None - x_216 = ( - mul_24 - + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_24 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_217 = torch.conv2d( - x_216, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_216 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_218 = torch.conv2d( - x_217, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_217 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_219 = x_211 + x_218 - x_211 = x_218 = None - x_220 = x_219.permute(0, 2, 3, 1) - x_221 = torch.nn.functional.layer_norm( - x_220, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_220 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) - x_222 = x_221.permute(0, 3, 1, 2) - x_221 = None - x_223 = torch.conv2d( - x_222, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_222 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_25 = torch.nn.functional.relu(x_223, inplace=False) - x_223 = None - pow_26 = relu_25**2 - relu_25 = None - mul_25 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - * pow_26 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( - pow_26 - ) = None - x_224 = ( - mul_25 - + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - ) - mul_25 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) - x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) - x_224 = None - x_226 = torch.conv2d( - x_225, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_225 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) - x_226 = None - x_228 = x_219 + x_227 - x_219 = x_227 = None - x_229 = x_228.permute(0, 2, 3, 1) - x_230 = torch.nn.functional.layer_norm( - x_229, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_229 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) - x_231 = x_230.permute(0, 3, 1, 2) - x_230 = None - x_232 = torch.conv2d( - x_231, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_231 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_26 = torch.nn.functional.relu(x_232, inplace=False) - x_232 = None - pow_27 = relu_26**2 - relu_26 = None - mul_26 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ - * pow_27 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_27 - ) = None - x_233 = ( - mul_26 - + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_26 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_234 = torch.conv2d( - x_233, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_233 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_235 = torch.conv2d( - x_234, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_234 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_236 = x_228 + x_235 - x_228 = x_235 = None - x_237 = x_236.permute(0, 2, 3, 1) - x_238 = torch.nn.functional.layer_norm( - x_237, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_237 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) - x_239 = x_238.permute(0, 3, 1, 2) - x_238 = None - x_240 = torch.conv2d( - x_239, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_239 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_27 = torch.nn.functional.relu(x_240, inplace=False) - x_240 = None - pow_28 = relu_27**2 - relu_27 = None - mul_27 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - * pow_28 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( - pow_28 - ) = None - x_241 = ( - mul_27 - + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - ) - mul_27 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) - x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) - x_241 = None - x_243 = torch.conv2d( - x_242, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_242 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_244 = torch.nn.functional.dropout(x_243, 0.0, False, False) - x_243 = None - x_245 = x_236 + x_244 - x_236 = x_244 = None - x_246 = x_245.permute(0, 2, 3, 1) - x_247 = torch.nn.functional.layer_norm( - x_246, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_246 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) - x_248 = x_247.permute(0, 3, 1, 2) - x_247 = None - x_249 = torch.conv2d( - x_248, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_248 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_28 = torch.nn.functional.relu(x_249, inplace=False) - x_249 = None - pow_29 = relu_28**2 - relu_28 = None - mul_28 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ - * pow_29 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_29 - ) = None - x_250 = ( - mul_28 - + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_28 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_251 = torch.conv2d( - x_250, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_250 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_252 = torch.conv2d( - x_251, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_251 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_253 = x_245 + x_252 - x_245 = x_252 = None - x_254 = x_253.permute(0, 2, 3, 1) - x_255 = torch.nn.functional.layer_norm( - x_254, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_254 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) - x_256 = x_255.permute(0, 3, 1, 2) - x_255 = None - x_257 = torch.conv2d( - x_256, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_256 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_29 = torch.nn.functional.relu(x_257, inplace=False) - x_257 = None - pow_30 = relu_29**2 - relu_29 = None - mul_29 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - * pow_30 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( - pow_30 - ) = None - x_258 = ( - mul_29 - + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - ) - mul_29 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) - x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) - x_258 = None - x_260 = torch.conv2d( - x_259, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_259 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_261 = torch.nn.functional.dropout(x_260, 0.0, False, False) - x_260 = None - x_262 = x_253 + x_261 - x_253 = x_261 = None - x_263 = x_262.permute(0, 2, 3, 1) - x_262 = None - x_264 = torch.nn.functional.layer_norm( - x_263, - (192,), - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_263 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) - x_265 = x_264.permute(0, 3, 1, 2) - x_264 = None - x_266 = torch.conv2d( - x_265, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_265 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_2 = sym_sum // 16 - sym_sum_3 = torch.sym_sum([1, floordiv_2]) - floordiv_2 = None - reshape = x_266.reshape(1, 384, -1) - x_266 = None - x_267 = reshape.transpose(1, 2) - reshape = None - view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_30 = x_267 * view - view = None - layer_norm_33 = torch.nn.functional.layer_norm( - x_267, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_52 = sym_sum_3**2 - linear = torch._C._nn.linear( - layer_norm_33, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_33 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_1 = linear.reshape(1, pow_52, 3, 12, 32) - linear = None - qkv = reshape_1.permute(2, 0, 3, 1, 4) - reshape_1 = None - unbind = qkv.unbind(0) - qkv = None - q = unbind[0] - k = unbind[1] - v = unbind[2] - unbind = None - x_268 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) - q = k = v = None - transpose_1 = x_268.transpose(1, 2) - x_268 = None - x_269 = transpose_1.reshape(1, pow_52, 384) - transpose_1 = None - x_270 = torch._C._nn.linear( - x_269, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_269 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_271 = torch.nn.functional.dropout(x_270, 0.0, False, False) - x_270 = None - x_272 = mul_30 + x_271 - mul_30 = x_271 = None - view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_31 = x_272 * view_1 - view_1 = None - layer_norm_34 = torch.nn.functional.layer_norm( - x_272, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_272 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_273 = torch._C._nn.linear( - layer_norm_34, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_34 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_30 = torch.nn.functional.relu(x_273, inplace=False) - x_273 = None - pow_31 = relu_30**2 - relu_30 = None - mul_32 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_31 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_31 - ) = None - x_274 = ( - mul_32 - + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_275 = torch.nn.functional.dropout(x_274, 0.0, False, False) - x_274 = None - x_276 = torch._C._nn.linear( - x_275, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_277 = torch.nn.functional.dropout(x_276, 0.0, False, False) - x_276 = None - x_278 = mul_31 + x_277 - mul_31 = x_277 = None - view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_33 = x_278 * view_2 - view_2 = None - layer_norm_35 = torch.nn.functional.layer_norm( - x_278, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_4 = torch._C._nn.linear( - layer_norm_35, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_3 = linear_4.reshape(1, pow_52, 3, 12, 32) - linear_4 = None - qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) - reshape_3 = None - unbind_1 = qkv_1.unbind(0) - qkv_1 = None - q_1 = unbind_1[0] - k_1 = unbind_1[1] - v_1 = unbind_1[2] - unbind_1 = None - x_279 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) - q_1 = k_1 = v_1 = None - transpose_2 = x_279.transpose(1, 2) - x_279 = None - x_280 = transpose_2.reshape(1, pow_52, 384) - transpose_2 = None - x_281 = torch._C._nn.linear( - x_280, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_280 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) - x_281 = None - x_283 = mul_33 + x_282 - mul_33 = x_282 = None - view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_34 = x_283 * view_3 - view_3 = None - layer_norm_36 = torch.nn.functional.layer_norm( - x_283, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_283 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_284 = torch._C._nn.linear( - layer_norm_36, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_36 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_31 = torch.nn.functional.relu(x_284, inplace=False) - x_284 = None - pow_32 = relu_31**2 - relu_31 = None - mul_35 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_32 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_32 - ) = None - x_285 = ( - mul_35 - + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_286 = torch.nn.functional.dropout(x_285, 0.0, False, False) - x_285 = None - x_287 = torch._C._nn.linear( - x_286, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) - x_287 = None - x_289 = mul_34 + x_288 - mul_34 = x_288 = None - view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_36 = x_289 * view_4 - view_4 = None - layer_norm_37 = torch.nn.functional.layer_norm( - x_289, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_8 = torch._C._nn.linear( - layer_norm_37, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_37 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_5 = linear_8.reshape(1, pow_52, 3, 12, 32) - linear_8 = None - qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) - reshape_5 = None - unbind_2 = qkv_2.unbind(0) - qkv_2 = None - q_2 = unbind_2[0] - k_2 = unbind_2[1] - v_2 = unbind_2[2] - unbind_2 = None - x_290 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) - q_2 = k_2 = v_2 = None - transpose_3 = x_290.transpose(1, 2) - x_290 = None - x_291 = transpose_3.reshape(1, pow_52, 384) - transpose_3 = None - x_292 = torch._C._nn.linear( - x_291, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_291 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_293 = torch.nn.functional.dropout(x_292, 0.0, False, False) - x_292 = None - x_294 = mul_36 + x_293 - mul_36 = x_293 = None - view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_37 = x_294 * view_5 - view_5 = None - layer_norm_38 = torch.nn.functional.layer_norm( - x_294, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_294 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_295 = torch._C._nn.linear( - layer_norm_38, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_32 = torch.nn.functional.relu(x_295, inplace=False) - x_295 = None - pow_33 = relu_32**2 - relu_32 = None - mul_38 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_33 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_33 - ) = None - x_296 = ( - mul_38 - + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_297 = torch.nn.functional.dropout(x_296, 0.0, False, False) - x_296 = None - x_298 = torch._C._nn.linear( - x_297, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_299 = torch.nn.functional.dropout(x_298, 0.0, False, False) - x_298 = None - x_300 = mul_37 + x_299 - mul_37 = x_299 = None - view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_39 = x_300 * view_6 - view_6 = None - layer_norm_39 = torch.nn.functional.layer_norm( - x_300, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) - linear_12 = torch._C._nn.linear( - layer_norm_39, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_39 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_7 = linear_12.reshape(1, pow_52, 3, 12, 32) - linear_12 = None - qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) - reshape_7 = None - unbind_3 = qkv_3.unbind(0) - qkv_3 = None - q_3 = unbind_3[0] - k_3 = unbind_3[1] - v_3 = unbind_3[2] - unbind_3 = None - x_301 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) - q_3 = k_3 = v_3 = None - transpose_4 = x_301.transpose(1, 2) - x_301 = None - x_302 = transpose_4.reshape(1, pow_52, 384) - transpose_4 = None - x_303 = torch._C._nn.linear( - x_302, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_302 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_304 = torch.nn.functional.dropout(x_303, 0.0, False, False) - x_303 = None - x_305 = mul_39 + x_304 - mul_39 = x_304 = None - view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_40 = x_305 * view_7 - view_7 = None - layer_norm_40 = torch.nn.functional.layer_norm( - x_305, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_305 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) - x_306 = torch._C._nn.linear( - layer_norm_40, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_40 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_33 = torch.nn.functional.relu(x_306, inplace=False) - x_306 = None - pow_34 = relu_33**2 - relu_33 = None - mul_41 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - * pow_34 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( - pow_34 - ) = None - x_307 = ( - mul_41 - + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - ) - mul_41 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) - x_308 = torch.nn.functional.dropout(x_307, 0.0, False, False) - x_307 = None - x_309 = torch._C._nn.linear( - x_308, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) - x_309 = None - x_311 = mul_40 + x_310 - mul_40 = x_310 = None - view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_42 = x_311 * view_8 - view_8 = None - layer_norm_41 = torch.nn.functional.layer_norm( - x_311, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) - linear_16 = torch._C._nn.linear( - layer_norm_41, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_41 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_9 = linear_16.reshape(1, pow_52, 3, 12, 32) - linear_16 = None - qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) - reshape_9 = None - unbind_4 = qkv_4.unbind(0) - qkv_4 = None - q_4 = unbind_4[0] - k_4 = unbind_4[1] - v_4 = unbind_4[2] - unbind_4 = None - x_312 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) - q_4 = k_4 = v_4 = None - transpose_5 = x_312.transpose(1, 2) - x_312 = None - x_313 = transpose_5.reshape(1, pow_52, 384) - transpose_5 = None - x_314 = torch._C._nn.linear( - x_313, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_313 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) - x_314 = None - x_316 = mul_42 + x_315 - mul_42 = x_315 = None - view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_43 = x_316 * view_9 - view_9 = None - layer_norm_42 = torch.nn.functional.layer_norm( - x_316, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_316 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) - x_317 = torch._C._nn.linear( - layer_norm_42, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_42 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_34 = torch.nn.functional.relu(x_317, inplace=False) - x_317 = None - pow_35 = relu_34**2 - relu_34 = None - mul_44 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - * pow_35 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( - pow_35 - ) = None - x_318 = ( - mul_44 - + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - ) - mul_44 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) - x_319 = torch.nn.functional.dropout(x_318, 0.0, False, False) - x_318 = None - x_320 = torch._C._nn.linear( - x_319, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_321 = torch.nn.functional.dropout(x_320, 0.0, False, False) - x_320 = None - x_322 = mul_43 + x_321 - mul_43 = x_321 = None - view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_45 = x_322 * view_10 - view_10 = None - layer_norm_43 = torch.nn.functional.layer_norm( - x_322, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) - linear_20 = torch._C._nn.linear( - layer_norm_43, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_43 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_11 = linear_20.reshape(1, pow_52, 3, 12, 32) - linear_20 = None - qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) - reshape_11 = None - unbind_5 = qkv_5.unbind(0) - qkv_5 = None - q_5 = unbind_5[0] - k_5 = unbind_5[1] - v_5 = unbind_5[2] - unbind_5 = None - x_323 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) - q_5 = k_5 = v_5 = None - transpose_6 = x_323.transpose(1, 2) - x_323 = None - x_324 = transpose_6.reshape(1, pow_52, 384) - transpose_6 = None - x_325 = torch._C._nn.linear( - x_324, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_324 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_326 = torch.nn.functional.dropout(x_325, 0.0, False, False) - x_325 = None - x_327 = mul_45 + x_326 - mul_45 = x_326 = None - view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_46 = x_327 * view_11 - view_11 = None - layer_norm_44 = torch.nn.functional.layer_norm( - x_327, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_327 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) - x_328 = torch._C._nn.linear( - layer_norm_44, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_44 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_35 = torch.nn.functional.relu(x_328, inplace=False) - x_328 = None - pow_36 = relu_35**2 - relu_35 = None - mul_47 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - * pow_36 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( - pow_36 - ) = None - x_329 = ( - mul_47 - + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - ) - mul_47 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) - x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) - x_329 = None - x_331 = torch._C._nn.linear( - x_330, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_332 = torch.nn.functional.dropout(x_331, 0.0, False, False) - x_331 = None - x_333 = mul_46 + x_332 - mul_46 = x_332 = None - view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_48 = x_333 * view_12 - view_12 = None - layer_norm_45 = torch.nn.functional.layer_norm( - x_333, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) - linear_24 = torch._C._nn.linear( - layer_norm_45, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_45 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_13 = linear_24.reshape(1, pow_52, 3, 12, 32) - linear_24 = None - qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) - reshape_13 = None - unbind_6 = qkv_6.unbind(0) - qkv_6 = None - q_6 = unbind_6[0] - k_6 = unbind_6[1] - v_6 = unbind_6[2] - unbind_6 = None - x_334 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) - q_6 = k_6 = v_6 = None - transpose_7 = x_334.transpose(1, 2) - x_334 = None - x_335 = transpose_7.reshape(1, pow_52, 384) - transpose_7 = None - x_336 = torch._C._nn.linear( - x_335, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_335 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_337 = torch.nn.functional.dropout(x_336, 0.0, False, False) - x_336 = None - x_338 = mul_48 + x_337 - mul_48 = x_337 = None - view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_49 = x_338 * view_13 - view_13 = None - layer_norm_46 = torch.nn.functional.layer_norm( - x_338, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_338 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) - x_339 = torch._C._nn.linear( - layer_norm_46, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_46 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_36 = torch.nn.functional.relu(x_339, inplace=False) - x_339 = None - pow_37 = relu_36**2 - relu_36 = None - mul_50 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - * pow_37 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( - pow_37 - ) = None - x_340 = ( - mul_50 - + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - ) - mul_50 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) - x_341 = torch.nn.functional.dropout(x_340, 0.0, False, False) - x_340 = None - x_342 = torch._C._nn.linear( - x_341, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_343 = torch.nn.functional.dropout(x_342, 0.0, False, False) - x_342 = None - x_344 = mul_49 + x_343 - mul_49 = x_343 = None - view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_51 = x_344 * view_14 - view_14 = None - layer_norm_47 = torch.nn.functional.layer_norm( - x_344, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) - linear_28 = torch._C._nn.linear( - layer_norm_47, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_47 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_15 = linear_28.reshape(1, pow_52, 3, 12, 32) - linear_28 = None - qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) - reshape_15 = None - unbind_7 = qkv_7.unbind(0) - qkv_7 = None - q_7 = unbind_7[0] - k_7 = unbind_7[1] - v_7 = unbind_7[2] - unbind_7 = None - x_345 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) - q_7 = k_7 = v_7 = None - transpose_8 = x_345.transpose(1, 2) - x_345 = None - x_346 = transpose_8.reshape(1, pow_52, 384) - transpose_8 = None - x_347 = torch._C._nn.linear( - x_346, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_346 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) - x_347 = None - x_349 = mul_51 + x_348 - mul_51 = x_348 = None - view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_52 = x_349 * view_15 - view_15 = None - layer_norm_48 = torch.nn.functional.layer_norm( - x_349, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_349 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) - x_350 = torch._C._nn.linear( - layer_norm_48, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_48 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_37 = torch.nn.functional.relu(x_350, inplace=False) - x_350 = None - pow_38 = relu_37**2 - relu_37 = None - mul_53 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - * pow_38 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( - pow_38 - ) = None - x_351 = ( - mul_53 - + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - ) - mul_53 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) - x_352 = torch.nn.functional.dropout(x_351, 0.0, False, False) - x_351 = None - x_353 = torch._C._nn.linear( - x_352, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_354 = torch.nn.functional.dropout(x_353, 0.0, False, False) - x_353 = None - x_355 = mul_52 + x_354 - mul_52 = x_354 = None - view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_54 = x_355 * view_16 - view_16 = None - layer_norm_49 = torch.nn.functional.layer_norm( - x_355, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) - linear_32 = torch._C._nn.linear( - layer_norm_49, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_49 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_17 = linear_32.reshape(1, pow_52, 3, 12, 32) - linear_32 = None - qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) - reshape_17 = None - unbind_8 = qkv_8.unbind(0) - qkv_8 = None - q_8 = unbind_8[0] - k_8 = unbind_8[1] - v_8 = unbind_8[2] - unbind_8 = None - x_356 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) - q_8 = k_8 = v_8 = None - transpose_9 = x_356.transpose(1, 2) - x_356 = None - x_357 = transpose_9.reshape(1, pow_52, 384) - transpose_9 = None - x_358 = torch._C._nn.linear( - x_357, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_357 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_359 = torch.nn.functional.dropout(x_358, 0.0, False, False) - x_358 = None - x_360 = mul_54 + x_359 - mul_54 = x_359 = None - view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_55 = x_360 * view_17 - view_17 = None - layer_norm_50 = torch.nn.functional.layer_norm( - x_360, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_360 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) - x_361 = torch._C._nn.linear( - layer_norm_50, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_50 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_38 = torch.nn.functional.relu(x_361, inplace=False) - x_361 = None - pow_39 = relu_38**2 - relu_38 = None - mul_56 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - * pow_39 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( - pow_39 - ) = None - x_362 = ( - mul_56 - + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - ) - mul_56 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) - x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) - x_362 = None - x_364 = torch._C._nn.linear( - x_363, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_365 = torch.nn.functional.dropout(x_364, 0.0, False, False) - x_364 = None - x_366 = mul_55 + x_365 - mul_55 = x_365 = None - view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_57 = x_366 * view_18 - view_18 = None - layer_norm_51 = torch.nn.functional.layer_norm( - x_366, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) - linear_36 = torch._C._nn.linear( - layer_norm_51, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_51 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_19 = linear_36.reshape(1, pow_52, 3, 12, 32) - linear_36 = None - qkv_9 = reshape_19.permute(2, 0, 3, 1, 4) - reshape_19 = None - unbind_9 = qkv_9.unbind(0) - qkv_9 = None - q_9 = unbind_9[0] - k_9 = unbind_9[1] - v_9 = unbind_9[2] - unbind_9 = None - x_367 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) - q_9 = k_9 = v_9 = None - transpose_10 = x_367.transpose(1, 2) - x_367 = None - x_368 = transpose_10.reshape(1, pow_52, 384) - transpose_10 = None - x_369 = torch._C._nn.linear( - x_368, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_368 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_370 = torch.nn.functional.dropout(x_369, 0.0, False, False) - x_369 = None - x_371 = mul_57 + x_370 - mul_57 = x_370 = None - view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_58 = x_371 * view_19 - view_19 = None - layer_norm_52 = torch.nn.functional.layer_norm( - x_371, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_371 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) - x_372 = torch._C._nn.linear( - layer_norm_52, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_52 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_39 = torch.nn.functional.relu(x_372, inplace=False) - x_372 = None - pow_40 = relu_39**2 - relu_39 = None - mul_59 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - * pow_40 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( - pow_40 - ) = None - x_373 = ( - mul_59 - + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - ) - mul_59 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) - x_374 = torch.nn.functional.dropout(x_373, 0.0, False, False) - x_373 = None - x_375 = torch._C._nn.linear( - x_374, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_374 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_376 = torch.nn.functional.dropout(x_375, 0.0, False, False) - x_375 = None - x_377 = mul_58 + x_376 - mul_58 = x_376 = None - view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_60 = x_377 * view_20 - view_20 = None - layer_norm_53 = torch.nn.functional.layer_norm( - x_377, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_377 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) - linear_40 = torch._C._nn.linear( - layer_norm_53, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_53 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_21 = linear_40.reshape(1, pow_52, 3, 12, 32) - linear_40 = None - qkv_10 = reshape_21.permute(2, 0, 3, 1, 4) - reshape_21 = None - unbind_10 = qkv_10.unbind(0) - qkv_10 = None - q_10 = unbind_10[0] - k_10 = unbind_10[1] - v_10 = unbind_10[2] - unbind_10 = None - x_378 = torch._C._nn.scaled_dot_product_attention( - q_10, k_10, v_10, dropout_p=0.0 - ) - q_10 = k_10 = v_10 = None - transpose_11 = x_378.transpose(1, 2) - x_378 = None - x_379 = transpose_11.reshape(1, pow_52, 384) - transpose_11 = None - x_380 = torch._C._nn.linear( - x_379, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_379 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_381 = torch.nn.functional.dropout(x_380, 0.0, False, False) - x_380 = None - x_382 = mul_60 + x_381 - mul_60 = x_381 = None - view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_61 = x_382 * view_21 - view_21 = None - layer_norm_54 = torch.nn.functional.layer_norm( - x_382, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_382 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) - x_383 = torch._C._nn.linear( - layer_norm_54, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_54 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_40 = torch.nn.functional.relu(x_383, inplace=False) - x_383 = None - pow_41 = relu_40**2 - relu_40 = None - mul_62 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - * pow_41 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( - pow_41 - ) = None - x_384 = ( - mul_62 - + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - ) - mul_62 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) - x_385 = torch.nn.functional.dropout(x_384, 0.0, False, False) - x_384 = None - x_386 = torch._C._nn.linear( - x_385, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_385 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_387 = torch.nn.functional.dropout(x_386, 0.0, False, False) - x_386 = None - x_388 = mul_61 + x_387 - mul_61 = x_387 = None - view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_63 = x_388 * view_22 - view_22 = None - layer_norm_55 = torch.nn.functional.layer_norm( - x_388, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_388 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) - linear_44 = torch._C._nn.linear( - layer_norm_55, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_55 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_23 = linear_44.reshape(1, pow_52, 3, 12, 32) - linear_44 = None - qkv_11 = reshape_23.permute(2, 0, 3, 1, 4) - reshape_23 = None - unbind_11 = qkv_11.unbind(0) - qkv_11 = None - q_11 = unbind_11[0] - k_11 = unbind_11[1] - v_11 = unbind_11[2] - unbind_11 = None - x_389 = torch._C._nn.scaled_dot_product_attention( - q_11, k_11, v_11, dropout_p=0.0 - ) - q_11 = k_11 = v_11 = None - transpose_12 = x_389.transpose(1, 2) - x_389 = None - x_390 = transpose_12.reshape(1, pow_52, 384) - transpose_12 = None - x_391 = torch._C._nn.linear( - x_390, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_390 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_392 = torch.nn.functional.dropout(x_391, 0.0, False, False) - x_391 = None - x_393 = mul_63 + x_392 - mul_63 = x_392 = None - view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_64 = x_393 * view_23 - view_23 = None - layer_norm_56 = torch.nn.functional.layer_norm( - x_393, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_393 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) - x_394 = torch._C._nn.linear( - layer_norm_56, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_56 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_41 = torch.nn.functional.relu(x_394, inplace=False) - x_394 = None - pow_42 = relu_41**2 - relu_41 = None - mul_65 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - * pow_42 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( - pow_42 - ) = None - x_395 = ( - mul_65 - + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - ) - mul_65 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) - x_396 = torch.nn.functional.dropout(x_395, 0.0, False, False) - x_395 = None - x_397 = torch._C._nn.linear( - x_396, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_396 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_398 = torch.nn.functional.dropout(x_397, 0.0, False, False) - x_397 = None - x_399 = mul_64 + x_398 - mul_64 = x_398 = None - view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_66 = x_399 * view_24 - view_24 = None - layer_norm_57 = torch.nn.functional.layer_norm( - x_399, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_399 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = (None) - linear_48 = torch._C._nn.linear( - layer_norm_57, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_57 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_25 = linear_48.reshape(1, pow_52, 3, 12, 32) - linear_48 = None - qkv_12 = reshape_25.permute(2, 0, 3, 1, 4) - reshape_25 = None - unbind_12 = qkv_12.unbind(0) - qkv_12 = None - q_12 = unbind_12[0] - k_12 = unbind_12[1] - v_12 = unbind_12[2] - unbind_12 = None - x_400 = torch._C._nn.scaled_dot_product_attention( - q_12, k_12, v_12, dropout_p=0.0 - ) - q_12 = k_12 = v_12 = None - transpose_13 = x_400.transpose(1, 2) - x_400 = None - x_401 = transpose_13.reshape(1, pow_52, 384) - transpose_13 = None - x_402 = torch._C._nn.linear( - x_401, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_401 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_403 = torch.nn.functional.dropout(x_402, 0.0, False, False) - x_402 = None - x_404 = mul_66 + x_403 - mul_66 = x_403 = None - view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_67 = x_404 * view_25 - view_25 = None - layer_norm_58 = torch.nn.functional.layer_norm( - x_404, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_404 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = (None) - x_405 = torch._C._nn.linear( - layer_norm_58, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_58 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_42 = torch.nn.functional.relu(x_405, inplace=False) - x_405 = None - pow_43 = relu_42**2 - relu_42 = None - mul_68 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ - * pow_43 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = ( - pow_43 - ) = None - x_406 = ( - mul_68 - + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ - ) - mul_68 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = (None) - x_407 = torch.nn.functional.dropout(x_406, 0.0, False, False) - x_406 = None - x_408 = torch._C._nn.linear( - x_407, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_407 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_409 = torch.nn.functional.dropout(x_408, 0.0, False, False) - x_408 = None - x_410 = mul_67 + x_409 - mul_67 = x_409 = None - view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_69 = x_410 * view_26 - view_26 = None - layer_norm_59 = torch.nn.functional.layer_norm( - x_410, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_410 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = (None) - linear_52 = torch._C._nn.linear( - layer_norm_59, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_59 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_27 = linear_52.reshape(1, pow_52, 3, 12, 32) - linear_52 = None - qkv_13 = reshape_27.permute(2, 0, 3, 1, 4) - reshape_27 = None - unbind_13 = qkv_13.unbind(0) - qkv_13 = None - q_13 = unbind_13[0] - k_13 = unbind_13[1] - v_13 = unbind_13[2] - unbind_13 = None - x_411 = torch._C._nn.scaled_dot_product_attention( - q_13, k_13, v_13, dropout_p=0.0 - ) - q_13 = k_13 = v_13 = None - transpose_14 = x_411.transpose(1, 2) - x_411 = None - x_412 = transpose_14.reshape(1, pow_52, 384) - transpose_14 = None - x_413 = torch._C._nn.linear( - x_412, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_412 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_414 = torch.nn.functional.dropout(x_413, 0.0, False, False) - x_413 = None - x_415 = mul_69 + x_414 - mul_69 = x_414 = None - view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_70 = x_415 * view_27 - view_27 = None - layer_norm_60 = torch.nn.functional.layer_norm( - x_415, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_415 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = (None) - x_416 = torch._C._nn.linear( - layer_norm_60, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_60 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_43 = torch.nn.functional.relu(x_416, inplace=False) - x_416 = None - pow_44 = relu_43**2 - relu_43 = None - mul_71 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ - * pow_44 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = ( - pow_44 - ) = None - x_417 = ( - mul_71 - + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ - ) - mul_71 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = (None) - x_418 = torch.nn.functional.dropout(x_417, 0.0, False, False) - x_417 = None - x_419 = torch._C._nn.linear( - x_418, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_418 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_420 = torch.nn.functional.dropout(x_419, 0.0, False, False) - x_419 = None - x_421 = mul_70 + x_420 - mul_70 = x_420 = None - view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_72 = x_421 * view_28 - view_28 = None - layer_norm_61 = torch.nn.functional.layer_norm( - x_421, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_421 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = (None) - linear_56 = torch._C._nn.linear( - layer_norm_61, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_61 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_29 = linear_56.reshape(1, pow_52, 3, 12, 32) - linear_56 = None - qkv_14 = reshape_29.permute(2, 0, 3, 1, 4) - reshape_29 = None - unbind_14 = qkv_14.unbind(0) - qkv_14 = None - q_14 = unbind_14[0] - k_14 = unbind_14[1] - v_14 = unbind_14[2] - unbind_14 = None - x_422 = torch._C._nn.scaled_dot_product_attention( - q_14, k_14, v_14, dropout_p=0.0 - ) - q_14 = k_14 = v_14 = None - transpose_15 = x_422.transpose(1, 2) - x_422 = None - x_423 = transpose_15.reshape(1, pow_52, 384) - transpose_15 = None - x_424 = torch._C._nn.linear( - x_423, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_423 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_425 = torch.nn.functional.dropout(x_424, 0.0, False, False) - x_424 = None - x_426 = mul_72 + x_425 - mul_72 = x_425 = None - view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_73 = x_426 * view_29 - view_29 = None - layer_norm_62 = torch.nn.functional.layer_norm( - x_426, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_426 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = (None) - x_427 = torch._C._nn.linear( - layer_norm_62, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_62 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_44 = torch.nn.functional.relu(x_427, inplace=False) - x_427 = None - pow_45 = relu_44**2 - relu_44 = None - mul_74 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ - * pow_45 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = ( - pow_45 - ) = None - x_428 = ( - mul_74 - + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ - ) - mul_74 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = (None) - x_429 = torch.nn.functional.dropout(x_428, 0.0, False, False) - x_428 = None - x_430 = torch._C._nn.linear( - x_429, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_429 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_431 = torch.nn.functional.dropout(x_430, 0.0, False, False) - x_430 = None - x_432 = mul_73 + x_431 - mul_73 = x_431 = None - view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_75 = x_432 * view_30 - view_30 = None - layer_norm_63 = torch.nn.functional.layer_norm( - x_432, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_432 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = (None) - linear_60 = torch._C._nn.linear( - layer_norm_63, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_63 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_31 = linear_60.reshape(1, pow_52, 3, 12, 32) - linear_60 = None - qkv_15 = reshape_31.permute(2, 0, 3, 1, 4) - reshape_31 = None - unbind_15 = qkv_15.unbind(0) - qkv_15 = None - q_15 = unbind_15[0] - k_15 = unbind_15[1] - v_15 = unbind_15[2] - unbind_15 = None - x_433 = torch._C._nn.scaled_dot_product_attention( - q_15, k_15, v_15, dropout_p=0.0 - ) - q_15 = k_15 = v_15 = None - transpose_16 = x_433.transpose(1, 2) - x_433 = None - x_434 = transpose_16.reshape(1, pow_52, 384) - transpose_16 = None - x_435 = torch._C._nn.linear( - x_434, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_434 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_436 = torch.nn.functional.dropout(x_435, 0.0, False, False) - x_435 = None - x_437 = mul_75 + x_436 - mul_75 = x_436 = None - view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_76 = x_437 * view_31 - view_31 = None - layer_norm_64 = torch.nn.functional.layer_norm( - x_437, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_437 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = (None) - x_438 = torch._C._nn.linear( - layer_norm_64, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_64 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_45 = torch.nn.functional.relu(x_438, inplace=False) - x_438 = None - pow_46 = relu_45**2 - relu_45 = None - mul_77 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ - * pow_46 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = ( - pow_46 - ) = None - x_439 = ( - mul_77 - + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ - ) - mul_77 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = (None) - x_440 = torch.nn.functional.dropout(x_439, 0.0, False, False) - x_439 = None - x_441 = torch._C._nn.linear( - x_440, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_440 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_442 = torch.nn.functional.dropout(x_441, 0.0, False, False) - x_441 = None - x_443 = mul_76 + x_442 - mul_76 = x_442 = None - view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_78 = x_443 * view_32 - view_32 = None - layer_norm_65 = torch.nn.functional.layer_norm( - x_443, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_443 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = (None) - linear_64 = torch._C._nn.linear( - layer_norm_65, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_65 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_33 = linear_64.reshape(1, pow_52, 3, 12, 32) - linear_64 = None - qkv_16 = reshape_33.permute(2, 0, 3, 1, 4) - reshape_33 = None - unbind_16 = qkv_16.unbind(0) - qkv_16 = None - q_16 = unbind_16[0] - k_16 = unbind_16[1] - v_16 = unbind_16[2] - unbind_16 = None - x_444 = torch._C._nn.scaled_dot_product_attention( - q_16, k_16, v_16, dropout_p=0.0 - ) - q_16 = k_16 = v_16 = None - transpose_17 = x_444.transpose(1, 2) - x_444 = None - x_445 = transpose_17.reshape(1, pow_52, 384) - transpose_17 = None - x_446 = torch._C._nn.linear( - x_445, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_445 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_447 = torch.nn.functional.dropout(x_446, 0.0, False, False) - x_446 = None - x_448 = mul_78 + x_447 - mul_78 = x_447 = None - view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_79 = x_448 * view_33 - view_33 = None - layer_norm_66 = torch.nn.functional.layer_norm( - x_448, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_448 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = (None) - x_449 = torch._C._nn.linear( - layer_norm_66, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_66 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_46 = torch.nn.functional.relu(x_449, inplace=False) - x_449 = None - pow_47 = relu_46**2 - relu_46 = None - mul_80 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ - * pow_47 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = ( - pow_47 - ) = None - x_450 = ( - mul_80 - + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ - ) - mul_80 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = (None) - x_451 = torch.nn.functional.dropout(x_450, 0.0, False, False) - x_450 = None - x_452 = torch._C._nn.linear( - x_451, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_451 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_453 = torch.nn.functional.dropout(x_452, 0.0, False, False) - x_452 = None - x_454 = mul_79 + x_453 - mul_79 = x_453 = None - view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_81 = x_454 * view_34 - view_34 = None - layer_norm_67 = torch.nn.functional.layer_norm( - x_454, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_454 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = (None) - linear_68 = torch._C._nn.linear( - layer_norm_67, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_67 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_35 = linear_68.reshape(1, pow_52, 3, 12, 32) - linear_68 = None - qkv_17 = reshape_35.permute(2, 0, 3, 1, 4) - reshape_35 = None - unbind_17 = qkv_17.unbind(0) - qkv_17 = None - q_17 = unbind_17[0] - k_17 = unbind_17[1] - v_17 = unbind_17[2] - unbind_17 = None - x_455 = torch._C._nn.scaled_dot_product_attention( - q_17, k_17, v_17, dropout_p=0.0 - ) - q_17 = k_17 = v_17 = None - transpose_18 = x_455.transpose(1, 2) - x_455 = None - x_456 = transpose_18.reshape(1, pow_52, 384) - transpose_18 = pow_52 = None - x_457 = torch._C._nn.linear( - x_456, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_456 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_458 = torch.nn.functional.dropout(x_457, 0.0, False, False) - x_457 = None - x_459 = mul_81 + x_458 - mul_81 = x_458 = None - view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_82 = x_459 * view_35 - view_35 = None - layer_norm_68 = torch.nn.functional.layer_norm( - x_459, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_459 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = (None) - x_460 = torch._C._nn.linear( - layer_norm_68, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_68 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_47 = torch.nn.functional.relu(x_460, inplace=False) - x_460 = None - pow_48 = relu_47**2 - relu_47 = None - mul_83 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ - * pow_48 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = ( - pow_48 - ) = None - x_461 = ( - mul_83 - + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ - ) - mul_83 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = (None) - x_462 = torch.nn.functional.dropout(x_461, 0.0, False, False) - x_461 = None - x_463 = torch._C._nn.linear( - x_462, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_462 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_464 = torch.nn.functional.dropout(x_463, 0.0, False, False) - x_463 = None - x_465 = mul_82 + x_464 - mul_82 = x_464 = None - transpose_19 = x_465.transpose(1, 2) - x_465 = None - x_466 = transpose_19.reshape(1, 384, sym_sum_3, sym_sum_3) - transpose_19 = sym_sum_3 = None - x_467 = x_466.permute(0, 2, 3, 1) - x_466 = None - x_468 = torch.nn.functional.layer_norm( - x_467, - (384,), - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_467 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) - x_469 = x_468.permute(0, 3, 1, 2) - x_468 = None - x_470 = torch.conv2d( - x_469, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_469 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_3 = sym_sum // 32 - sym_sum = None - sym_sum_4 = torch.sym_sum([1, floordiv_3]) - floordiv_3 = None - reshape_38 = x_470.reshape(1, 576, -1) - x_470 = None - x_471 = reshape_38.transpose(1, 2) - reshape_38 = None - view_36 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_84 = x_471 * view_36 - view_36 = None - layer_norm_70 = torch.nn.functional.layer_norm( - x_471, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_471 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_53 = sym_sum_4**2 - linear_72 = torch._C._nn.linear( - layer_norm_70, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_70 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_39 = linear_72.reshape(1, pow_53, 3, 18, 32) - linear_72 = None - qkv_18 = reshape_39.permute(2, 0, 3, 1, 4) - reshape_39 = None - unbind_18 = qkv_18.unbind(0) - qkv_18 = None - q_18 = unbind_18[0] - k_18 = unbind_18[1] - v_18 = unbind_18[2] - unbind_18 = None - x_472 = torch._C._nn.scaled_dot_product_attention( - q_18, k_18, v_18, dropout_p=0.0 - ) - q_18 = k_18 = v_18 = None - transpose_21 = x_472.transpose(1, 2) - x_472 = None - x_473 = transpose_21.reshape(1, pow_53, 576) - transpose_21 = None - x_474 = torch._C._nn.linear( - x_473, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_473 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_475 = torch.nn.functional.dropout(x_474, 0.0, False, False) - x_474 = None - x_476 = mul_84 + x_475 - mul_84 = x_475 = None - view_37 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_85 = x_476 * view_37 - view_37 = None - layer_norm_71 = torch.nn.functional.layer_norm( - x_476, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_476 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_477 = torch._C._nn.linear( - layer_norm_71, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_71 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_48 = torch.nn.functional.relu(x_477, inplace=False) - x_477 = None - pow_49 = relu_48**2 - relu_48 = None - mul_86 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_49 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_49 - ) = None - x_478 = ( - mul_86 - + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_86 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_479 = torch.nn.functional.dropout(x_478, 0.0, False, False) - x_478 = None - x_480 = torch._C._nn.linear( - x_479, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_479 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_481 = torch.nn.functional.dropout(x_480, 0.0, False, False) - x_480 = None - x_482 = mul_85 + x_481 - mul_85 = x_481 = None - view_38 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_87 = x_482 * view_38 - view_38 = None - layer_norm_72 = torch.nn.functional.layer_norm( - x_482, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_482 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_76 = torch._C._nn.linear( - layer_norm_72, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_72 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_41 = linear_76.reshape(1, pow_53, 3, 18, 32) - linear_76 = None - qkv_19 = reshape_41.permute(2, 0, 3, 1, 4) - reshape_41 = None - unbind_19 = qkv_19.unbind(0) - qkv_19 = None - q_19 = unbind_19[0] - k_19 = unbind_19[1] - v_19 = unbind_19[2] - unbind_19 = None - x_483 = torch._C._nn.scaled_dot_product_attention( - q_19, k_19, v_19, dropout_p=0.0 - ) - q_19 = k_19 = v_19 = None - transpose_22 = x_483.transpose(1, 2) - x_483 = None - x_484 = transpose_22.reshape(1, pow_53, 576) - transpose_22 = None - x_485 = torch._C._nn.linear( - x_484, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_484 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_486 = torch.nn.functional.dropout(x_485, 0.0, False, False) - x_485 = None - x_487 = mul_87 + x_486 - mul_87 = x_486 = None - view_39 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_88 = x_487 * view_39 - view_39 = None - layer_norm_73 = torch.nn.functional.layer_norm( - x_487, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_487 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_488 = torch._C._nn.linear( - layer_norm_73, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_73 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_49 = torch.nn.functional.relu(x_488, inplace=False) - x_488 = None - pow_50 = relu_49**2 - relu_49 = None - mul_89 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_50 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_50 - ) = None - x_489 = ( - mul_89 - + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_89 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_490 = torch.nn.functional.dropout(x_489, 0.0, False, False) - x_489 = None - x_491 = torch._C._nn.linear( - x_490, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_490 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_492 = torch.nn.functional.dropout(x_491, 0.0, False, False) - x_491 = None - x_493 = mul_88 + x_492 - mul_88 = x_492 = None - view_40 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_90 = x_493 * view_40 - view_40 = None - layer_norm_74 = torch.nn.functional.layer_norm( - x_493, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_493 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_80 = torch._C._nn.linear( - layer_norm_74, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_74 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_43 = linear_80.reshape(1, pow_53, 3, 18, 32) - linear_80 = None - qkv_20 = reshape_43.permute(2, 0, 3, 1, 4) - reshape_43 = None - unbind_20 = qkv_20.unbind(0) - qkv_20 = None - q_20 = unbind_20[0] - k_20 = unbind_20[1] - v_20 = unbind_20[2] - unbind_20 = None - x_494 = torch._C._nn.scaled_dot_product_attention( - q_20, k_20, v_20, dropout_p=0.0 - ) - q_20 = k_20 = v_20 = None - transpose_23 = x_494.transpose(1, 2) - x_494 = None - x_495 = transpose_23.reshape(1, pow_53, 576) - transpose_23 = pow_53 = None - x_496 = torch._C._nn.linear( - x_495, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_495 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_497 = torch.nn.functional.dropout(x_496, 0.0, False, False) - x_496 = None - x_498 = mul_90 + x_497 - mul_90 = x_497 = None - view_41 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_91 = x_498 * view_41 - view_41 = None - layer_norm_75 = torch.nn.functional.layer_norm( - x_498, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_498 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_499 = torch._C._nn.linear( - layer_norm_75, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_75 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_50 = torch.nn.functional.relu(x_499, inplace=False) - x_499 = None - pow_51 = relu_50**2 - relu_50 = None - mul_92 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_51 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_51 - ) = None - x_500 = ( - mul_92 - + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_92 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_501 = torch.nn.functional.dropout(x_500, 0.0, False, False) - x_500 = None - x_502 = torch._C._nn.linear( - x_501, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_501 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_503 = torch.nn.functional.dropout(x_502, 0.0, False, False) - x_502 = None - x_504 = mul_91 + x_503 - mul_91 = x_503 = None - transpose_24 = x_504.transpose(1, 2) - x_504 = None - x_505 = transpose_24.reshape(1, 576, sym_sum_4, sym_sum_4) - transpose_24 = sym_sum_4 = None - x_506 = torch.nn.functional.adaptive_avg_pool2d(x_505, 1) - x_505 = None - x_507 = x_506.permute(0, 2, 3, 1) - x_506 = None - x_508 = torch.nn.functional.layer_norm( - x_507, - (576,), - l_self_modules_head_modules_norm_parameters_weight_, - l_self_modules_head_modules_norm_parameters_bias_, - 1e-06, - ) - x_507 = ( - l_self_modules_head_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_norm_parameters_bias_ = None - x_509 = x_508.permute(0, 3, 1, 2) - x_508 = None - x_510 = x_509.flatten(1, -1) - x_509 = None - x_511 = torch.nn.functional.dropout(x_510, 0.0, False, False) - x_510 = None - x_512 = torch._C._nn.linear( - x_511, - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, - ) - x_511 = ( - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None - relu_51 = torch.nn.functional.relu(x_512, inplace=False) - x_512 = None - x_513 = torch.square(relu_51) - relu_51 = None - x_514 = torch.nn.functional.layer_norm( - x_513, - (2304,), - l_self_modules_head_modules_fc_modules_norm_parameters_weight_, - l_self_modules_head_modules_fc_modules_norm_parameters_bias_, - 1e-06, - ) - x_513 = ( - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None - x_515 = torch.nn.functional.dropout(x_514, 0.0, False, False) - x_514 = None - x_516 = torch._C._nn.linear( - x_515, - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, - ) - x_515 = ( - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None - return (x_516,) diff --git a/samples/timm/caformer_m36.sail_in22k/weight_meta.py b/samples/timm/caformer_m36.sail_in22k/weight_meta.py deleted file mode 100644 index 7f331146f7..0000000000 --- a/samples/timm/caformer_m36.sail_in22k/weight_meta.py +++ /dev/null @@ -1,3968 +0,0 @@ -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: - name = "L_self_modules_stem_modules_conv_parameters_weight_" - shape = [96, 3, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: - name = "L_self_modules_stem_modules_conv_parameters_bias_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_s1: - name = "s1" - shape = [] - dtype = "torch.int64" - device = "cpu" - mean = None - std = None - data = [4] - - -class Program_weight_tensor_meta_L_x_: - name = "L_x_" - shape = [1, 3, 224, 224] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.227 - std = 1.285 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: - name = "L_self_modules_stem_modules_norm_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [192, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [192, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [96, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [384, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [96, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [192, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [192, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [96, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [384, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [96, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [192, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [192, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [96, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [384, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [96, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" - shape = [192, 96, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" - shape = [384, 192, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" - shape = [576, 384, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1728, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [576, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [2304, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [576, 2304] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1728, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [576, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [2304, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [576, 2304] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1728, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [576, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [2304, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [576, 2304] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_norm_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_norm_parameters_bias_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" - shape = [2304, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" - shape = [2304] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" - shape = [2304] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" - shape = [2304] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" - shape = [21841, 2304] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" - shape = [21841] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None diff --git a/samples/timm/caformer_m36.sail_in22k_ft_in1k/graph_net.json b/samples/timm/caformer_m36.sail_in22k_ft_in1k/graph_net.json deleted file mode 100644 index 1373fe3b58..0000000000 --- a/samples/timm/caformer_m36.sail_in22k_ft_in1k/graph_net.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "framework": "torch", - "num_devices_required": 1, - "num_nodes_required": 1 -} \ No newline at end of file diff --git a/samples/timm/caformer_m36.sail_in22k_ft_in1k/input_meta.py b/samples/timm/caformer_m36.sail_in22k_ft_in1k/input_meta.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_m36.sail_in22k_ft_in1k/input_tensor_constraints.py b/samples/timm/caformer_m36.sail_in22k_ft_in1k/input_tensor_constraints.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_m36.sail_in22k_ft_in1k/model.py b/samples/timm/caformer_m36.sail_in22k_ft_in1k/model.py deleted file mode 100644 index dcce286c5c..0000000000 --- a/samples/timm/caformer_m36.sail_in22k_ft_in1k/model.py +++ /dev/null @@ -1,4754 +0,0 @@ -import torch - - -class GraphModule(torch.nn.Module): - def forward( - self, - L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - s1: torch.SymInt, - L_x_: torch.Tensor, - L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - ): - l_self_modules_stem_modules_conv_parameters_weight_ = ( - L_self_modules_stem_modules_conv_parameters_weight_ - ) - l_self_modules_stem_modules_conv_parameters_bias_ = ( - L_self_modules_stem_modules_conv_parameters_bias_ - ) - l_x_ = L_x_ - l_self_modules_stem_modules_norm_parameters_weight_ = ( - L_self_modules_stem_modules_norm_parameters_weight_ - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_head_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ - ) - x = torch.conv2d( - l_x_, - l_self_modules_stem_modules_conv_parameters_weight_, - l_self_modules_stem_modules_conv_parameters_bias_, - (4, 4), - (2, 2), - (1, 1), - 1, - ) - l_x_ = ( - l_self_modules_stem_modules_conv_parameters_weight_ - ) = l_self_modules_stem_modules_conv_parameters_bias_ = None - x_1 = x.permute(0, 2, 3, 1) - x = None - x_2 = torch.nn.functional.layer_norm( - x_1, (96,), l_self_modules_stem_modules_norm_parameters_weight_, None, 1e-06 - ) - x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None - x_3 = x_2.permute(0, 3, 1, 2) - x_2 = None - sym_sum = torch.sym_sum([-3, s1]) - s1 = None - floordiv = sym_sum // 4 - sym_sum_1 = torch.sym_sum([1, floordiv]) - floordiv = sym_sum_1 = None - x_4 = x_3.permute(0, 2, 3, 1) - x_5 = torch.nn.functional.layer_norm( - x_4, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_6 = x_5.permute(0, 3, 1, 2) - x_5 = None - x_7 = torch.conv2d( - x_6, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu = torch.nn.functional.relu(x_7, inplace=False) - x_7 = None - pow_1 = relu**2 - relu = None - mul = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_1 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_1 - ) = None - x_8 = ( - mul - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_9 = torch.conv2d( - x_8, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 192, - ) - x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_10 = torch.conv2d( - x_9, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_11 = x_3 + x_10 - x_3 = x_10 = None - x_12 = x_11.permute(0, 2, 3, 1) - x_13 = torch.nn.functional.layer_norm( - x_12, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_14 = x_13.permute(0, 3, 1, 2) - x_13 = None - x_15 = torch.conv2d( - x_14, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_1 = torch.nn.functional.relu(x_15, inplace=False) - x_15 = None - pow_2 = relu_1**2 - relu_1 = None - mul_1 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_2 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_2 - ) = None - x_16 = ( - mul_1 - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) - x_16 = None - x_18 = torch.conv2d( - x_17, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) - x_18 = None - x_20 = x_11 + x_19 - x_11 = x_19 = None - x_21 = x_20.permute(0, 2, 3, 1) - x_22 = torch.nn.functional.layer_norm( - x_21, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_23 = x_22.permute(0, 3, 1, 2) - x_22 = None - x_24 = torch.conv2d( - x_23, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_2 = torch.nn.functional.relu(x_24, inplace=False) - x_24 = None - pow_3 = relu_2**2 - relu_2 = None - mul_2 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_3 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_3 - ) = None - x_25 = ( - mul_2 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_26 = torch.conv2d( - x_25, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 192, - ) - x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_27 = torch.conv2d( - x_26, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_28 = x_20 + x_27 - x_20 = x_27 = None - x_29 = x_28.permute(0, 2, 3, 1) - x_30 = torch.nn.functional.layer_norm( - x_29, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_31 = x_30.permute(0, 3, 1, 2) - x_30 = None - x_32 = torch.conv2d( - x_31, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_3 = torch.nn.functional.relu(x_32, inplace=False) - x_32 = None - pow_4 = relu_3**2 - relu_3 = None - mul_3 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_4 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_4 - ) = None - x_33 = ( - mul_3 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) - x_33 = None - x_35 = torch.conv2d( - x_34, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) - x_35 = None - x_37 = x_28 + x_36 - x_28 = x_36 = None - x_38 = x_37.permute(0, 2, 3, 1) - x_39 = torch.nn.functional.layer_norm( - x_38, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_40 = x_39.permute(0, 3, 1, 2) - x_39 = None - x_41 = torch.conv2d( - x_40, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_4 = torch.nn.functional.relu(x_41, inplace=False) - x_41 = None - pow_5 = relu_4**2 - relu_4 = None - mul_4 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_5 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_5 - ) = None - x_42 = ( - mul_4 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_43 = torch.conv2d( - x_42, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 192, - ) - x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_44 = torch.conv2d( - x_43, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_45 = x_37 + x_44 - x_37 = x_44 = None - x_46 = x_45.permute(0, 2, 3, 1) - x_47 = torch.nn.functional.layer_norm( - x_46, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_48 = x_47.permute(0, 3, 1, 2) - x_47 = None - x_49 = torch.conv2d( - x_48, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_5 = torch.nn.functional.relu(x_49, inplace=False) - x_49 = None - pow_6 = relu_5**2 - relu_5 = None - mul_5 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_6 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_6 - ) = None - x_50 = ( - mul_5 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) - x_50 = None - x_52 = torch.conv2d( - x_51, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) - x_52 = None - x_54 = x_45 + x_53 - x_45 = x_53 = None - x_55 = x_54.permute(0, 2, 3, 1) - x_54 = None - x_56 = torch.nn.functional.layer_norm( - x_55, - (96,), - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) - x_57 = x_56.permute(0, 3, 1, 2) - x_56 = None - x_58 = torch.conv2d( - x_57, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_1 = sym_sum // 8 - sym_sum_2 = torch.sym_sum([1, floordiv_1]) - floordiv_1 = sym_sum_2 = None - x_59 = x_58.permute(0, 2, 3, 1) - x_60 = torch.nn.functional.layer_norm( - x_59, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_61 = x_60.permute(0, 3, 1, 2) - x_60 = None - x_62 = torch.conv2d( - x_61, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_6 = torch.nn.functional.relu(x_62, inplace=False) - x_62 = None - pow_7 = relu_6**2 - relu_6 = None - mul_6 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_7 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_7 - ) = None - x_63 = ( - mul_6 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_64 = torch.conv2d( - x_63, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_65 = torch.conv2d( - x_64, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_66 = x_58 + x_65 - x_58 = x_65 = None - x_67 = x_66.permute(0, 2, 3, 1) - x_68 = torch.nn.functional.layer_norm( - x_67, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_69 = x_68.permute(0, 3, 1, 2) - x_68 = None - x_70 = torch.conv2d( - x_69, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_7 = torch.nn.functional.relu(x_70, inplace=False) - x_70 = None - pow_8 = relu_7**2 - relu_7 = None - mul_7 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_8 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_8 - ) = None - x_71 = ( - mul_7 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) - x_71 = None - x_73 = torch.conv2d( - x_72, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) - x_73 = None - x_75 = x_66 + x_74 - x_66 = x_74 = None - x_76 = x_75.permute(0, 2, 3, 1) - x_77 = torch.nn.functional.layer_norm( - x_76, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_78 = x_77.permute(0, 3, 1, 2) - x_77 = None - x_79 = torch.conv2d( - x_78, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_8 = torch.nn.functional.relu(x_79, inplace=False) - x_79 = None - pow_9 = relu_8**2 - relu_8 = None - mul_8 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_9 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_9 - ) = None - x_80 = ( - mul_8 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_81 = torch.conv2d( - x_80, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_82 = torch.conv2d( - x_81, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_83 = x_75 + x_82 - x_75 = x_82 = None - x_84 = x_83.permute(0, 2, 3, 1) - x_85 = torch.nn.functional.layer_norm( - x_84, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_86 = x_85.permute(0, 3, 1, 2) - x_85 = None - x_87 = torch.conv2d( - x_86, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_9 = torch.nn.functional.relu(x_87, inplace=False) - x_87 = None - pow_10 = relu_9**2 - relu_9 = None - mul_9 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_10 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_10 - ) = None - x_88 = ( - mul_9 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) - x_88 = None - x_90 = torch.conv2d( - x_89, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) - x_90 = None - x_92 = x_83 + x_91 - x_83 = x_91 = None - x_93 = x_92.permute(0, 2, 3, 1) - x_94 = torch.nn.functional.layer_norm( - x_93, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_95 = x_94.permute(0, 3, 1, 2) - x_94 = None - x_96 = torch.conv2d( - x_95, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_10 = torch.nn.functional.relu(x_96, inplace=False) - x_96 = None - pow_11 = relu_10**2 - relu_10 = None - mul_10 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_11 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_11 - ) = None - x_97 = ( - mul_10 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_98 = torch.conv2d( - x_97, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_99 = torch.conv2d( - x_98, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_100 = x_92 + x_99 - x_92 = x_99 = None - x_101 = x_100.permute(0, 2, 3, 1) - x_102 = torch.nn.functional.layer_norm( - x_101, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_103 = x_102.permute(0, 3, 1, 2) - x_102 = None - x_104 = torch.conv2d( - x_103, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_11 = torch.nn.functional.relu(x_104, inplace=False) - x_104 = None - pow_12 = relu_11**2 - relu_11 = None - mul_11 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_12 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_12 - ) = None - x_105 = ( - mul_11 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) - x_105 = None - x_107 = torch.conv2d( - x_106, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) - x_107 = None - x_109 = x_100 + x_108 - x_100 = x_108 = None - x_110 = x_109.permute(0, 2, 3, 1) - x_111 = torch.nn.functional.layer_norm( - x_110, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_110 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) - x_112 = x_111.permute(0, 3, 1, 2) - x_111 = None - x_113 = torch.conv2d( - x_112, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_112 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_12 = torch.nn.functional.relu(x_113, inplace=False) - x_113 = None - pow_13 = relu_12**2 - relu_12 = None - mul_12 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ - * pow_13 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_13 - ) = None - x_114 = ( - mul_12 - + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_12 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_115 = torch.conv2d( - x_114, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_114 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_116 = torch.conv2d( - x_115, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_115 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_117 = x_109 + x_116 - x_109 = x_116 = None - x_118 = x_117.permute(0, 2, 3, 1) - x_119 = torch.nn.functional.layer_norm( - x_118, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_118 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) - x_120 = x_119.permute(0, 3, 1, 2) - x_119 = None - x_121 = torch.conv2d( - x_120, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_120 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_13 = torch.nn.functional.relu(x_121, inplace=False) - x_121 = None - pow_14 = relu_13**2 - relu_13 = None - mul_13 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - * pow_14 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( - pow_14 - ) = None - x_122 = ( - mul_13 - + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - ) - mul_13 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) - x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) - x_122 = None - x_124 = torch.conv2d( - x_123, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_123 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) - x_124 = None - x_126 = x_117 + x_125 - x_117 = x_125 = None - x_127 = x_126.permute(0, 2, 3, 1) - x_128 = torch.nn.functional.layer_norm( - x_127, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_127 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) - x_129 = x_128.permute(0, 3, 1, 2) - x_128 = None - x_130 = torch.conv2d( - x_129, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_129 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_14 = torch.nn.functional.relu(x_130, inplace=False) - x_130 = None - pow_15 = relu_14**2 - relu_14 = None - mul_14 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ - * pow_15 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_15 - ) = None - x_131 = ( - mul_14 - + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_14 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_132 = torch.conv2d( - x_131, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_131 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_133 = torch.conv2d( - x_132, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_132 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_134 = x_126 + x_133 - x_126 = x_133 = None - x_135 = x_134.permute(0, 2, 3, 1) - x_136 = torch.nn.functional.layer_norm( - x_135, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_135 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) - x_137 = x_136.permute(0, 3, 1, 2) - x_136 = None - x_138 = torch.conv2d( - x_137, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_137 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_15 = torch.nn.functional.relu(x_138, inplace=False) - x_138 = None - pow_16 = relu_15**2 - relu_15 = None - mul_15 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - * pow_16 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( - pow_16 - ) = None - x_139 = ( - mul_15 - + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - ) - mul_15 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) - x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) - x_139 = None - x_141 = torch.conv2d( - x_140, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_140 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) - x_141 = None - x_143 = x_134 + x_142 - x_134 = x_142 = None - x_144 = x_143.permute(0, 2, 3, 1) - x_145 = torch.nn.functional.layer_norm( - x_144, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_144 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) - x_146 = x_145.permute(0, 3, 1, 2) - x_145 = None - x_147 = torch.conv2d( - x_146, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_146 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_16 = torch.nn.functional.relu(x_147, inplace=False) - x_147 = None - pow_17 = relu_16**2 - relu_16 = None - mul_16 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ - * pow_17 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_17 - ) = None - x_148 = ( - mul_16 - + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_16 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_149 = torch.conv2d( - x_148, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_148 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_150 = torch.conv2d( - x_149, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_149 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_151 = x_143 + x_150 - x_143 = x_150 = None - x_152 = x_151.permute(0, 2, 3, 1) - x_153 = torch.nn.functional.layer_norm( - x_152, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_152 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) - x_154 = x_153.permute(0, 3, 1, 2) - x_153 = None - x_155 = torch.conv2d( - x_154, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_154 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_17 = torch.nn.functional.relu(x_155, inplace=False) - x_155 = None - pow_18 = relu_17**2 - relu_17 = None - mul_17 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - * pow_18 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( - pow_18 - ) = None - x_156 = ( - mul_17 - + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - ) - mul_17 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) - x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) - x_156 = None - x_158 = torch.conv2d( - x_157, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_157 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) - x_158 = None - x_160 = x_151 + x_159 - x_151 = x_159 = None - x_161 = x_160.permute(0, 2, 3, 1) - x_162 = torch.nn.functional.layer_norm( - x_161, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_161 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) - x_163 = x_162.permute(0, 3, 1, 2) - x_162 = None - x_164 = torch.conv2d( - x_163, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_163 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_18 = torch.nn.functional.relu(x_164, inplace=False) - x_164 = None - pow_19 = relu_18**2 - relu_18 = None - mul_18 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ - * pow_19 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_19 - ) = None - x_165 = ( - mul_18 - + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_18 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_166 = torch.conv2d( - x_165, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_165 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_167 = torch.conv2d( - x_166, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_166 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_168 = x_160 + x_167 - x_160 = x_167 = None - x_169 = x_168.permute(0, 2, 3, 1) - x_170 = torch.nn.functional.layer_norm( - x_169, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_169 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) - x_171 = x_170.permute(0, 3, 1, 2) - x_170 = None - x_172 = torch.conv2d( - x_171, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_171 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_19 = torch.nn.functional.relu(x_172, inplace=False) - x_172 = None - pow_20 = relu_19**2 - relu_19 = None - mul_19 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - * pow_20 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( - pow_20 - ) = None - x_173 = ( - mul_19 - + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - ) - mul_19 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) - x_174 = torch.nn.functional.dropout(x_173, 0.0, False, False) - x_173 = None - x_175 = torch.conv2d( - x_174, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_174 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) - x_175 = None - x_177 = x_168 + x_176 - x_168 = x_176 = None - x_178 = x_177.permute(0, 2, 3, 1) - x_179 = torch.nn.functional.layer_norm( - x_178, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_178 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) - x_180 = x_179.permute(0, 3, 1, 2) - x_179 = None - x_181 = torch.conv2d( - x_180, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_180 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_20 = torch.nn.functional.relu(x_181, inplace=False) - x_181 = None - pow_21 = relu_20**2 - relu_20 = None - mul_20 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ - * pow_21 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_21 - ) = None - x_182 = ( - mul_20 - + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_20 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_183 = torch.conv2d( - x_182, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_182 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_184 = torch.conv2d( - x_183, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_183 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_185 = x_177 + x_184 - x_177 = x_184 = None - x_186 = x_185.permute(0, 2, 3, 1) - x_187 = torch.nn.functional.layer_norm( - x_186, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_186 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) - x_188 = x_187.permute(0, 3, 1, 2) - x_187 = None - x_189 = torch.conv2d( - x_188, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_188 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_21 = torch.nn.functional.relu(x_189, inplace=False) - x_189 = None - pow_22 = relu_21**2 - relu_21 = None - mul_21 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - * pow_22 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( - pow_22 - ) = None - x_190 = ( - mul_21 - + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - ) - mul_21 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) - x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) - x_190 = None - x_192 = torch.conv2d( - x_191, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_191 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) - x_192 = None - x_194 = x_185 + x_193 - x_185 = x_193 = None - x_195 = x_194.permute(0, 2, 3, 1) - x_196 = torch.nn.functional.layer_norm( - x_195, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_195 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) - x_197 = x_196.permute(0, 3, 1, 2) - x_196 = None - x_198 = torch.conv2d( - x_197, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_197 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_22 = torch.nn.functional.relu(x_198, inplace=False) - x_198 = None - pow_23 = relu_22**2 - relu_22 = None - mul_22 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ - * pow_23 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_23 - ) = None - x_199 = ( - mul_22 - + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_22 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_200 = torch.conv2d( - x_199, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_199 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_201 = torch.conv2d( - x_200, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_200 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_202 = x_194 + x_201 - x_194 = x_201 = None - x_203 = x_202.permute(0, 2, 3, 1) - x_204 = torch.nn.functional.layer_norm( - x_203, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_203 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) - x_205 = x_204.permute(0, 3, 1, 2) - x_204 = None - x_206 = torch.conv2d( - x_205, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_205 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_23 = torch.nn.functional.relu(x_206, inplace=False) - x_206 = None - pow_24 = relu_23**2 - relu_23 = None - mul_23 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - * pow_24 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( - pow_24 - ) = None - x_207 = ( - mul_23 - + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - ) - mul_23 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) - x_208 = torch.nn.functional.dropout(x_207, 0.0, False, False) - x_207 = None - x_209 = torch.conv2d( - x_208, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_208 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) - x_209 = None - x_211 = x_202 + x_210 - x_202 = x_210 = None - x_212 = x_211.permute(0, 2, 3, 1) - x_213 = torch.nn.functional.layer_norm( - x_212, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_212 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) - x_214 = x_213.permute(0, 3, 1, 2) - x_213 = None - x_215 = torch.conv2d( - x_214, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_214 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_24 = torch.nn.functional.relu(x_215, inplace=False) - x_215 = None - pow_25 = relu_24**2 - relu_24 = None - mul_24 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ - * pow_25 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_25 - ) = None - x_216 = ( - mul_24 - + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_24 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_217 = torch.conv2d( - x_216, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_216 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_218 = torch.conv2d( - x_217, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_217 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_219 = x_211 + x_218 - x_211 = x_218 = None - x_220 = x_219.permute(0, 2, 3, 1) - x_221 = torch.nn.functional.layer_norm( - x_220, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_220 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) - x_222 = x_221.permute(0, 3, 1, 2) - x_221 = None - x_223 = torch.conv2d( - x_222, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_222 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_25 = torch.nn.functional.relu(x_223, inplace=False) - x_223 = None - pow_26 = relu_25**2 - relu_25 = None - mul_25 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - * pow_26 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( - pow_26 - ) = None - x_224 = ( - mul_25 - + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - ) - mul_25 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) - x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) - x_224 = None - x_226 = torch.conv2d( - x_225, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_225 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) - x_226 = None - x_228 = x_219 + x_227 - x_219 = x_227 = None - x_229 = x_228.permute(0, 2, 3, 1) - x_230 = torch.nn.functional.layer_norm( - x_229, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_229 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) - x_231 = x_230.permute(0, 3, 1, 2) - x_230 = None - x_232 = torch.conv2d( - x_231, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_231 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_26 = torch.nn.functional.relu(x_232, inplace=False) - x_232 = None - pow_27 = relu_26**2 - relu_26 = None - mul_26 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ - * pow_27 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_27 - ) = None - x_233 = ( - mul_26 - + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_26 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_234 = torch.conv2d( - x_233, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_233 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_235 = torch.conv2d( - x_234, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_234 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_236 = x_228 + x_235 - x_228 = x_235 = None - x_237 = x_236.permute(0, 2, 3, 1) - x_238 = torch.nn.functional.layer_norm( - x_237, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_237 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) - x_239 = x_238.permute(0, 3, 1, 2) - x_238 = None - x_240 = torch.conv2d( - x_239, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_239 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_27 = torch.nn.functional.relu(x_240, inplace=False) - x_240 = None - pow_28 = relu_27**2 - relu_27 = None - mul_27 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - * pow_28 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( - pow_28 - ) = None - x_241 = ( - mul_27 - + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - ) - mul_27 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) - x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) - x_241 = None - x_243 = torch.conv2d( - x_242, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_242 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_244 = torch.nn.functional.dropout(x_243, 0.0, False, False) - x_243 = None - x_245 = x_236 + x_244 - x_236 = x_244 = None - x_246 = x_245.permute(0, 2, 3, 1) - x_247 = torch.nn.functional.layer_norm( - x_246, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_246 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) - x_248 = x_247.permute(0, 3, 1, 2) - x_247 = None - x_249 = torch.conv2d( - x_248, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_248 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_28 = torch.nn.functional.relu(x_249, inplace=False) - x_249 = None - pow_29 = relu_28**2 - relu_28 = None - mul_28 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ - * pow_29 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_29 - ) = None - x_250 = ( - mul_28 - + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_28 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_251 = torch.conv2d( - x_250, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_250 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_252 = torch.conv2d( - x_251, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_251 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_253 = x_245 + x_252 - x_245 = x_252 = None - x_254 = x_253.permute(0, 2, 3, 1) - x_255 = torch.nn.functional.layer_norm( - x_254, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_254 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) - x_256 = x_255.permute(0, 3, 1, 2) - x_255 = None - x_257 = torch.conv2d( - x_256, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_256 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_29 = torch.nn.functional.relu(x_257, inplace=False) - x_257 = None - pow_30 = relu_29**2 - relu_29 = None - mul_29 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - * pow_30 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( - pow_30 - ) = None - x_258 = ( - mul_29 - + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - ) - mul_29 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) - x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) - x_258 = None - x_260 = torch.conv2d( - x_259, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_259 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_261 = torch.nn.functional.dropout(x_260, 0.0, False, False) - x_260 = None - x_262 = x_253 + x_261 - x_253 = x_261 = None - x_263 = x_262.permute(0, 2, 3, 1) - x_262 = None - x_264 = torch.nn.functional.layer_norm( - x_263, - (192,), - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_263 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) - x_265 = x_264.permute(0, 3, 1, 2) - x_264 = None - x_266 = torch.conv2d( - x_265, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_265 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_2 = sym_sum // 16 - sym_sum_3 = torch.sym_sum([1, floordiv_2]) - floordiv_2 = None - reshape = x_266.reshape(1, 384, -1) - x_266 = None - x_267 = reshape.transpose(1, 2) - reshape = None - view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_30 = x_267 * view - view = None - layer_norm_33 = torch.nn.functional.layer_norm( - x_267, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_52 = sym_sum_3**2 - linear = torch._C._nn.linear( - layer_norm_33, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_33 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_1 = linear.reshape(1, pow_52, 3, 12, 32) - linear = None - qkv = reshape_1.permute(2, 0, 3, 1, 4) - reshape_1 = None - unbind = qkv.unbind(0) - qkv = None - q = unbind[0] - k = unbind[1] - v = unbind[2] - unbind = None - x_268 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) - q = k = v = None - transpose_1 = x_268.transpose(1, 2) - x_268 = None - x_269 = transpose_1.reshape(1, pow_52, 384) - transpose_1 = None - x_270 = torch._C._nn.linear( - x_269, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_269 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_271 = torch.nn.functional.dropout(x_270, 0.0, False, False) - x_270 = None - x_272 = mul_30 + x_271 - mul_30 = x_271 = None - view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_31 = x_272 * view_1 - view_1 = None - layer_norm_34 = torch.nn.functional.layer_norm( - x_272, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_272 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_273 = torch._C._nn.linear( - layer_norm_34, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_34 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_30 = torch.nn.functional.relu(x_273, inplace=False) - x_273 = None - pow_31 = relu_30**2 - relu_30 = None - mul_32 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_31 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_31 - ) = None - x_274 = ( - mul_32 - + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_275 = torch.nn.functional.dropout(x_274, 0.0, False, False) - x_274 = None - x_276 = torch._C._nn.linear( - x_275, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_277 = torch.nn.functional.dropout(x_276, 0.0, False, False) - x_276 = None - x_278 = mul_31 + x_277 - mul_31 = x_277 = None - view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_33 = x_278 * view_2 - view_2 = None - layer_norm_35 = torch.nn.functional.layer_norm( - x_278, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_4 = torch._C._nn.linear( - layer_norm_35, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_3 = linear_4.reshape(1, pow_52, 3, 12, 32) - linear_4 = None - qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) - reshape_3 = None - unbind_1 = qkv_1.unbind(0) - qkv_1 = None - q_1 = unbind_1[0] - k_1 = unbind_1[1] - v_1 = unbind_1[2] - unbind_1 = None - x_279 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) - q_1 = k_1 = v_1 = None - transpose_2 = x_279.transpose(1, 2) - x_279 = None - x_280 = transpose_2.reshape(1, pow_52, 384) - transpose_2 = None - x_281 = torch._C._nn.linear( - x_280, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_280 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) - x_281 = None - x_283 = mul_33 + x_282 - mul_33 = x_282 = None - view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_34 = x_283 * view_3 - view_3 = None - layer_norm_36 = torch.nn.functional.layer_norm( - x_283, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_283 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_284 = torch._C._nn.linear( - layer_norm_36, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_36 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_31 = torch.nn.functional.relu(x_284, inplace=False) - x_284 = None - pow_32 = relu_31**2 - relu_31 = None - mul_35 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_32 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_32 - ) = None - x_285 = ( - mul_35 - + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_286 = torch.nn.functional.dropout(x_285, 0.0, False, False) - x_285 = None - x_287 = torch._C._nn.linear( - x_286, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) - x_287 = None - x_289 = mul_34 + x_288 - mul_34 = x_288 = None - view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_36 = x_289 * view_4 - view_4 = None - layer_norm_37 = torch.nn.functional.layer_norm( - x_289, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_8 = torch._C._nn.linear( - layer_norm_37, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_37 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_5 = linear_8.reshape(1, pow_52, 3, 12, 32) - linear_8 = None - qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) - reshape_5 = None - unbind_2 = qkv_2.unbind(0) - qkv_2 = None - q_2 = unbind_2[0] - k_2 = unbind_2[1] - v_2 = unbind_2[2] - unbind_2 = None - x_290 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) - q_2 = k_2 = v_2 = None - transpose_3 = x_290.transpose(1, 2) - x_290 = None - x_291 = transpose_3.reshape(1, pow_52, 384) - transpose_3 = None - x_292 = torch._C._nn.linear( - x_291, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_291 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_293 = torch.nn.functional.dropout(x_292, 0.0, False, False) - x_292 = None - x_294 = mul_36 + x_293 - mul_36 = x_293 = None - view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_37 = x_294 * view_5 - view_5 = None - layer_norm_38 = torch.nn.functional.layer_norm( - x_294, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_294 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_295 = torch._C._nn.linear( - layer_norm_38, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_32 = torch.nn.functional.relu(x_295, inplace=False) - x_295 = None - pow_33 = relu_32**2 - relu_32 = None - mul_38 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_33 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_33 - ) = None - x_296 = ( - mul_38 - + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_297 = torch.nn.functional.dropout(x_296, 0.0, False, False) - x_296 = None - x_298 = torch._C._nn.linear( - x_297, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_299 = torch.nn.functional.dropout(x_298, 0.0, False, False) - x_298 = None - x_300 = mul_37 + x_299 - mul_37 = x_299 = None - view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_39 = x_300 * view_6 - view_6 = None - layer_norm_39 = torch.nn.functional.layer_norm( - x_300, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) - linear_12 = torch._C._nn.linear( - layer_norm_39, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_39 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_7 = linear_12.reshape(1, pow_52, 3, 12, 32) - linear_12 = None - qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) - reshape_7 = None - unbind_3 = qkv_3.unbind(0) - qkv_3 = None - q_3 = unbind_3[0] - k_3 = unbind_3[1] - v_3 = unbind_3[2] - unbind_3 = None - x_301 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) - q_3 = k_3 = v_3 = None - transpose_4 = x_301.transpose(1, 2) - x_301 = None - x_302 = transpose_4.reshape(1, pow_52, 384) - transpose_4 = None - x_303 = torch._C._nn.linear( - x_302, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_302 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_304 = torch.nn.functional.dropout(x_303, 0.0, False, False) - x_303 = None - x_305 = mul_39 + x_304 - mul_39 = x_304 = None - view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_40 = x_305 * view_7 - view_7 = None - layer_norm_40 = torch.nn.functional.layer_norm( - x_305, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_305 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) - x_306 = torch._C._nn.linear( - layer_norm_40, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_40 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_33 = torch.nn.functional.relu(x_306, inplace=False) - x_306 = None - pow_34 = relu_33**2 - relu_33 = None - mul_41 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - * pow_34 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( - pow_34 - ) = None - x_307 = ( - mul_41 - + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - ) - mul_41 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) - x_308 = torch.nn.functional.dropout(x_307, 0.0, False, False) - x_307 = None - x_309 = torch._C._nn.linear( - x_308, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) - x_309 = None - x_311 = mul_40 + x_310 - mul_40 = x_310 = None - view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_42 = x_311 * view_8 - view_8 = None - layer_norm_41 = torch.nn.functional.layer_norm( - x_311, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) - linear_16 = torch._C._nn.linear( - layer_norm_41, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_41 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_9 = linear_16.reshape(1, pow_52, 3, 12, 32) - linear_16 = None - qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) - reshape_9 = None - unbind_4 = qkv_4.unbind(0) - qkv_4 = None - q_4 = unbind_4[0] - k_4 = unbind_4[1] - v_4 = unbind_4[2] - unbind_4 = None - x_312 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) - q_4 = k_4 = v_4 = None - transpose_5 = x_312.transpose(1, 2) - x_312 = None - x_313 = transpose_5.reshape(1, pow_52, 384) - transpose_5 = None - x_314 = torch._C._nn.linear( - x_313, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_313 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) - x_314 = None - x_316 = mul_42 + x_315 - mul_42 = x_315 = None - view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_43 = x_316 * view_9 - view_9 = None - layer_norm_42 = torch.nn.functional.layer_norm( - x_316, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_316 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) - x_317 = torch._C._nn.linear( - layer_norm_42, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_42 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_34 = torch.nn.functional.relu(x_317, inplace=False) - x_317 = None - pow_35 = relu_34**2 - relu_34 = None - mul_44 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - * pow_35 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( - pow_35 - ) = None - x_318 = ( - mul_44 - + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - ) - mul_44 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) - x_319 = torch.nn.functional.dropout(x_318, 0.0, False, False) - x_318 = None - x_320 = torch._C._nn.linear( - x_319, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_321 = torch.nn.functional.dropout(x_320, 0.0, False, False) - x_320 = None - x_322 = mul_43 + x_321 - mul_43 = x_321 = None - view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_45 = x_322 * view_10 - view_10 = None - layer_norm_43 = torch.nn.functional.layer_norm( - x_322, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) - linear_20 = torch._C._nn.linear( - layer_norm_43, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_43 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_11 = linear_20.reshape(1, pow_52, 3, 12, 32) - linear_20 = None - qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) - reshape_11 = None - unbind_5 = qkv_5.unbind(0) - qkv_5 = None - q_5 = unbind_5[0] - k_5 = unbind_5[1] - v_5 = unbind_5[2] - unbind_5 = None - x_323 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) - q_5 = k_5 = v_5 = None - transpose_6 = x_323.transpose(1, 2) - x_323 = None - x_324 = transpose_6.reshape(1, pow_52, 384) - transpose_6 = None - x_325 = torch._C._nn.linear( - x_324, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_324 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_326 = torch.nn.functional.dropout(x_325, 0.0, False, False) - x_325 = None - x_327 = mul_45 + x_326 - mul_45 = x_326 = None - view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_46 = x_327 * view_11 - view_11 = None - layer_norm_44 = torch.nn.functional.layer_norm( - x_327, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_327 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) - x_328 = torch._C._nn.linear( - layer_norm_44, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_44 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_35 = torch.nn.functional.relu(x_328, inplace=False) - x_328 = None - pow_36 = relu_35**2 - relu_35 = None - mul_47 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - * pow_36 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( - pow_36 - ) = None - x_329 = ( - mul_47 - + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - ) - mul_47 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) - x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) - x_329 = None - x_331 = torch._C._nn.linear( - x_330, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_332 = torch.nn.functional.dropout(x_331, 0.0, False, False) - x_331 = None - x_333 = mul_46 + x_332 - mul_46 = x_332 = None - view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_48 = x_333 * view_12 - view_12 = None - layer_norm_45 = torch.nn.functional.layer_norm( - x_333, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) - linear_24 = torch._C._nn.linear( - layer_norm_45, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_45 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_13 = linear_24.reshape(1, pow_52, 3, 12, 32) - linear_24 = None - qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) - reshape_13 = None - unbind_6 = qkv_6.unbind(0) - qkv_6 = None - q_6 = unbind_6[0] - k_6 = unbind_6[1] - v_6 = unbind_6[2] - unbind_6 = None - x_334 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) - q_6 = k_6 = v_6 = None - transpose_7 = x_334.transpose(1, 2) - x_334 = None - x_335 = transpose_7.reshape(1, pow_52, 384) - transpose_7 = None - x_336 = torch._C._nn.linear( - x_335, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_335 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_337 = torch.nn.functional.dropout(x_336, 0.0, False, False) - x_336 = None - x_338 = mul_48 + x_337 - mul_48 = x_337 = None - view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_49 = x_338 * view_13 - view_13 = None - layer_norm_46 = torch.nn.functional.layer_norm( - x_338, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_338 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) - x_339 = torch._C._nn.linear( - layer_norm_46, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_46 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_36 = torch.nn.functional.relu(x_339, inplace=False) - x_339 = None - pow_37 = relu_36**2 - relu_36 = None - mul_50 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - * pow_37 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( - pow_37 - ) = None - x_340 = ( - mul_50 - + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - ) - mul_50 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) - x_341 = torch.nn.functional.dropout(x_340, 0.0, False, False) - x_340 = None - x_342 = torch._C._nn.linear( - x_341, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_343 = torch.nn.functional.dropout(x_342, 0.0, False, False) - x_342 = None - x_344 = mul_49 + x_343 - mul_49 = x_343 = None - view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_51 = x_344 * view_14 - view_14 = None - layer_norm_47 = torch.nn.functional.layer_norm( - x_344, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) - linear_28 = torch._C._nn.linear( - layer_norm_47, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_47 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_15 = linear_28.reshape(1, pow_52, 3, 12, 32) - linear_28 = None - qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) - reshape_15 = None - unbind_7 = qkv_7.unbind(0) - qkv_7 = None - q_7 = unbind_7[0] - k_7 = unbind_7[1] - v_7 = unbind_7[2] - unbind_7 = None - x_345 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) - q_7 = k_7 = v_7 = None - transpose_8 = x_345.transpose(1, 2) - x_345 = None - x_346 = transpose_8.reshape(1, pow_52, 384) - transpose_8 = None - x_347 = torch._C._nn.linear( - x_346, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_346 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) - x_347 = None - x_349 = mul_51 + x_348 - mul_51 = x_348 = None - view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_52 = x_349 * view_15 - view_15 = None - layer_norm_48 = torch.nn.functional.layer_norm( - x_349, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_349 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) - x_350 = torch._C._nn.linear( - layer_norm_48, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_48 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_37 = torch.nn.functional.relu(x_350, inplace=False) - x_350 = None - pow_38 = relu_37**2 - relu_37 = None - mul_53 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - * pow_38 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( - pow_38 - ) = None - x_351 = ( - mul_53 - + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - ) - mul_53 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) - x_352 = torch.nn.functional.dropout(x_351, 0.0, False, False) - x_351 = None - x_353 = torch._C._nn.linear( - x_352, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_354 = torch.nn.functional.dropout(x_353, 0.0, False, False) - x_353 = None - x_355 = mul_52 + x_354 - mul_52 = x_354 = None - view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_54 = x_355 * view_16 - view_16 = None - layer_norm_49 = torch.nn.functional.layer_norm( - x_355, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) - linear_32 = torch._C._nn.linear( - layer_norm_49, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_49 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_17 = linear_32.reshape(1, pow_52, 3, 12, 32) - linear_32 = None - qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) - reshape_17 = None - unbind_8 = qkv_8.unbind(0) - qkv_8 = None - q_8 = unbind_8[0] - k_8 = unbind_8[1] - v_8 = unbind_8[2] - unbind_8 = None - x_356 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) - q_8 = k_8 = v_8 = None - transpose_9 = x_356.transpose(1, 2) - x_356 = None - x_357 = transpose_9.reshape(1, pow_52, 384) - transpose_9 = None - x_358 = torch._C._nn.linear( - x_357, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_357 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_359 = torch.nn.functional.dropout(x_358, 0.0, False, False) - x_358 = None - x_360 = mul_54 + x_359 - mul_54 = x_359 = None - view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_55 = x_360 * view_17 - view_17 = None - layer_norm_50 = torch.nn.functional.layer_norm( - x_360, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_360 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) - x_361 = torch._C._nn.linear( - layer_norm_50, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_50 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_38 = torch.nn.functional.relu(x_361, inplace=False) - x_361 = None - pow_39 = relu_38**2 - relu_38 = None - mul_56 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - * pow_39 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( - pow_39 - ) = None - x_362 = ( - mul_56 - + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - ) - mul_56 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) - x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) - x_362 = None - x_364 = torch._C._nn.linear( - x_363, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_365 = torch.nn.functional.dropout(x_364, 0.0, False, False) - x_364 = None - x_366 = mul_55 + x_365 - mul_55 = x_365 = None - view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_57 = x_366 * view_18 - view_18 = None - layer_norm_51 = torch.nn.functional.layer_norm( - x_366, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) - linear_36 = torch._C._nn.linear( - layer_norm_51, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_51 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_19 = linear_36.reshape(1, pow_52, 3, 12, 32) - linear_36 = None - qkv_9 = reshape_19.permute(2, 0, 3, 1, 4) - reshape_19 = None - unbind_9 = qkv_9.unbind(0) - qkv_9 = None - q_9 = unbind_9[0] - k_9 = unbind_9[1] - v_9 = unbind_9[2] - unbind_9 = None - x_367 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) - q_9 = k_9 = v_9 = None - transpose_10 = x_367.transpose(1, 2) - x_367 = None - x_368 = transpose_10.reshape(1, pow_52, 384) - transpose_10 = None - x_369 = torch._C._nn.linear( - x_368, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_368 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_370 = torch.nn.functional.dropout(x_369, 0.0, False, False) - x_369 = None - x_371 = mul_57 + x_370 - mul_57 = x_370 = None - view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_58 = x_371 * view_19 - view_19 = None - layer_norm_52 = torch.nn.functional.layer_norm( - x_371, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_371 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) - x_372 = torch._C._nn.linear( - layer_norm_52, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_52 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_39 = torch.nn.functional.relu(x_372, inplace=False) - x_372 = None - pow_40 = relu_39**2 - relu_39 = None - mul_59 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - * pow_40 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( - pow_40 - ) = None - x_373 = ( - mul_59 - + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - ) - mul_59 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) - x_374 = torch.nn.functional.dropout(x_373, 0.0, False, False) - x_373 = None - x_375 = torch._C._nn.linear( - x_374, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_374 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_376 = torch.nn.functional.dropout(x_375, 0.0, False, False) - x_375 = None - x_377 = mul_58 + x_376 - mul_58 = x_376 = None - view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_60 = x_377 * view_20 - view_20 = None - layer_norm_53 = torch.nn.functional.layer_norm( - x_377, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_377 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) - linear_40 = torch._C._nn.linear( - layer_norm_53, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_53 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_21 = linear_40.reshape(1, pow_52, 3, 12, 32) - linear_40 = None - qkv_10 = reshape_21.permute(2, 0, 3, 1, 4) - reshape_21 = None - unbind_10 = qkv_10.unbind(0) - qkv_10 = None - q_10 = unbind_10[0] - k_10 = unbind_10[1] - v_10 = unbind_10[2] - unbind_10 = None - x_378 = torch._C._nn.scaled_dot_product_attention( - q_10, k_10, v_10, dropout_p=0.0 - ) - q_10 = k_10 = v_10 = None - transpose_11 = x_378.transpose(1, 2) - x_378 = None - x_379 = transpose_11.reshape(1, pow_52, 384) - transpose_11 = None - x_380 = torch._C._nn.linear( - x_379, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_379 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_381 = torch.nn.functional.dropout(x_380, 0.0, False, False) - x_380 = None - x_382 = mul_60 + x_381 - mul_60 = x_381 = None - view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_61 = x_382 * view_21 - view_21 = None - layer_norm_54 = torch.nn.functional.layer_norm( - x_382, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_382 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) - x_383 = torch._C._nn.linear( - layer_norm_54, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_54 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_40 = torch.nn.functional.relu(x_383, inplace=False) - x_383 = None - pow_41 = relu_40**2 - relu_40 = None - mul_62 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - * pow_41 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( - pow_41 - ) = None - x_384 = ( - mul_62 - + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - ) - mul_62 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) - x_385 = torch.nn.functional.dropout(x_384, 0.0, False, False) - x_384 = None - x_386 = torch._C._nn.linear( - x_385, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_385 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_387 = torch.nn.functional.dropout(x_386, 0.0, False, False) - x_386 = None - x_388 = mul_61 + x_387 - mul_61 = x_387 = None - view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_63 = x_388 * view_22 - view_22 = None - layer_norm_55 = torch.nn.functional.layer_norm( - x_388, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_388 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) - linear_44 = torch._C._nn.linear( - layer_norm_55, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_55 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_23 = linear_44.reshape(1, pow_52, 3, 12, 32) - linear_44 = None - qkv_11 = reshape_23.permute(2, 0, 3, 1, 4) - reshape_23 = None - unbind_11 = qkv_11.unbind(0) - qkv_11 = None - q_11 = unbind_11[0] - k_11 = unbind_11[1] - v_11 = unbind_11[2] - unbind_11 = None - x_389 = torch._C._nn.scaled_dot_product_attention( - q_11, k_11, v_11, dropout_p=0.0 - ) - q_11 = k_11 = v_11 = None - transpose_12 = x_389.transpose(1, 2) - x_389 = None - x_390 = transpose_12.reshape(1, pow_52, 384) - transpose_12 = None - x_391 = torch._C._nn.linear( - x_390, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_390 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_392 = torch.nn.functional.dropout(x_391, 0.0, False, False) - x_391 = None - x_393 = mul_63 + x_392 - mul_63 = x_392 = None - view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_64 = x_393 * view_23 - view_23 = None - layer_norm_56 = torch.nn.functional.layer_norm( - x_393, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_393 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) - x_394 = torch._C._nn.linear( - layer_norm_56, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_56 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_41 = torch.nn.functional.relu(x_394, inplace=False) - x_394 = None - pow_42 = relu_41**2 - relu_41 = None - mul_65 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - * pow_42 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( - pow_42 - ) = None - x_395 = ( - mul_65 - + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - ) - mul_65 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) - x_396 = torch.nn.functional.dropout(x_395, 0.0, False, False) - x_395 = None - x_397 = torch._C._nn.linear( - x_396, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_396 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_398 = torch.nn.functional.dropout(x_397, 0.0, False, False) - x_397 = None - x_399 = mul_64 + x_398 - mul_64 = x_398 = None - view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_66 = x_399 * view_24 - view_24 = None - layer_norm_57 = torch.nn.functional.layer_norm( - x_399, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_399 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = (None) - linear_48 = torch._C._nn.linear( - layer_norm_57, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_57 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_25 = linear_48.reshape(1, pow_52, 3, 12, 32) - linear_48 = None - qkv_12 = reshape_25.permute(2, 0, 3, 1, 4) - reshape_25 = None - unbind_12 = qkv_12.unbind(0) - qkv_12 = None - q_12 = unbind_12[0] - k_12 = unbind_12[1] - v_12 = unbind_12[2] - unbind_12 = None - x_400 = torch._C._nn.scaled_dot_product_attention( - q_12, k_12, v_12, dropout_p=0.0 - ) - q_12 = k_12 = v_12 = None - transpose_13 = x_400.transpose(1, 2) - x_400 = None - x_401 = transpose_13.reshape(1, pow_52, 384) - transpose_13 = None - x_402 = torch._C._nn.linear( - x_401, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_401 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_403 = torch.nn.functional.dropout(x_402, 0.0, False, False) - x_402 = None - x_404 = mul_66 + x_403 - mul_66 = x_403 = None - view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_67 = x_404 * view_25 - view_25 = None - layer_norm_58 = torch.nn.functional.layer_norm( - x_404, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_404 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = (None) - x_405 = torch._C._nn.linear( - layer_norm_58, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_58 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_42 = torch.nn.functional.relu(x_405, inplace=False) - x_405 = None - pow_43 = relu_42**2 - relu_42 = None - mul_68 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ - * pow_43 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = ( - pow_43 - ) = None - x_406 = ( - mul_68 - + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ - ) - mul_68 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = (None) - x_407 = torch.nn.functional.dropout(x_406, 0.0, False, False) - x_406 = None - x_408 = torch._C._nn.linear( - x_407, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_407 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_409 = torch.nn.functional.dropout(x_408, 0.0, False, False) - x_408 = None - x_410 = mul_67 + x_409 - mul_67 = x_409 = None - view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_69 = x_410 * view_26 - view_26 = None - layer_norm_59 = torch.nn.functional.layer_norm( - x_410, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_410 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = (None) - linear_52 = torch._C._nn.linear( - layer_norm_59, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_59 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_27 = linear_52.reshape(1, pow_52, 3, 12, 32) - linear_52 = None - qkv_13 = reshape_27.permute(2, 0, 3, 1, 4) - reshape_27 = None - unbind_13 = qkv_13.unbind(0) - qkv_13 = None - q_13 = unbind_13[0] - k_13 = unbind_13[1] - v_13 = unbind_13[2] - unbind_13 = None - x_411 = torch._C._nn.scaled_dot_product_attention( - q_13, k_13, v_13, dropout_p=0.0 - ) - q_13 = k_13 = v_13 = None - transpose_14 = x_411.transpose(1, 2) - x_411 = None - x_412 = transpose_14.reshape(1, pow_52, 384) - transpose_14 = None - x_413 = torch._C._nn.linear( - x_412, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_412 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_414 = torch.nn.functional.dropout(x_413, 0.0, False, False) - x_413 = None - x_415 = mul_69 + x_414 - mul_69 = x_414 = None - view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_70 = x_415 * view_27 - view_27 = None - layer_norm_60 = torch.nn.functional.layer_norm( - x_415, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_415 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = (None) - x_416 = torch._C._nn.linear( - layer_norm_60, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_60 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_43 = torch.nn.functional.relu(x_416, inplace=False) - x_416 = None - pow_44 = relu_43**2 - relu_43 = None - mul_71 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ - * pow_44 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = ( - pow_44 - ) = None - x_417 = ( - mul_71 - + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ - ) - mul_71 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = (None) - x_418 = torch.nn.functional.dropout(x_417, 0.0, False, False) - x_417 = None - x_419 = torch._C._nn.linear( - x_418, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_418 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_420 = torch.nn.functional.dropout(x_419, 0.0, False, False) - x_419 = None - x_421 = mul_70 + x_420 - mul_70 = x_420 = None - view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_72 = x_421 * view_28 - view_28 = None - layer_norm_61 = torch.nn.functional.layer_norm( - x_421, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_421 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = (None) - linear_56 = torch._C._nn.linear( - layer_norm_61, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_61 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_29 = linear_56.reshape(1, pow_52, 3, 12, 32) - linear_56 = None - qkv_14 = reshape_29.permute(2, 0, 3, 1, 4) - reshape_29 = None - unbind_14 = qkv_14.unbind(0) - qkv_14 = None - q_14 = unbind_14[0] - k_14 = unbind_14[1] - v_14 = unbind_14[2] - unbind_14 = None - x_422 = torch._C._nn.scaled_dot_product_attention( - q_14, k_14, v_14, dropout_p=0.0 - ) - q_14 = k_14 = v_14 = None - transpose_15 = x_422.transpose(1, 2) - x_422 = None - x_423 = transpose_15.reshape(1, pow_52, 384) - transpose_15 = None - x_424 = torch._C._nn.linear( - x_423, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_423 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_425 = torch.nn.functional.dropout(x_424, 0.0, False, False) - x_424 = None - x_426 = mul_72 + x_425 - mul_72 = x_425 = None - view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_73 = x_426 * view_29 - view_29 = None - layer_norm_62 = torch.nn.functional.layer_norm( - x_426, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_426 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = (None) - x_427 = torch._C._nn.linear( - layer_norm_62, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_62 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_44 = torch.nn.functional.relu(x_427, inplace=False) - x_427 = None - pow_45 = relu_44**2 - relu_44 = None - mul_74 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ - * pow_45 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = ( - pow_45 - ) = None - x_428 = ( - mul_74 - + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ - ) - mul_74 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = (None) - x_429 = torch.nn.functional.dropout(x_428, 0.0, False, False) - x_428 = None - x_430 = torch._C._nn.linear( - x_429, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_429 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_431 = torch.nn.functional.dropout(x_430, 0.0, False, False) - x_430 = None - x_432 = mul_73 + x_431 - mul_73 = x_431 = None - view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_75 = x_432 * view_30 - view_30 = None - layer_norm_63 = torch.nn.functional.layer_norm( - x_432, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_432 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = (None) - linear_60 = torch._C._nn.linear( - layer_norm_63, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_63 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_31 = linear_60.reshape(1, pow_52, 3, 12, 32) - linear_60 = None - qkv_15 = reshape_31.permute(2, 0, 3, 1, 4) - reshape_31 = None - unbind_15 = qkv_15.unbind(0) - qkv_15 = None - q_15 = unbind_15[0] - k_15 = unbind_15[1] - v_15 = unbind_15[2] - unbind_15 = None - x_433 = torch._C._nn.scaled_dot_product_attention( - q_15, k_15, v_15, dropout_p=0.0 - ) - q_15 = k_15 = v_15 = None - transpose_16 = x_433.transpose(1, 2) - x_433 = None - x_434 = transpose_16.reshape(1, pow_52, 384) - transpose_16 = None - x_435 = torch._C._nn.linear( - x_434, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_434 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_436 = torch.nn.functional.dropout(x_435, 0.0, False, False) - x_435 = None - x_437 = mul_75 + x_436 - mul_75 = x_436 = None - view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_76 = x_437 * view_31 - view_31 = None - layer_norm_64 = torch.nn.functional.layer_norm( - x_437, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_437 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = (None) - x_438 = torch._C._nn.linear( - layer_norm_64, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_64 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_45 = torch.nn.functional.relu(x_438, inplace=False) - x_438 = None - pow_46 = relu_45**2 - relu_45 = None - mul_77 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ - * pow_46 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = ( - pow_46 - ) = None - x_439 = ( - mul_77 - + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ - ) - mul_77 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = (None) - x_440 = torch.nn.functional.dropout(x_439, 0.0, False, False) - x_439 = None - x_441 = torch._C._nn.linear( - x_440, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_440 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_442 = torch.nn.functional.dropout(x_441, 0.0, False, False) - x_441 = None - x_443 = mul_76 + x_442 - mul_76 = x_442 = None - view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_78 = x_443 * view_32 - view_32 = None - layer_norm_65 = torch.nn.functional.layer_norm( - x_443, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_443 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = (None) - linear_64 = torch._C._nn.linear( - layer_norm_65, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_65 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_33 = linear_64.reshape(1, pow_52, 3, 12, 32) - linear_64 = None - qkv_16 = reshape_33.permute(2, 0, 3, 1, 4) - reshape_33 = None - unbind_16 = qkv_16.unbind(0) - qkv_16 = None - q_16 = unbind_16[0] - k_16 = unbind_16[1] - v_16 = unbind_16[2] - unbind_16 = None - x_444 = torch._C._nn.scaled_dot_product_attention( - q_16, k_16, v_16, dropout_p=0.0 - ) - q_16 = k_16 = v_16 = None - transpose_17 = x_444.transpose(1, 2) - x_444 = None - x_445 = transpose_17.reshape(1, pow_52, 384) - transpose_17 = None - x_446 = torch._C._nn.linear( - x_445, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_445 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_447 = torch.nn.functional.dropout(x_446, 0.0, False, False) - x_446 = None - x_448 = mul_78 + x_447 - mul_78 = x_447 = None - view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_79 = x_448 * view_33 - view_33 = None - layer_norm_66 = torch.nn.functional.layer_norm( - x_448, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_448 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = (None) - x_449 = torch._C._nn.linear( - layer_norm_66, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_66 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_46 = torch.nn.functional.relu(x_449, inplace=False) - x_449 = None - pow_47 = relu_46**2 - relu_46 = None - mul_80 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ - * pow_47 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = ( - pow_47 - ) = None - x_450 = ( - mul_80 - + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ - ) - mul_80 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = (None) - x_451 = torch.nn.functional.dropout(x_450, 0.0, False, False) - x_450 = None - x_452 = torch._C._nn.linear( - x_451, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_451 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_453 = torch.nn.functional.dropout(x_452, 0.0, False, False) - x_452 = None - x_454 = mul_79 + x_453 - mul_79 = x_453 = None - view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_81 = x_454 * view_34 - view_34 = None - layer_norm_67 = torch.nn.functional.layer_norm( - x_454, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_454 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = (None) - linear_68 = torch._C._nn.linear( - layer_norm_67, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_67 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_35 = linear_68.reshape(1, pow_52, 3, 12, 32) - linear_68 = None - qkv_17 = reshape_35.permute(2, 0, 3, 1, 4) - reshape_35 = None - unbind_17 = qkv_17.unbind(0) - qkv_17 = None - q_17 = unbind_17[0] - k_17 = unbind_17[1] - v_17 = unbind_17[2] - unbind_17 = None - x_455 = torch._C._nn.scaled_dot_product_attention( - q_17, k_17, v_17, dropout_p=0.0 - ) - q_17 = k_17 = v_17 = None - transpose_18 = x_455.transpose(1, 2) - x_455 = None - x_456 = transpose_18.reshape(1, pow_52, 384) - transpose_18 = pow_52 = None - x_457 = torch._C._nn.linear( - x_456, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_456 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_458 = torch.nn.functional.dropout(x_457, 0.0, False, False) - x_457 = None - x_459 = mul_81 + x_458 - mul_81 = x_458 = None - view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_82 = x_459 * view_35 - view_35 = None - layer_norm_68 = torch.nn.functional.layer_norm( - x_459, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_459 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = (None) - x_460 = torch._C._nn.linear( - layer_norm_68, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_68 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_47 = torch.nn.functional.relu(x_460, inplace=False) - x_460 = None - pow_48 = relu_47**2 - relu_47 = None - mul_83 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ - * pow_48 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = ( - pow_48 - ) = None - x_461 = ( - mul_83 - + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ - ) - mul_83 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = (None) - x_462 = torch.nn.functional.dropout(x_461, 0.0, False, False) - x_461 = None - x_463 = torch._C._nn.linear( - x_462, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_462 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_464 = torch.nn.functional.dropout(x_463, 0.0, False, False) - x_463 = None - x_465 = mul_82 + x_464 - mul_82 = x_464 = None - transpose_19 = x_465.transpose(1, 2) - x_465 = None - x_466 = transpose_19.reshape(1, 384, sym_sum_3, sym_sum_3) - transpose_19 = sym_sum_3 = None - x_467 = x_466.permute(0, 2, 3, 1) - x_466 = None - x_468 = torch.nn.functional.layer_norm( - x_467, - (384,), - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_467 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) - x_469 = x_468.permute(0, 3, 1, 2) - x_468 = None - x_470 = torch.conv2d( - x_469, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_469 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_3 = sym_sum // 32 - sym_sum = None - sym_sum_4 = torch.sym_sum([1, floordiv_3]) - floordiv_3 = None - reshape_38 = x_470.reshape(1, 576, -1) - x_470 = None - x_471 = reshape_38.transpose(1, 2) - reshape_38 = None - view_36 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_84 = x_471 * view_36 - view_36 = None - layer_norm_70 = torch.nn.functional.layer_norm( - x_471, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_471 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_53 = sym_sum_4**2 - linear_72 = torch._C._nn.linear( - layer_norm_70, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_70 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_39 = linear_72.reshape(1, pow_53, 3, 18, 32) - linear_72 = None - qkv_18 = reshape_39.permute(2, 0, 3, 1, 4) - reshape_39 = None - unbind_18 = qkv_18.unbind(0) - qkv_18 = None - q_18 = unbind_18[0] - k_18 = unbind_18[1] - v_18 = unbind_18[2] - unbind_18 = None - x_472 = torch._C._nn.scaled_dot_product_attention( - q_18, k_18, v_18, dropout_p=0.0 - ) - q_18 = k_18 = v_18 = None - transpose_21 = x_472.transpose(1, 2) - x_472 = None - x_473 = transpose_21.reshape(1, pow_53, 576) - transpose_21 = None - x_474 = torch._C._nn.linear( - x_473, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_473 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_475 = torch.nn.functional.dropout(x_474, 0.0, False, False) - x_474 = None - x_476 = mul_84 + x_475 - mul_84 = x_475 = None - view_37 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_85 = x_476 * view_37 - view_37 = None - layer_norm_71 = torch.nn.functional.layer_norm( - x_476, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_476 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_477 = torch._C._nn.linear( - layer_norm_71, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_71 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_48 = torch.nn.functional.relu(x_477, inplace=False) - x_477 = None - pow_49 = relu_48**2 - relu_48 = None - mul_86 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_49 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_49 - ) = None - x_478 = ( - mul_86 - + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_86 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_479 = torch.nn.functional.dropout(x_478, 0.0, False, False) - x_478 = None - x_480 = torch._C._nn.linear( - x_479, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_479 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_481 = torch.nn.functional.dropout(x_480, 0.0, False, False) - x_480 = None - x_482 = mul_85 + x_481 - mul_85 = x_481 = None - view_38 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_87 = x_482 * view_38 - view_38 = None - layer_norm_72 = torch.nn.functional.layer_norm( - x_482, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_482 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_76 = torch._C._nn.linear( - layer_norm_72, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_72 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_41 = linear_76.reshape(1, pow_53, 3, 18, 32) - linear_76 = None - qkv_19 = reshape_41.permute(2, 0, 3, 1, 4) - reshape_41 = None - unbind_19 = qkv_19.unbind(0) - qkv_19 = None - q_19 = unbind_19[0] - k_19 = unbind_19[1] - v_19 = unbind_19[2] - unbind_19 = None - x_483 = torch._C._nn.scaled_dot_product_attention( - q_19, k_19, v_19, dropout_p=0.0 - ) - q_19 = k_19 = v_19 = None - transpose_22 = x_483.transpose(1, 2) - x_483 = None - x_484 = transpose_22.reshape(1, pow_53, 576) - transpose_22 = None - x_485 = torch._C._nn.linear( - x_484, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_484 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_486 = torch.nn.functional.dropout(x_485, 0.0, False, False) - x_485 = None - x_487 = mul_87 + x_486 - mul_87 = x_486 = None - view_39 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_88 = x_487 * view_39 - view_39 = None - layer_norm_73 = torch.nn.functional.layer_norm( - x_487, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_487 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_488 = torch._C._nn.linear( - layer_norm_73, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_73 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_49 = torch.nn.functional.relu(x_488, inplace=False) - x_488 = None - pow_50 = relu_49**2 - relu_49 = None - mul_89 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_50 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_50 - ) = None - x_489 = ( - mul_89 - + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_89 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_490 = torch.nn.functional.dropout(x_489, 0.0, False, False) - x_489 = None - x_491 = torch._C._nn.linear( - x_490, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_490 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_492 = torch.nn.functional.dropout(x_491, 0.0, False, False) - x_491 = None - x_493 = mul_88 + x_492 - mul_88 = x_492 = None - view_40 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_90 = x_493 * view_40 - view_40 = None - layer_norm_74 = torch.nn.functional.layer_norm( - x_493, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_493 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_80 = torch._C._nn.linear( - layer_norm_74, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_74 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_43 = linear_80.reshape(1, pow_53, 3, 18, 32) - linear_80 = None - qkv_20 = reshape_43.permute(2, 0, 3, 1, 4) - reshape_43 = None - unbind_20 = qkv_20.unbind(0) - qkv_20 = None - q_20 = unbind_20[0] - k_20 = unbind_20[1] - v_20 = unbind_20[2] - unbind_20 = None - x_494 = torch._C._nn.scaled_dot_product_attention( - q_20, k_20, v_20, dropout_p=0.0 - ) - q_20 = k_20 = v_20 = None - transpose_23 = x_494.transpose(1, 2) - x_494 = None - x_495 = transpose_23.reshape(1, pow_53, 576) - transpose_23 = pow_53 = None - x_496 = torch._C._nn.linear( - x_495, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_495 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_497 = torch.nn.functional.dropout(x_496, 0.0, False, False) - x_496 = None - x_498 = mul_90 + x_497 - mul_90 = x_497 = None - view_41 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_91 = x_498 * view_41 - view_41 = None - layer_norm_75 = torch.nn.functional.layer_norm( - x_498, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_498 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_499 = torch._C._nn.linear( - layer_norm_75, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_75 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_50 = torch.nn.functional.relu(x_499, inplace=False) - x_499 = None - pow_51 = relu_50**2 - relu_50 = None - mul_92 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_51 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_51 - ) = None - x_500 = ( - mul_92 - + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_92 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_501 = torch.nn.functional.dropout(x_500, 0.0, False, False) - x_500 = None - x_502 = torch._C._nn.linear( - x_501, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_501 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_503 = torch.nn.functional.dropout(x_502, 0.0, False, False) - x_502 = None - x_504 = mul_91 + x_503 - mul_91 = x_503 = None - transpose_24 = x_504.transpose(1, 2) - x_504 = None - x_505 = transpose_24.reshape(1, 576, sym_sum_4, sym_sum_4) - transpose_24 = sym_sum_4 = None - x_506 = torch.nn.functional.adaptive_avg_pool2d(x_505, 1) - x_505 = None - x_507 = x_506.permute(0, 2, 3, 1) - x_506 = None - x_508 = torch.nn.functional.layer_norm( - x_507, - (576,), - l_self_modules_head_modules_norm_parameters_weight_, - l_self_modules_head_modules_norm_parameters_bias_, - 1e-06, - ) - x_507 = ( - l_self_modules_head_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_norm_parameters_bias_ = None - x_509 = x_508.permute(0, 3, 1, 2) - x_508 = None - x_510 = x_509.flatten(1, -1) - x_509 = None - x_511 = torch.nn.functional.dropout(x_510, 0.0, False, False) - x_510 = None - x_512 = torch._C._nn.linear( - x_511, - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, - ) - x_511 = ( - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None - relu_51 = torch.nn.functional.relu(x_512, inplace=False) - x_512 = None - x_513 = torch.square(relu_51) - relu_51 = None - x_514 = torch.nn.functional.layer_norm( - x_513, - (2304,), - l_self_modules_head_modules_fc_modules_norm_parameters_weight_, - l_self_modules_head_modules_fc_modules_norm_parameters_bias_, - 1e-06, - ) - x_513 = ( - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None - x_515 = torch.nn.functional.dropout(x_514, 0.0, False, False) - x_514 = None - x_516 = torch._C._nn.linear( - x_515, - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, - ) - x_515 = ( - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None - return (x_516,) diff --git a/samples/timm/caformer_m36.sail_in22k_ft_in1k/weight_meta.py b/samples/timm/caformer_m36.sail_in22k_ft_in1k/weight_meta.py deleted file mode 100644 index aa8ca0fde6..0000000000 --- a/samples/timm/caformer_m36.sail_in22k_ft_in1k/weight_meta.py +++ /dev/null @@ -1,3968 +0,0 @@ -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: - name = "L_self_modules_stem_modules_conv_parameters_weight_" - shape = [96, 3, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: - name = "L_self_modules_stem_modules_conv_parameters_bias_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_s1: - name = "s1" - shape = [] - dtype = "torch.int64" - device = "cpu" - mean = None - std = None - data = [4] - - -class Program_weight_tensor_meta_L_x_: - name = "L_x_" - shape = [1, 3, 224, 224] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.228 - std = 1.286 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: - name = "L_self_modules_stem_modules_norm_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [192, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [192, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [96, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [384, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [96, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [192, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [192, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [96, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [384, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [96, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [192, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [192, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [96, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [384, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [96, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" - shape = [192, 96, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" - shape = [384, 192, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" - shape = [576, 384, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1728, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [576, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [2304, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [576, 2304] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1728, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [576, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [2304, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [576, 2304] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1728, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [576, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [2304, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [576, 2304] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_norm_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_norm_parameters_bias_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" - shape = [2304, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" - shape = [2304] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" - shape = [2304] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" - shape = [2304] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" - shape = [1000, 2304] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" - shape = [1000] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None diff --git a/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/graph_net.json b/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/graph_net.json deleted file mode 100644 index 1373fe3b58..0000000000 --- a/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/graph_net.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "framework": "torch", - "num_devices_required": 1, - "num_nodes_required": 1 -} \ No newline at end of file diff --git a/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/input_meta.py b/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/input_meta.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/input_tensor_constraints.py b/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/input_tensor_constraints.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/model.py b/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/model.py deleted file mode 100644 index dcce286c5c..0000000000 --- a/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/model.py +++ /dev/null @@ -1,4754 +0,0 @@ -import torch - - -class GraphModule(torch.nn.Module): - def forward( - self, - L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - s1: torch.SymInt, - L_x_: torch.Tensor, - L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - ): - l_self_modules_stem_modules_conv_parameters_weight_ = ( - L_self_modules_stem_modules_conv_parameters_weight_ - ) - l_self_modules_stem_modules_conv_parameters_bias_ = ( - L_self_modules_stem_modules_conv_parameters_bias_ - ) - l_x_ = L_x_ - l_self_modules_stem_modules_norm_parameters_weight_ = ( - L_self_modules_stem_modules_norm_parameters_weight_ - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_head_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ - ) - x = torch.conv2d( - l_x_, - l_self_modules_stem_modules_conv_parameters_weight_, - l_self_modules_stem_modules_conv_parameters_bias_, - (4, 4), - (2, 2), - (1, 1), - 1, - ) - l_x_ = ( - l_self_modules_stem_modules_conv_parameters_weight_ - ) = l_self_modules_stem_modules_conv_parameters_bias_ = None - x_1 = x.permute(0, 2, 3, 1) - x = None - x_2 = torch.nn.functional.layer_norm( - x_1, (96,), l_self_modules_stem_modules_norm_parameters_weight_, None, 1e-06 - ) - x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None - x_3 = x_2.permute(0, 3, 1, 2) - x_2 = None - sym_sum = torch.sym_sum([-3, s1]) - s1 = None - floordiv = sym_sum // 4 - sym_sum_1 = torch.sym_sum([1, floordiv]) - floordiv = sym_sum_1 = None - x_4 = x_3.permute(0, 2, 3, 1) - x_5 = torch.nn.functional.layer_norm( - x_4, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_6 = x_5.permute(0, 3, 1, 2) - x_5 = None - x_7 = torch.conv2d( - x_6, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu = torch.nn.functional.relu(x_7, inplace=False) - x_7 = None - pow_1 = relu**2 - relu = None - mul = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_1 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_1 - ) = None - x_8 = ( - mul - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_9 = torch.conv2d( - x_8, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 192, - ) - x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_10 = torch.conv2d( - x_9, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_11 = x_3 + x_10 - x_3 = x_10 = None - x_12 = x_11.permute(0, 2, 3, 1) - x_13 = torch.nn.functional.layer_norm( - x_12, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_14 = x_13.permute(0, 3, 1, 2) - x_13 = None - x_15 = torch.conv2d( - x_14, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_1 = torch.nn.functional.relu(x_15, inplace=False) - x_15 = None - pow_2 = relu_1**2 - relu_1 = None - mul_1 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_2 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_2 - ) = None - x_16 = ( - mul_1 - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) - x_16 = None - x_18 = torch.conv2d( - x_17, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) - x_18 = None - x_20 = x_11 + x_19 - x_11 = x_19 = None - x_21 = x_20.permute(0, 2, 3, 1) - x_22 = torch.nn.functional.layer_norm( - x_21, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_23 = x_22.permute(0, 3, 1, 2) - x_22 = None - x_24 = torch.conv2d( - x_23, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_2 = torch.nn.functional.relu(x_24, inplace=False) - x_24 = None - pow_3 = relu_2**2 - relu_2 = None - mul_2 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_3 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_3 - ) = None - x_25 = ( - mul_2 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_26 = torch.conv2d( - x_25, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 192, - ) - x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_27 = torch.conv2d( - x_26, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_28 = x_20 + x_27 - x_20 = x_27 = None - x_29 = x_28.permute(0, 2, 3, 1) - x_30 = torch.nn.functional.layer_norm( - x_29, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_31 = x_30.permute(0, 3, 1, 2) - x_30 = None - x_32 = torch.conv2d( - x_31, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_3 = torch.nn.functional.relu(x_32, inplace=False) - x_32 = None - pow_4 = relu_3**2 - relu_3 = None - mul_3 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_4 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_4 - ) = None - x_33 = ( - mul_3 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) - x_33 = None - x_35 = torch.conv2d( - x_34, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) - x_35 = None - x_37 = x_28 + x_36 - x_28 = x_36 = None - x_38 = x_37.permute(0, 2, 3, 1) - x_39 = torch.nn.functional.layer_norm( - x_38, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_40 = x_39.permute(0, 3, 1, 2) - x_39 = None - x_41 = torch.conv2d( - x_40, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_4 = torch.nn.functional.relu(x_41, inplace=False) - x_41 = None - pow_5 = relu_4**2 - relu_4 = None - mul_4 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_5 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_5 - ) = None - x_42 = ( - mul_4 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_43 = torch.conv2d( - x_42, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 192, - ) - x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_44 = torch.conv2d( - x_43, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_45 = x_37 + x_44 - x_37 = x_44 = None - x_46 = x_45.permute(0, 2, 3, 1) - x_47 = torch.nn.functional.layer_norm( - x_46, - (96,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_48 = x_47.permute(0, 3, 1, 2) - x_47 = None - x_49 = torch.conv2d( - x_48, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_5 = torch.nn.functional.relu(x_49, inplace=False) - x_49 = None - pow_6 = relu_5**2 - relu_5 = None - mul_5 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_6 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_6 - ) = None - x_50 = ( - mul_5 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) - x_50 = None - x_52 = torch.conv2d( - x_51, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) - x_52 = None - x_54 = x_45 + x_53 - x_45 = x_53 = None - x_55 = x_54.permute(0, 2, 3, 1) - x_54 = None - x_56 = torch.nn.functional.layer_norm( - x_55, - (96,), - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) - x_57 = x_56.permute(0, 3, 1, 2) - x_56 = None - x_58 = torch.conv2d( - x_57, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_1 = sym_sum // 8 - sym_sum_2 = torch.sym_sum([1, floordiv_1]) - floordiv_1 = sym_sum_2 = None - x_59 = x_58.permute(0, 2, 3, 1) - x_60 = torch.nn.functional.layer_norm( - x_59, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_61 = x_60.permute(0, 3, 1, 2) - x_60 = None - x_62 = torch.conv2d( - x_61, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_6 = torch.nn.functional.relu(x_62, inplace=False) - x_62 = None - pow_7 = relu_6**2 - relu_6 = None - mul_6 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_7 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_7 - ) = None - x_63 = ( - mul_6 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_64 = torch.conv2d( - x_63, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_65 = torch.conv2d( - x_64, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_66 = x_58 + x_65 - x_58 = x_65 = None - x_67 = x_66.permute(0, 2, 3, 1) - x_68 = torch.nn.functional.layer_norm( - x_67, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_69 = x_68.permute(0, 3, 1, 2) - x_68 = None - x_70 = torch.conv2d( - x_69, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_7 = torch.nn.functional.relu(x_70, inplace=False) - x_70 = None - pow_8 = relu_7**2 - relu_7 = None - mul_7 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_8 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_8 - ) = None - x_71 = ( - mul_7 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) - x_71 = None - x_73 = torch.conv2d( - x_72, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) - x_73 = None - x_75 = x_66 + x_74 - x_66 = x_74 = None - x_76 = x_75.permute(0, 2, 3, 1) - x_77 = torch.nn.functional.layer_norm( - x_76, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_78 = x_77.permute(0, 3, 1, 2) - x_77 = None - x_79 = torch.conv2d( - x_78, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_8 = torch.nn.functional.relu(x_79, inplace=False) - x_79 = None - pow_9 = relu_8**2 - relu_8 = None - mul_8 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_9 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_9 - ) = None - x_80 = ( - mul_8 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_81 = torch.conv2d( - x_80, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_82 = torch.conv2d( - x_81, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_83 = x_75 + x_82 - x_75 = x_82 = None - x_84 = x_83.permute(0, 2, 3, 1) - x_85 = torch.nn.functional.layer_norm( - x_84, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_86 = x_85.permute(0, 3, 1, 2) - x_85 = None - x_87 = torch.conv2d( - x_86, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_9 = torch.nn.functional.relu(x_87, inplace=False) - x_87 = None - pow_10 = relu_9**2 - relu_9 = None - mul_9 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_10 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_10 - ) = None - x_88 = ( - mul_9 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) - x_88 = None - x_90 = torch.conv2d( - x_89, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) - x_90 = None - x_92 = x_83 + x_91 - x_83 = x_91 = None - x_93 = x_92.permute(0, 2, 3, 1) - x_94 = torch.nn.functional.layer_norm( - x_93, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_95 = x_94.permute(0, 3, 1, 2) - x_94 = None - x_96 = torch.conv2d( - x_95, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_10 = torch.nn.functional.relu(x_96, inplace=False) - x_96 = None - pow_11 = relu_10**2 - relu_10 = None - mul_10 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_11 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_11 - ) = None - x_97 = ( - mul_10 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_98 = torch.conv2d( - x_97, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_99 = torch.conv2d( - x_98, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_100 = x_92 + x_99 - x_92 = x_99 = None - x_101 = x_100.permute(0, 2, 3, 1) - x_102 = torch.nn.functional.layer_norm( - x_101, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_103 = x_102.permute(0, 3, 1, 2) - x_102 = None - x_104 = torch.conv2d( - x_103, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_11 = torch.nn.functional.relu(x_104, inplace=False) - x_104 = None - pow_12 = relu_11**2 - relu_11 = None - mul_11 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_12 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_12 - ) = None - x_105 = ( - mul_11 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) - x_105 = None - x_107 = torch.conv2d( - x_106, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) - x_107 = None - x_109 = x_100 + x_108 - x_100 = x_108 = None - x_110 = x_109.permute(0, 2, 3, 1) - x_111 = torch.nn.functional.layer_norm( - x_110, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_110 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) - x_112 = x_111.permute(0, 3, 1, 2) - x_111 = None - x_113 = torch.conv2d( - x_112, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_112 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_12 = torch.nn.functional.relu(x_113, inplace=False) - x_113 = None - pow_13 = relu_12**2 - relu_12 = None - mul_12 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ - * pow_13 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_13 - ) = None - x_114 = ( - mul_12 - + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_12 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_115 = torch.conv2d( - x_114, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_114 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_116 = torch.conv2d( - x_115, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_115 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_117 = x_109 + x_116 - x_109 = x_116 = None - x_118 = x_117.permute(0, 2, 3, 1) - x_119 = torch.nn.functional.layer_norm( - x_118, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_118 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) - x_120 = x_119.permute(0, 3, 1, 2) - x_119 = None - x_121 = torch.conv2d( - x_120, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_120 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_13 = torch.nn.functional.relu(x_121, inplace=False) - x_121 = None - pow_14 = relu_13**2 - relu_13 = None - mul_13 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - * pow_14 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( - pow_14 - ) = None - x_122 = ( - mul_13 - + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - ) - mul_13 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) - x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) - x_122 = None - x_124 = torch.conv2d( - x_123, - l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_123 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) - x_124 = None - x_126 = x_117 + x_125 - x_117 = x_125 = None - x_127 = x_126.permute(0, 2, 3, 1) - x_128 = torch.nn.functional.layer_norm( - x_127, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_127 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) - x_129 = x_128.permute(0, 3, 1, 2) - x_128 = None - x_130 = torch.conv2d( - x_129, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_129 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_14 = torch.nn.functional.relu(x_130, inplace=False) - x_130 = None - pow_15 = relu_14**2 - relu_14 = None - mul_14 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ - * pow_15 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_15 - ) = None - x_131 = ( - mul_14 - + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_14 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_132 = torch.conv2d( - x_131, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_131 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_133 = torch.conv2d( - x_132, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_132 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_134 = x_126 + x_133 - x_126 = x_133 = None - x_135 = x_134.permute(0, 2, 3, 1) - x_136 = torch.nn.functional.layer_norm( - x_135, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_135 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) - x_137 = x_136.permute(0, 3, 1, 2) - x_136 = None - x_138 = torch.conv2d( - x_137, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_137 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_15 = torch.nn.functional.relu(x_138, inplace=False) - x_138 = None - pow_16 = relu_15**2 - relu_15 = None - mul_15 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - * pow_16 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( - pow_16 - ) = None - x_139 = ( - mul_15 - + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - ) - mul_15 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) - x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) - x_139 = None - x_141 = torch.conv2d( - x_140, - l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_140 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) - x_141 = None - x_143 = x_134 + x_142 - x_134 = x_142 = None - x_144 = x_143.permute(0, 2, 3, 1) - x_145 = torch.nn.functional.layer_norm( - x_144, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_144 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) - x_146 = x_145.permute(0, 3, 1, 2) - x_145 = None - x_147 = torch.conv2d( - x_146, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_146 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_16 = torch.nn.functional.relu(x_147, inplace=False) - x_147 = None - pow_17 = relu_16**2 - relu_16 = None - mul_16 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ - * pow_17 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_17 - ) = None - x_148 = ( - mul_16 - + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_16 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_149 = torch.conv2d( - x_148, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_148 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_150 = torch.conv2d( - x_149, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_149 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_151 = x_143 + x_150 - x_143 = x_150 = None - x_152 = x_151.permute(0, 2, 3, 1) - x_153 = torch.nn.functional.layer_norm( - x_152, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_152 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) - x_154 = x_153.permute(0, 3, 1, 2) - x_153 = None - x_155 = torch.conv2d( - x_154, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_154 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_17 = torch.nn.functional.relu(x_155, inplace=False) - x_155 = None - pow_18 = relu_17**2 - relu_17 = None - mul_17 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - * pow_18 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( - pow_18 - ) = None - x_156 = ( - mul_17 - + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - ) - mul_17 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) - x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) - x_156 = None - x_158 = torch.conv2d( - x_157, - l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_157 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) - x_158 = None - x_160 = x_151 + x_159 - x_151 = x_159 = None - x_161 = x_160.permute(0, 2, 3, 1) - x_162 = torch.nn.functional.layer_norm( - x_161, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_161 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) - x_163 = x_162.permute(0, 3, 1, 2) - x_162 = None - x_164 = torch.conv2d( - x_163, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_163 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_18 = torch.nn.functional.relu(x_164, inplace=False) - x_164 = None - pow_19 = relu_18**2 - relu_18 = None - mul_18 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ - * pow_19 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_19 - ) = None - x_165 = ( - mul_18 - + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_18 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_166 = torch.conv2d( - x_165, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_165 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_167 = torch.conv2d( - x_166, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_166 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_168 = x_160 + x_167 - x_160 = x_167 = None - x_169 = x_168.permute(0, 2, 3, 1) - x_170 = torch.nn.functional.layer_norm( - x_169, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_169 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) - x_171 = x_170.permute(0, 3, 1, 2) - x_170 = None - x_172 = torch.conv2d( - x_171, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_171 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_19 = torch.nn.functional.relu(x_172, inplace=False) - x_172 = None - pow_20 = relu_19**2 - relu_19 = None - mul_19 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - * pow_20 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( - pow_20 - ) = None - x_173 = ( - mul_19 - + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - ) - mul_19 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) - x_174 = torch.nn.functional.dropout(x_173, 0.0, False, False) - x_173 = None - x_175 = torch.conv2d( - x_174, - l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_174 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) - x_175 = None - x_177 = x_168 + x_176 - x_168 = x_176 = None - x_178 = x_177.permute(0, 2, 3, 1) - x_179 = torch.nn.functional.layer_norm( - x_178, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_178 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) - x_180 = x_179.permute(0, 3, 1, 2) - x_179 = None - x_181 = torch.conv2d( - x_180, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_180 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_20 = torch.nn.functional.relu(x_181, inplace=False) - x_181 = None - pow_21 = relu_20**2 - relu_20 = None - mul_20 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ - * pow_21 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_21 - ) = None - x_182 = ( - mul_20 - + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_20 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_183 = torch.conv2d( - x_182, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_182 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_184 = torch.conv2d( - x_183, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_183 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_185 = x_177 + x_184 - x_177 = x_184 = None - x_186 = x_185.permute(0, 2, 3, 1) - x_187 = torch.nn.functional.layer_norm( - x_186, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_186 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) - x_188 = x_187.permute(0, 3, 1, 2) - x_187 = None - x_189 = torch.conv2d( - x_188, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_188 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_21 = torch.nn.functional.relu(x_189, inplace=False) - x_189 = None - pow_22 = relu_21**2 - relu_21 = None - mul_21 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - * pow_22 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( - pow_22 - ) = None - x_190 = ( - mul_21 - + l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - ) - mul_21 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) - x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) - x_190 = None - x_192 = torch.conv2d( - x_191, - l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_191 = l_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) - x_192 = None - x_194 = x_185 + x_193 - x_185 = x_193 = None - x_195 = x_194.permute(0, 2, 3, 1) - x_196 = torch.nn.functional.layer_norm( - x_195, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_195 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) - x_197 = x_196.permute(0, 3, 1, 2) - x_196 = None - x_198 = torch.conv2d( - x_197, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_197 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_22 = torch.nn.functional.relu(x_198, inplace=False) - x_198 = None - pow_23 = relu_22**2 - relu_22 = None - mul_22 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ - * pow_23 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_23 - ) = None - x_199 = ( - mul_22 - + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_22 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_200 = torch.conv2d( - x_199, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_199 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_201 = torch.conv2d( - x_200, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_200 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_202 = x_194 + x_201 - x_194 = x_201 = None - x_203 = x_202.permute(0, 2, 3, 1) - x_204 = torch.nn.functional.layer_norm( - x_203, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_203 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) - x_205 = x_204.permute(0, 3, 1, 2) - x_204 = None - x_206 = torch.conv2d( - x_205, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_205 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_23 = torch.nn.functional.relu(x_206, inplace=False) - x_206 = None - pow_24 = relu_23**2 - relu_23 = None - mul_23 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - * pow_24 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( - pow_24 - ) = None - x_207 = ( - mul_23 - + l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - ) - mul_23 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) - x_208 = torch.nn.functional.dropout(x_207, 0.0, False, False) - x_207 = None - x_209 = torch.conv2d( - x_208, - l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_208 = l_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) - x_209 = None - x_211 = x_202 + x_210 - x_202 = x_210 = None - x_212 = x_211.permute(0, 2, 3, 1) - x_213 = torch.nn.functional.layer_norm( - x_212, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_212 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) - x_214 = x_213.permute(0, 3, 1, 2) - x_213 = None - x_215 = torch.conv2d( - x_214, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_214 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_24 = torch.nn.functional.relu(x_215, inplace=False) - x_215 = None - pow_25 = relu_24**2 - relu_24 = None - mul_24 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ - * pow_25 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_25 - ) = None - x_216 = ( - mul_24 - + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_24 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_217 = torch.conv2d( - x_216, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_216 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_218 = torch.conv2d( - x_217, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_217 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_219 = x_211 + x_218 - x_211 = x_218 = None - x_220 = x_219.permute(0, 2, 3, 1) - x_221 = torch.nn.functional.layer_norm( - x_220, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_220 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) - x_222 = x_221.permute(0, 3, 1, 2) - x_221 = None - x_223 = torch.conv2d( - x_222, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_222 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_25 = torch.nn.functional.relu(x_223, inplace=False) - x_223 = None - pow_26 = relu_25**2 - relu_25 = None - mul_25 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - * pow_26 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( - pow_26 - ) = None - x_224 = ( - mul_25 - + l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - ) - mul_25 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) - x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) - x_224 = None - x_226 = torch.conv2d( - x_225, - l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_225 = l_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) - x_226 = None - x_228 = x_219 + x_227 - x_219 = x_227 = None - x_229 = x_228.permute(0, 2, 3, 1) - x_230 = torch.nn.functional.layer_norm( - x_229, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_229 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) - x_231 = x_230.permute(0, 3, 1, 2) - x_230 = None - x_232 = torch.conv2d( - x_231, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_231 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_26 = torch.nn.functional.relu(x_232, inplace=False) - x_232 = None - pow_27 = relu_26**2 - relu_26 = None - mul_26 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ - * pow_27 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_27 - ) = None - x_233 = ( - mul_26 - + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_26 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_234 = torch.conv2d( - x_233, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_233 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_235 = torch.conv2d( - x_234, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_234 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_236 = x_228 + x_235 - x_228 = x_235 = None - x_237 = x_236.permute(0, 2, 3, 1) - x_238 = torch.nn.functional.layer_norm( - x_237, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_237 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) - x_239 = x_238.permute(0, 3, 1, 2) - x_238 = None - x_240 = torch.conv2d( - x_239, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_239 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_27 = torch.nn.functional.relu(x_240, inplace=False) - x_240 = None - pow_28 = relu_27**2 - relu_27 = None - mul_27 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - * pow_28 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( - pow_28 - ) = None - x_241 = ( - mul_27 - + l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - ) - mul_27 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) - x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) - x_241 = None - x_243 = torch.conv2d( - x_242, - l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_242 = l_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_244 = torch.nn.functional.dropout(x_243, 0.0, False, False) - x_243 = None - x_245 = x_236 + x_244 - x_236 = x_244 = None - x_246 = x_245.permute(0, 2, 3, 1) - x_247 = torch.nn.functional.layer_norm( - x_246, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_246 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) - x_248 = x_247.permute(0, 3, 1, 2) - x_247 = None - x_249 = torch.conv2d( - x_248, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_248 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_28 = torch.nn.functional.relu(x_249, inplace=False) - x_249 = None - pow_29 = relu_28**2 - relu_28 = None - mul_28 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ - * pow_29 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_29 - ) = None - x_250 = ( - mul_28 - + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_28 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_251 = torch.conv2d( - x_250, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 384, - ) - x_250 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_252 = torch.conv2d( - x_251, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_251 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_253 = x_245 + x_252 - x_245 = x_252 = None - x_254 = x_253.permute(0, 2, 3, 1) - x_255 = torch.nn.functional.layer_norm( - x_254, - (192,), - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_254 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) - x_256 = x_255.permute(0, 3, 1, 2) - x_255 = None - x_257 = torch.conv2d( - x_256, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_256 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_29 = torch.nn.functional.relu(x_257, inplace=False) - x_257 = None - pow_30 = relu_29**2 - relu_29 = None - mul_29 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - * pow_30 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( - pow_30 - ) = None - x_258 = ( - mul_29 - + l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - ) - mul_29 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) - x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) - x_258 = None - x_260 = torch.conv2d( - x_259, - l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_259 = l_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_261 = torch.nn.functional.dropout(x_260, 0.0, False, False) - x_260 = None - x_262 = x_253 + x_261 - x_253 = x_261 = None - x_263 = x_262.permute(0, 2, 3, 1) - x_262 = None - x_264 = torch.nn.functional.layer_norm( - x_263, - (192,), - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_263 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) - x_265 = x_264.permute(0, 3, 1, 2) - x_264 = None - x_266 = torch.conv2d( - x_265, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_265 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_2 = sym_sum // 16 - sym_sum_3 = torch.sym_sum([1, floordiv_2]) - floordiv_2 = None - reshape = x_266.reshape(1, 384, -1) - x_266 = None - x_267 = reshape.transpose(1, 2) - reshape = None - view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_30 = x_267 * view - view = None - layer_norm_33 = torch.nn.functional.layer_norm( - x_267, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_52 = sym_sum_3**2 - linear = torch._C._nn.linear( - layer_norm_33, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_33 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_1 = linear.reshape(1, pow_52, 3, 12, 32) - linear = None - qkv = reshape_1.permute(2, 0, 3, 1, 4) - reshape_1 = None - unbind = qkv.unbind(0) - qkv = None - q = unbind[0] - k = unbind[1] - v = unbind[2] - unbind = None - x_268 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) - q = k = v = None - transpose_1 = x_268.transpose(1, 2) - x_268 = None - x_269 = transpose_1.reshape(1, pow_52, 384) - transpose_1 = None - x_270 = torch._C._nn.linear( - x_269, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_269 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_271 = torch.nn.functional.dropout(x_270, 0.0, False, False) - x_270 = None - x_272 = mul_30 + x_271 - mul_30 = x_271 = None - view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_31 = x_272 * view_1 - view_1 = None - layer_norm_34 = torch.nn.functional.layer_norm( - x_272, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_272 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_273 = torch._C._nn.linear( - layer_norm_34, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_34 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_30 = torch.nn.functional.relu(x_273, inplace=False) - x_273 = None - pow_31 = relu_30**2 - relu_30 = None - mul_32 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_31 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_31 - ) = None - x_274 = ( - mul_32 - + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_275 = torch.nn.functional.dropout(x_274, 0.0, False, False) - x_274 = None - x_276 = torch._C._nn.linear( - x_275, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_277 = torch.nn.functional.dropout(x_276, 0.0, False, False) - x_276 = None - x_278 = mul_31 + x_277 - mul_31 = x_277 = None - view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_33 = x_278 * view_2 - view_2 = None - layer_norm_35 = torch.nn.functional.layer_norm( - x_278, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_4 = torch._C._nn.linear( - layer_norm_35, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_3 = linear_4.reshape(1, pow_52, 3, 12, 32) - linear_4 = None - qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) - reshape_3 = None - unbind_1 = qkv_1.unbind(0) - qkv_1 = None - q_1 = unbind_1[0] - k_1 = unbind_1[1] - v_1 = unbind_1[2] - unbind_1 = None - x_279 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) - q_1 = k_1 = v_1 = None - transpose_2 = x_279.transpose(1, 2) - x_279 = None - x_280 = transpose_2.reshape(1, pow_52, 384) - transpose_2 = None - x_281 = torch._C._nn.linear( - x_280, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_280 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) - x_281 = None - x_283 = mul_33 + x_282 - mul_33 = x_282 = None - view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_34 = x_283 * view_3 - view_3 = None - layer_norm_36 = torch.nn.functional.layer_norm( - x_283, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_283 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_284 = torch._C._nn.linear( - layer_norm_36, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_36 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_31 = torch.nn.functional.relu(x_284, inplace=False) - x_284 = None - pow_32 = relu_31**2 - relu_31 = None - mul_35 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_32 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_32 - ) = None - x_285 = ( - mul_35 - + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_286 = torch.nn.functional.dropout(x_285, 0.0, False, False) - x_285 = None - x_287 = torch._C._nn.linear( - x_286, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) - x_287 = None - x_289 = mul_34 + x_288 - mul_34 = x_288 = None - view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_36 = x_289 * view_4 - view_4 = None - layer_norm_37 = torch.nn.functional.layer_norm( - x_289, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_8 = torch._C._nn.linear( - layer_norm_37, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_37 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_5 = linear_8.reshape(1, pow_52, 3, 12, 32) - linear_8 = None - qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) - reshape_5 = None - unbind_2 = qkv_2.unbind(0) - qkv_2 = None - q_2 = unbind_2[0] - k_2 = unbind_2[1] - v_2 = unbind_2[2] - unbind_2 = None - x_290 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) - q_2 = k_2 = v_2 = None - transpose_3 = x_290.transpose(1, 2) - x_290 = None - x_291 = transpose_3.reshape(1, pow_52, 384) - transpose_3 = None - x_292 = torch._C._nn.linear( - x_291, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_291 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_293 = torch.nn.functional.dropout(x_292, 0.0, False, False) - x_292 = None - x_294 = mul_36 + x_293 - mul_36 = x_293 = None - view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_37 = x_294 * view_5 - view_5 = None - layer_norm_38 = torch.nn.functional.layer_norm( - x_294, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_294 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_295 = torch._C._nn.linear( - layer_norm_38, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_32 = torch.nn.functional.relu(x_295, inplace=False) - x_295 = None - pow_33 = relu_32**2 - relu_32 = None - mul_38 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_33 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_33 - ) = None - x_296 = ( - mul_38 - + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_297 = torch.nn.functional.dropout(x_296, 0.0, False, False) - x_296 = None - x_298 = torch._C._nn.linear( - x_297, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_299 = torch.nn.functional.dropout(x_298, 0.0, False, False) - x_298 = None - x_300 = mul_37 + x_299 - mul_37 = x_299 = None - view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_39 = x_300 * view_6 - view_6 = None - layer_norm_39 = torch.nn.functional.layer_norm( - x_300, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) - linear_12 = torch._C._nn.linear( - layer_norm_39, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_39 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_7 = linear_12.reshape(1, pow_52, 3, 12, 32) - linear_12 = None - qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) - reshape_7 = None - unbind_3 = qkv_3.unbind(0) - qkv_3 = None - q_3 = unbind_3[0] - k_3 = unbind_3[1] - v_3 = unbind_3[2] - unbind_3 = None - x_301 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) - q_3 = k_3 = v_3 = None - transpose_4 = x_301.transpose(1, 2) - x_301 = None - x_302 = transpose_4.reshape(1, pow_52, 384) - transpose_4 = None - x_303 = torch._C._nn.linear( - x_302, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_302 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_304 = torch.nn.functional.dropout(x_303, 0.0, False, False) - x_303 = None - x_305 = mul_39 + x_304 - mul_39 = x_304 = None - view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_40 = x_305 * view_7 - view_7 = None - layer_norm_40 = torch.nn.functional.layer_norm( - x_305, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_305 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) - x_306 = torch._C._nn.linear( - layer_norm_40, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_40 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_33 = torch.nn.functional.relu(x_306, inplace=False) - x_306 = None - pow_34 = relu_33**2 - relu_33 = None - mul_41 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - * pow_34 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( - pow_34 - ) = None - x_307 = ( - mul_41 - + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - ) - mul_41 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) - x_308 = torch.nn.functional.dropout(x_307, 0.0, False, False) - x_307 = None - x_309 = torch._C._nn.linear( - x_308, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) - x_309 = None - x_311 = mul_40 + x_310 - mul_40 = x_310 = None - view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_42 = x_311 * view_8 - view_8 = None - layer_norm_41 = torch.nn.functional.layer_norm( - x_311, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) - linear_16 = torch._C._nn.linear( - layer_norm_41, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_41 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_9 = linear_16.reshape(1, pow_52, 3, 12, 32) - linear_16 = None - qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) - reshape_9 = None - unbind_4 = qkv_4.unbind(0) - qkv_4 = None - q_4 = unbind_4[0] - k_4 = unbind_4[1] - v_4 = unbind_4[2] - unbind_4 = None - x_312 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) - q_4 = k_4 = v_4 = None - transpose_5 = x_312.transpose(1, 2) - x_312 = None - x_313 = transpose_5.reshape(1, pow_52, 384) - transpose_5 = None - x_314 = torch._C._nn.linear( - x_313, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_313 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) - x_314 = None - x_316 = mul_42 + x_315 - mul_42 = x_315 = None - view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_43 = x_316 * view_9 - view_9 = None - layer_norm_42 = torch.nn.functional.layer_norm( - x_316, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_316 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) - x_317 = torch._C._nn.linear( - layer_norm_42, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_42 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_34 = torch.nn.functional.relu(x_317, inplace=False) - x_317 = None - pow_35 = relu_34**2 - relu_34 = None - mul_44 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - * pow_35 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( - pow_35 - ) = None - x_318 = ( - mul_44 - + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - ) - mul_44 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) - x_319 = torch.nn.functional.dropout(x_318, 0.0, False, False) - x_318 = None - x_320 = torch._C._nn.linear( - x_319, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_321 = torch.nn.functional.dropout(x_320, 0.0, False, False) - x_320 = None - x_322 = mul_43 + x_321 - mul_43 = x_321 = None - view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_45 = x_322 * view_10 - view_10 = None - layer_norm_43 = torch.nn.functional.layer_norm( - x_322, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) - linear_20 = torch._C._nn.linear( - layer_norm_43, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_43 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_11 = linear_20.reshape(1, pow_52, 3, 12, 32) - linear_20 = None - qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) - reshape_11 = None - unbind_5 = qkv_5.unbind(0) - qkv_5 = None - q_5 = unbind_5[0] - k_5 = unbind_5[1] - v_5 = unbind_5[2] - unbind_5 = None - x_323 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) - q_5 = k_5 = v_5 = None - transpose_6 = x_323.transpose(1, 2) - x_323 = None - x_324 = transpose_6.reshape(1, pow_52, 384) - transpose_6 = None - x_325 = torch._C._nn.linear( - x_324, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_324 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_326 = torch.nn.functional.dropout(x_325, 0.0, False, False) - x_325 = None - x_327 = mul_45 + x_326 - mul_45 = x_326 = None - view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_46 = x_327 * view_11 - view_11 = None - layer_norm_44 = torch.nn.functional.layer_norm( - x_327, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_327 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) - x_328 = torch._C._nn.linear( - layer_norm_44, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_44 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_35 = torch.nn.functional.relu(x_328, inplace=False) - x_328 = None - pow_36 = relu_35**2 - relu_35 = None - mul_47 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - * pow_36 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( - pow_36 - ) = None - x_329 = ( - mul_47 - + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - ) - mul_47 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) - x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) - x_329 = None - x_331 = torch._C._nn.linear( - x_330, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_332 = torch.nn.functional.dropout(x_331, 0.0, False, False) - x_331 = None - x_333 = mul_46 + x_332 - mul_46 = x_332 = None - view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_48 = x_333 * view_12 - view_12 = None - layer_norm_45 = torch.nn.functional.layer_norm( - x_333, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) - linear_24 = torch._C._nn.linear( - layer_norm_45, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_45 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_13 = linear_24.reshape(1, pow_52, 3, 12, 32) - linear_24 = None - qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) - reshape_13 = None - unbind_6 = qkv_6.unbind(0) - qkv_6 = None - q_6 = unbind_6[0] - k_6 = unbind_6[1] - v_6 = unbind_6[2] - unbind_6 = None - x_334 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) - q_6 = k_6 = v_6 = None - transpose_7 = x_334.transpose(1, 2) - x_334 = None - x_335 = transpose_7.reshape(1, pow_52, 384) - transpose_7 = None - x_336 = torch._C._nn.linear( - x_335, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_335 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_337 = torch.nn.functional.dropout(x_336, 0.0, False, False) - x_336 = None - x_338 = mul_48 + x_337 - mul_48 = x_337 = None - view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_49 = x_338 * view_13 - view_13 = None - layer_norm_46 = torch.nn.functional.layer_norm( - x_338, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_338 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) - x_339 = torch._C._nn.linear( - layer_norm_46, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_46 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_36 = torch.nn.functional.relu(x_339, inplace=False) - x_339 = None - pow_37 = relu_36**2 - relu_36 = None - mul_50 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - * pow_37 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( - pow_37 - ) = None - x_340 = ( - mul_50 - + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - ) - mul_50 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) - x_341 = torch.nn.functional.dropout(x_340, 0.0, False, False) - x_340 = None - x_342 = torch._C._nn.linear( - x_341, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_343 = torch.nn.functional.dropout(x_342, 0.0, False, False) - x_342 = None - x_344 = mul_49 + x_343 - mul_49 = x_343 = None - view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_51 = x_344 * view_14 - view_14 = None - layer_norm_47 = torch.nn.functional.layer_norm( - x_344, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) - linear_28 = torch._C._nn.linear( - layer_norm_47, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_47 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_15 = linear_28.reshape(1, pow_52, 3, 12, 32) - linear_28 = None - qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) - reshape_15 = None - unbind_7 = qkv_7.unbind(0) - qkv_7 = None - q_7 = unbind_7[0] - k_7 = unbind_7[1] - v_7 = unbind_7[2] - unbind_7 = None - x_345 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) - q_7 = k_7 = v_7 = None - transpose_8 = x_345.transpose(1, 2) - x_345 = None - x_346 = transpose_8.reshape(1, pow_52, 384) - transpose_8 = None - x_347 = torch._C._nn.linear( - x_346, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_346 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) - x_347 = None - x_349 = mul_51 + x_348 - mul_51 = x_348 = None - view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_52 = x_349 * view_15 - view_15 = None - layer_norm_48 = torch.nn.functional.layer_norm( - x_349, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_349 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) - x_350 = torch._C._nn.linear( - layer_norm_48, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_48 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_37 = torch.nn.functional.relu(x_350, inplace=False) - x_350 = None - pow_38 = relu_37**2 - relu_37 = None - mul_53 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - * pow_38 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( - pow_38 - ) = None - x_351 = ( - mul_53 - + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - ) - mul_53 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) - x_352 = torch.nn.functional.dropout(x_351, 0.0, False, False) - x_351 = None - x_353 = torch._C._nn.linear( - x_352, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_354 = torch.nn.functional.dropout(x_353, 0.0, False, False) - x_353 = None - x_355 = mul_52 + x_354 - mul_52 = x_354 = None - view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_54 = x_355 * view_16 - view_16 = None - layer_norm_49 = torch.nn.functional.layer_norm( - x_355, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) - linear_32 = torch._C._nn.linear( - layer_norm_49, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_49 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_17 = linear_32.reshape(1, pow_52, 3, 12, 32) - linear_32 = None - qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) - reshape_17 = None - unbind_8 = qkv_8.unbind(0) - qkv_8 = None - q_8 = unbind_8[0] - k_8 = unbind_8[1] - v_8 = unbind_8[2] - unbind_8 = None - x_356 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) - q_8 = k_8 = v_8 = None - transpose_9 = x_356.transpose(1, 2) - x_356 = None - x_357 = transpose_9.reshape(1, pow_52, 384) - transpose_9 = None - x_358 = torch._C._nn.linear( - x_357, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_357 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_359 = torch.nn.functional.dropout(x_358, 0.0, False, False) - x_358 = None - x_360 = mul_54 + x_359 - mul_54 = x_359 = None - view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_55 = x_360 * view_17 - view_17 = None - layer_norm_50 = torch.nn.functional.layer_norm( - x_360, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_360 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) - x_361 = torch._C._nn.linear( - layer_norm_50, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_50 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_38 = torch.nn.functional.relu(x_361, inplace=False) - x_361 = None - pow_39 = relu_38**2 - relu_38 = None - mul_56 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - * pow_39 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( - pow_39 - ) = None - x_362 = ( - mul_56 - + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - ) - mul_56 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) - x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) - x_362 = None - x_364 = torch._C._nn.linear( - x_363, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_365 = torch.nn.functional.dropout(x_364, 0.0, False, False) - x_364 = None - x_366 = mul_55 + x_365 - mul_55 = x_365 = None - view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_57 = x_366 * view_18 - view_18 = None - layer_norm_51 = torch.nn.functional.layer_norm( - x_366, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_ = (None) - linear_36 = torch._C._nn.linear( - layer_norm_51, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_51 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_19 = linear_36.reshape(1, pow_52, 3, 12, 32) - linear_36 = None - qkv_9 = reshape_19.permute(2, 0, 3, 1, 4) - reshape_19 = None - unbind_9 = qkv_9.unbind(0) - qkv_9 = None - q_9 = unbind_9[0] - k_9 = unbind_9[1] - v_9 = unbind_9[2] - unbind_9 = None - x_367 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) - q_9 = k_9 = v_9 = None - transpose_10 = x_367.transpose(1, 2) - x_367 = None - x_368 = transpose_10.reshape(1, pow_52, 384) - transpose_10 = None - x_369 = torch._C._nn.linear( - x_368, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_368 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_370 = torch.nn.functional.dropout(x_369, 0.0, False, False) - x_369 = None - x_371 = mul_57 + x_370 - mul_57 = x_370 = None - view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_58 = x_371 * view_19 - view_19 = None - layer_norm_52 = torch.nn.functional.layer_norm( - x_371, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_371 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_ = (None) - x_372 = torch._C._nn.linear( - layer_norm_52, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_52 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_39 = torch.nn.functional.relu(x_372, inplace=False) - x_372 = None - pow_40 = relu_39**2 - relu_39 = None - mul_59 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ - * pow_40 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_ = ( - pow_40 - ) = None - x_373 = ( - mul_59 - + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ - ) - mul_59 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_ = (None) - x_374 = torch.nn.functional.dropout(x_373, 0.0, False, False) - x_373 = None - x_375 = torch._C._nn.linear( - x_374, - l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_374 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_376 = torch.nn.functional.dropout(x_375, 0.0, False, False) - x_375 = None - x_377 = mul_58 + x_376 - mul_58 = x_376 = None - view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_60 = x_377 * view_20 - view_20 = None - layer_norm_53 = torch.nn.functional.layer_norm( - x_377, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_377 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_ = (None) - linear_40 = torch._C._nn.linear( - layer_norm_53, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_53 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_21 = linear_40.reshape(1, pow_52, 3, 12, 32) - linear_40 = None - qkv_10 = reshape_21.permute(2, 0, 3, 1, 4) - reshape_21 = None - unbind_10 = qkv_10.unbind(0) - qkv_10 = None - q_10 = unbind_10[0] - k_10 = unbind_10[1] - v_10 = unbind_10[2] - unbind_10 = None - x_378 = torch._C._nn.scaled_dot_product_attention( - q_10, k_10, v_10, dropout_p=0.0 - ) - q_10 = k_10 = v_10 = None - transpose_11 = x_378.transpose(1, 2) - x_378 = None - x_379 = transpose_11.reshape(1, pow_52, 384) - transpose_11 = None - x_380 = torch._C._nn.linear( - x_379, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_379 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_381 = torch.nn.functional.dropout(x_380, 0.0, False, False) - x_380 = None - x_382 = mul_60 + x_381 - mul_60 = x_381 = None - view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_61 = x_382 * view_21 - view_21 = None - layer_norm_54 = torch.nn.functional.layer_norm( - x_382, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_382 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_ = (None) - x_383 = torch._C._nn.linear( - layer_norm_54, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_54 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_40 = torch.nn.functional.relu(x_383, inplace=False) - x_383 = None - pow_41 = relu_40**2 - relu_40 = None - mul_62 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ - * pow_41 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_ = ( - pow_41 - ) = None - x_384 = ( - mul_62 - + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ - ) - mul_62 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_ = (None) - x_385 = torch.nn.functional.dropout(x_384, 0.0, False, False) - x_384 = None - x_386 = torch._C._nn.linear( - x_385, - l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_385 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_387 = torch.nn.functional.dropout(x_386, 0.0, False, False) - x_386 = None - x_388 = mul_61 + x_387 - mul_61 = x_387 = None - view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_63 = x_388 * view_22 - view_22 = None - layer_norm_55 = torch.nn.functional.layer_norm( - x_388, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_388 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_ = (None) - linear_44 = torch._C._nn.linear( - layer_norm_55, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_55 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_23 = linear_44.reshape(1, pow_52, 3, 12, 32) - linear_44 = None - qkv_11 = reshape_23.permute(2, 0, 3, 1, 4) - reshape_23 = None - unbind_11 = qkv_11.unbind(0) - qkv_11 = None - q_11 = unbind_11[0] - k_11 = unbind_11[1] - v_11 = unbind_11[2] - unbind_11 = None - x_389 = torch._C._nn.scaled_dot_product_attention( - q_11, k_11, v_11, dropout_p=0.0 - ) - q_11 = k_11 = v_11 = None - transpose_12 = x_389.transpose(1, 2) - x_389 = None - x_390 = transpose_12.reshape(1, pow_52, 384) - transpose_12 = None - x_391 = torch._C._nn.linear( - x_390, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_390 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_392 = torch.nn.functional.dropout(x_391, 0.0, False, False) - x_391 = None - x_393 = mul_63 + x_392 - mul_63 = x_392 = None - view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_64 = x_393 * view_23 - view_23 = None - layer_norm_56 = torch.nn.functional.layer_norm( - x_393, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_393 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_ = (None) - x_394 = torch._C._nn.linear( - layer_norm_56, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_56 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_41 = torch.nn.functional.relu(x_394, inplace=False) - x_394 = None - pow_42 = relu_41**2 - relu_41 = None - mul_65 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ - * pow_42 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_ = ( - pow_42 - ) = None - x_395 = ( - mul_65 - + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ - ) - mul_65 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_ = (None) - x_396 = torch.nn.functional.dropout(x_395, 0.0, False, False) - x_395 = None - x_397 = torch._C._nn.linear( - x_396, - l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_396 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_398 = torch.nn.functional.dropout(x_397, 0.0, False, False) - x_397 = None - x_399 = mul_64 + x_398 - mul_64 = x_398 = None - view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_66 = x_399 * view_24 - view_24 = None - layer_norm_57 = torch.nn.functional.layer_norm( - x_399, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_399 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_ = (None) - linear_48 = torch._C._nn.linear( - layer_norm_57, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_57 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_25 = linear_48.reshape(1, pow_52, 3, 12, 32) - linear_48 = None - qkv_12 = reshape_25.permute(2, 0, 3, 1, 4) - reshape_25 = None - unbind_12 = qkv_12.unbind(0) - qkv_12 = None - q_12 = unbind_12[0] - k_12 = unbind_12[1] - v_12 = unbind_12[2] - unbind_12 = None - x_400 = torch._C._nn.scaled_dot_product_attention( - q_12, k_12, v_12, dropout_p=0.0 - ) - q_12 = k_12 = v_12 = None - transpose_13 = x_400.transpose(1, 2) - x_400 = None - x_401 = transpose_13.reshape(1, pow_52, 384) - transpose_13 = None - x_402 = torch._C._nn.linear( - x_401, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_401 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_403 = torch.nn.functional.dropout(x_402, 0.0, False, False) - x_402 = None - x_404 = mul_66 + x_403 - mul_66 = x_403 = None - view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_67 = x_404 * view_25 - view_25 = None - layer_norm_58 = torch.nn.functional.layer_norm( - x_404, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_404 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_ = (None) - x_405 = torch._C._nn.linear( - layer_norm_58, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_58 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_42 = torch.nn.functional.relu(x_405, inplace=False) - x_405 = None - pow_43 = relu_42**2 - relu_42 = None - mul_68 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ - * pow_43 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_ = ( - pow_43 - ) = None - x_406 = ( - mul_68 - + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ - ) - mul_68 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_ = (None) - x_407 = torch.nn.functional.dropout(x_406, 0.0, False, False) - x_406 = None - x_408 = torch._C._nn.linear( - x_407, - l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_407 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_409 = torch.nn.functional.dropout(x_408, 0.0, False, False) - x_408 = None - x_410 = mul_67 + x_409 - mul_67 = x_409 = None - view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_69 = x_410 * view_26 - view_26 = None - layer_norm_59 = torch.nn.functional.layer_norm( - x_410, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_410 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_ = (None) - linear_52 = torch._C._nn.linear( - layer_norm_59, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_59 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_27 = linear_52.reshape(1, pow_52, 3, 12, 32) - linear_52 = None - qkv_13 = reshape_27.permute(2, 0, 3, 1, 4) - reshape_27 = None - unbind_13 = qkv_13.unbind(0) - qkv_13 = None - q_13 = unbind_13[0] - k_13 = unbind_13[1] - v_13 = unbind_13[2] - unbind_13 = None - x_411 = torch._C._nn.scaled_dot_product_attention( - q_13, k_13, v_13, dropout_p=0.0 - ) - q_13 = k_13 = v_13 = None - transpose_14 = x_411.transpose(1, 2) - x_411 = None - x_412 = transpose_14.reshape(1, pow_52, 384) - transpose_14 = None - x_413 = torch._C._nn.linear( - x_412, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_412 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_414 = torch.nn.functional.dropout(x_413, 0.0, False, False) - x_413 = None - x_415 = mul_69 + x_414 - mul_69 = x_414 = None - view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_70 = x_415 * view_27 - view_27 = None - layer_norm_60 = torch.nn.functional.layer_norm( - x_415, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_415 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_ = (None) - x_416 = torch._C._nn.linear( - layer_norm_60, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_60 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_43 = torch.nn.functional.relu(x_416, inplace=False) - x_416 = None - pow_44 = relu_43**2 - relu_43 = None - mul_71 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ - * pow_44 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_ = ( - pow_44 - ) = None - x_417 = ( - mul_71 - + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ - ) - mul_71 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_ = (None) - x_418 = torch.nn.functional.dropout(x_417, 0.0, False, False) - x_417 = None - x_419 = torch._C._nn.linear( - x_418, - l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_418 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_420 = torch.nn.functional.dropout(x_419, 0.0, False, False) - x_419 = None - x_421 = mul_70 + x_420 - mul_70 = x_420 = None - view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_72 = x_421 * view_28 - view_28 = None - layer_norm_61 = torch.nn.functional.layer_norm( - x_421, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_421 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_ = (None) - linear_56 = torch._C._nn.linear( - layer_norm_61, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_61 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_29 = linear_56.reshape(1, pow_52, 3, 12, 32) - linear_56 = None - qkv_14 = reshape_29.permute(2, 0, 3, 1, 4) - reshape_29 = None - unbind_14 = qkv_14.unbind(0) - qkv_14 = None - q_14 = unbind_14[0] - k_14 = unbind_14[1] - v_14 = unbind_14[2] - unbind_14 = None - x_422 = torch._C._nn.scaled_dot_product_attention( - q_14, k_14, v_14, dropout_p=0.0 - ) - q_14 = k_14 = v_14 = None - transpose_15 = x_422.transpose(1, 2) - x_422 = None - x_423 = transpose_15.reshape(1, pow_52, 384) - transpose_15 = None - x_424 = torch._C._nn.linear( - x_423, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_423 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_425 = torch.nn.functional.dropout(x_424, 0.0, False, False) - x_424 = None - x_426 = mul_72 + x_425 - mul_72 = x_425 = None - view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_73 = x_426 * view_29 - view_29 = None - layer_norm_62 = torch.nn.functional.layer_norm( - x_426, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_426 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_ = (None) - x_427 = torch._C._nn.linear( - layer_norm_62, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_62 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_44 = torch.nn.functional.relu(x_427, inplace=False) - x_427 = None - pow_45 = relu_44**2 - relu_44 = None - mul_74 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ - * pow_45 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_ = ( - pow_45 - ) = None - x_428 = ( - mul_74 - + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ - ) - mul_74 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_ = (None) - x_429 = torch.nn.functional.dropout(x_428, 0.0, False, False) - x_428 = None - x_430 = torch._C._nn.linear( - x_429, - l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_429 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_431 = torch.nn.functional.dropout(x_430, 0.0, False, False) - x_430 = None - x_432 = mul_73 + x_431 - mul_73 = x_431 = None - view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_75 = x_432 * view_30 - view_30 = None - layer_norm_63 = torch.nn.functional.layer_norm( - x_432, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_432 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_ = (None) - linear_60 = torch._C._nn.linear( - layer_norm_63, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_63 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_31 = linear_60.reshape(1, pow_52, 3, 12, 32) - linear_60 = None - qkv_15 = reshape_31.permute(2, 0, 3, 1, 4) - reshape_31 = None - unbind_15 = qkv_15.unbind(0) - qkv_15 = None - q_15 = unbind_15[0] - k_15 = unbind_15[1] - v_15 = unbind_15[2] - unbind_15 = None - x_433 = torch._C._nn.scaled_dot_product_attention( - q_15, k_15, v_15, dropout_p=0.0 - ) - q_15 = k_15 = v_15 = None - transpose_16 = x_433.transpose(1, 2) - x_433 = None - x_434 = transpose_16.reshape(1, pow_52, 384) - transpose_16 = None - x_435 = torch._C._nn.linear( - x_434, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_434 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_436 = torch.nn.functional.dropout(x_435, 0.0, False, False) - x_435 = None - x_437 = mul_75 + x_436 - mul_75 = x_436 = None - view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_76 = x_437 * view_31 - view_31 = None - layer_norm_64 = torch.nn.functional.layer_norm( - x_437, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_437 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_ = (None) - x_438 = torch._C._nn.linear( - layer_norm_64, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_64 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_45 = torch.nn.functional.relu(x_438, inplace=False) - x_438 = None - pow_46 = relu_45**2 - relu_45 = None - mul_77 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ - * pow_46 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_ = ( - pow_46 - ) = None - x_439 = ( - mul_77 - + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ - ) - mul_77 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_ = (None) - x_440 = torch.nn.functional.dropout(x_439, 0.0, False, False) - x_439 = None - x_441 = torch._C._nn.linear( - x_440, - l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_440 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_442 = torch.nn.functional.dropout(x_441, 0.0, False, False) - x_441 = None - x_443 = mul_76 + x_442 - mul_76 = x_442 = None - view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_78 = x_443 * view_32 - view_32 = None - layer_norm_65 = torch.nn.functional.layer_norm( - x_443, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_443 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_ = (None) - linear_64 = torch._C._nn.linear( - layer_norm_65, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_65 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_33 = linear_64.reshape(1, pow_52, 3, 12, 32) - linear_64 = None - qkv_16 = reshape_33.permute(2, 0, 3, 1, 4) - reshape_33 = None - unbind_16 = qkv_16.unbind(0) - qkv_16 = None - q_16 = unbind_16[0] - k_16 = unbind_16[1] - v_16 = unbind_16[2] - unbind_16 = None - x_444 = torch._C._nn.scaled_dot_product_attention( - q_16, k_16, v_16, dropout_p=0.0 - ) - q_16 = k_16 = v_16 = None - transpose_17 = x_444.transpose(1, 2) - x_444 = None - x_445 = transpose_17.reshape(1, pow_52, 384) - transpose_17 = None - x_446 = torch._C._nn.linear( - x_445, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_445 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_447 = torch.nn.functional.dropout(x_446, 0.0, False, False) - x_446 = None - x_448 = mul_78 + x_447 - mul_78 = x_447 = None - view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_79 = x_448 * view_33 - view_33 = None - layer_norm_66 = torch.nn.functional.layer_norm( - x_448, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_448 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_ = (None) - x_449 = torch._C._nn.linear( - layer_norm_66, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_66 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_46 = torch.nn.functional.relu(x_449, inplace=False) - x_449 = None - pow_47 = relu_46**2 - relu_46 = None - mul_80 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ - * pow_47 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_ = ( - pow_47 - ) = None - x_450 = ( - mul_80 - + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ - ) - mul_80 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_ = (None) - x_451 = torch.nn.functional.dropout(x_450, 0.0, False, False) - x_450 = None - x_452 = torch._C._nn.linear( - x_451, - l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_451 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_453 = torch.nn.functional.dropout(x_452, 0.0, False, False) - x_452 = None - x_454 = mul_79 + x_453 - mul_79 = x_453 = None - view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_81 = x_454 * view_34 - view_34 = None - layer_norm_67 = torch.nn.functional.layer_norm( - x_454, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_454 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_ = (None) - linear_68 = torch._C._nn.linear( - layer_norm_67, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_67 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_35 = linear_68.reshape(1, pow_52, 3, 12, 32) - linear_68 = None - qkv_17 = reshape_35.permute(2, 0, 3, 1, 4) - reshape_35 = None - unbind_17 = qkv_17.unbind(0) - qkv_17 = None - q_17 = unbind_17[0] - k_17 = unbind_17[1] - v_17 = unbind_17[2] - unbind_17 = None - x_455 = torch._C._nn.scaled_dot_product_attention( - q_17, k_17, v_17, dropout_p=0.0 - ) - q_17 = k_17 = v_17 = None - transpose_18 = x_455.transpose(1, 2) - x_455 = None - x_456 = transpose_18.reshape(1, pow_52, 384) - transpose_18 = pow_52 = None - x_457 = torch._C._nn.linear( - x_456, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_456 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_458 = torch.nn.functional.dropout(x_457, 0.0, False, False) - x_457 = None - x_459 = mul_81 + x_458 - mul_81 = x_458 = None - view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_.view( - (384,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_82 = x_459 * view_35 - view_35 = None - layer_norm_68 = torch.nn.functional.layer_norm( - x_459, - (384,), - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_459 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_ = (None) - x_460 = torch._C._nn.linear( - layer_norm_68, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_68 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_47 = torch.nn.functional.relu(x_460, inplace=False) - x_460 = None - pow_48 = relu_47**2 - relu_47 = None - mul_83 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ - * pow_48 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_ = ( - pow_48 - ) = None - x_461 = ( - mul_83 - + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ - ) - mul_83 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_ = (None) - x_462 = torch.nn.functional.dropout(x_461, 0.0, False, False) - x_461 = None - x_463 = torch._C._nn.linear( - x_462, - l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_462 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_464 = torch.nn.functional.dropout(x_463, 0.0, False, False) - x_463 = None - x_465 = mul_82 + x_464 - mul_82 = x_464 = None - transpose_19 = x_465.transpose(1, 2) - x_465 = None - x_466 = transpose_19.reshape(1, 384, sym_sum_3, sym_sum_3) - transpose_19 = sym_sum_3 = None - x_467 = x_466.permute(0, 2, 3, 1) - x_466 = None - x_468 = torch.nn.functional.layer_norm( - x_467, - (384,), - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_467 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) - x_469 = x_468.permute(0, 3, 1, 2) - x_468 = None - x_470 = torch.conv2d( - x_469, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_469 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_3 = sym_sum // 32 - sym_sum = None - sym_sum_4 = torch.sym_sum([1, floordiv_3]) - floordiv_3 = None - reshape_38 = x_470.reshape(1, 576, -1) - x_470 = None - x_471 = reshape_38.transpose(1, 2) - reshape_38 = None - view_36 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_84 = x_471 * view_36 - view_36 = None - layer_norm_70 = torch.nn.functional.layer_norm( - x_471, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_471 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_53 = sym_sum_4**2 - linear_72 = torch._C._nn.linear( - layer_norm_70, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_70 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_39 = linear_72.reshape(1, pow_53, 3, 18, 32) - linear_72 = None - qkv_18 = reshape_39.permute(2, 0, 3, 1, 4) - reshape_39 = None - unbind_18 = qkv_18.unbind(0) - qkv_18 = None - q_18 = unbind_18[0] - k_18 = unbind_18[1] - v_18 = unbind_18[2] - unbind_18 = None - x_472 = torch._C._nn.scaled_dot_product_attention( - q_18, k_18, v_18, dropout_p=0.0 - ) - q_18 = k_18 = v_18 = None - transpose_21 = x_472.transpose(1, 2) - x_472 = None - x_473 = transpose_21.reshape(1, pow_53, 576) - transpose_21 = None - x_474 = torch._C._nn.linear( - x_473, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_473 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_475 = torch.nn.functional.dropout(x_474, 0.0, False, False) - x_474 = None - x_476 = mul_84 + x_475 - mul_84 = x_475 = None - view_37 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_85 = x_476 * view_37 - view_37 = None - layer_norm_71 = torch.nn.functional.layer_norm( - x_476, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_476 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_477 = torch._C._nn.linear( - layer_norm_71, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_71 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_48 = torch.nn.functional.relu(x_477, inplace=False) - x_477 = None - pow_49 = relu_48**2 - relu_48 = None - mul_86 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_49 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_49 - ) = None - x_478 = ( - mul_86 - + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_86 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_479 = torch.nn.functional.dropout(x_478, 0.0, False, False) - x_478 = None - x_480 = torch._C._nn.linear( - x_479, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_479 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_481 = torch.nn.functional.dropout(x_480, 0.0, False, False) - x_480 = None - x_482 = mul_85 + x_481 - mul_85 = x_481 = None - view_38 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_87 = x_482 * view_38 - view_38 = None - layer_norm_72 = torch.nn.functional.layer_norm( - x_482, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_482 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_76 = torch._C._nn.linear( - layer_norm_72, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_72 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_41 = linear_76.reshape(1, pow_53, 3, 18, 32) - linear_76 = None - qkv_19 = reshape_41.permute(2, 0, 3, 1, 4) - reshape_41 = None - unbind_19 = qkv_19.unbind(0) - qkv_19 = None - q_19 = unbind_19[0] - k_19 = unbind_19[1] - v_19 = unbind_19[2] - unbind_19 = None - x_483 = torch._C._nn.scaled_dot_product_attention( - q_19, k_19, v_19, dropout_p=0.0 - ) - q_19 = k_19 = v_19 = None - transpose_22 = x_483.transpose(1, 2) - x_483 = None - x_484 = transpose_22.reshape(1, pow_53, 576) - transpose_22 = None - x_485 = torch._C._nn.linear( - x_484, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_484 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_486 = torch.nn.functional.dropout(x_485, 0.0, False, False) - x_485 = None - x_487 = mul_87 + x_486 - mul_87 = x_486 = None - view_39 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_88 = x_487 * view_39 - view_39 = None - layer_norm_73 = torch.nn.functional.layer_norm( - x_487, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_487 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_488 = torch._C._nn.linear( - layer_norm_73, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_73 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_49 = torch.nn.functional.relu(x_488, inplace=False) - x_488 = None - pow_50 = relu_49**2 - relu_49 = None - mul_89 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_50 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_50 - ) = None - x_489 = ( - mul_89 - + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_89 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_490 = torch.nn.functional.dropout(x_489, 0.0, False, False) - x_489 = None - x_491 = torch._C._nn.linear( - x_490, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_490 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_492 = torch.nn.functional.dropout(x_491, 0.0, False, False) - x_491 = None - x_493 = mul_88 + x_492 - mul_88 = x_492 = None - view_40 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_90 = x_493 * view_40 - view_40 = None - layer_norm_74 = torch.nn.functional.layer_norm( - x_493, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_493 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_80 = torch._C._nn.linear( - layer_norm_74, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_74 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_43 = linear_80.reshape(1, pow_53, 3, 18, 32) - linear_80 = None - qkv_20 = reshape_43.permute(2, 0, 3, 1, 4) - reshape_43 = None - unbind_20 = qkv_20.unbind(0) - qkv_20 = None - q_20 = unbind_20[0] - k_20 = unbind_20[1] - v_20 = unbind_20[2] - unbind_20 = None - x_494 = torch._C._nn.scaled_dot_product_attention( - q_20, k_20, v_20, dropout_p=0.0 - ) - q_20 = k_20 = v_20 = None - transpose_23 = x_494.transpose(1, 2) - x_494 = None - x_495 = transpose_23.reshape(1, pow_53, 576) - transpose_23 = pow_53 = None - x_496 = torch._C._nn.linear( - x_495, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_495 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_497 = torch.nn.functional.dropout(x_496, 0.0, False, False) - x_496 = None - x_498 = mul_90 + x_497 - mul_90 = x_497 = None - view_41 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (576,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_91 = x_498 * view_41 - view_41 = None - layer_norm_75 = torch.nn.functional.layer_norm( - x_498, - (576,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_498 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_499 = torch._C._nn.linear( - layer_norm_75, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_75 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_50 = torch.nn.functional.relu(x_499, inplace=False) - x_499 = None - pow_51 = relu_50**2 - relu_50 = None - mul_92 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_51 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_51 - ) = None - x_500 = ( - mul_92 - + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_92 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_501 = torch.nn.functional.dropout(x_500, 0.0, False, False) - x_500 = None - x_502 = torch._C._nn.linear( - x_501, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_501 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_503 = torch.nn.functional.dropout(x_502, 0.0, False, False) - x_502 = None - x_504 = mul_91 + x_503 - mul_91 = x_503 = None - transpose_24 = x_504.transpose(1, 2) - x_504 = None - x_505 = transpose_24.reshape(1, 576, sym_sum_4, sym_sum_4) - transpose_24 = sym_sum_4 = None - x_506 = torch.nn.functional.adaptive_avg_pool2d(x_505, 1) - x_505 = None - x_507 = x_506.permute(0, 2, 3, 1) - x_506 = None - x_508 = torch.nn.functional.layer_norm( - x_507, - (576,), - l_self_modules_head_modules_norm_parameters_weight_, - l_self_modules_head_modules_norm_parameters_bias_, - 1e-06, - ) - x_507 = ( - l_self_modules_head_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_norm_parameters_bias_ = None - x_509 = x_508.permute(0, 3, 1, 2) - x_508 = None - x_510 = x_509.flatten(1, -1) - x_509 = None - x_511 = torch.nn.functional.dropout(x_510, 0.0, False, False) - x_510 = None - x_512 = torch._C._nn.linear( - x_511, - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, - ) - x_511 = ( - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None - relu_51 = torch.nn.functional.relu(x_512, inplace=False) - x_512 = None - x_513 = torch.square(relu_51) - relu_51 = None - x_514 = torch.nn.functional.layer_norm( - x_513, - (2304,), - l_self_modules_head_modules_fc_modules_norm_parameters_weight_, - l_self_modules_head_modules_fc_modules_norm_parameters_bias_, - 1e-06, - ) - x_513 = ( - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None - x_515 = torch.nn.functional.dropout(x_514, 0.0, False, False) - x_514 = None - x_516 = torch._C._nn.linear( - x_515, - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, - ) - x_515 = ( - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None - return (x_516,) diff --git a/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/weight_meta.py b/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/weight_meta.py deleted file mode 100644 index 8e5c0d0fc0..0000000000 --- a/samples/timm/caformer_m36.sail_in22k_ft_in1k_384/weight_meta.py +++ /dev/null @@ -1,3968 +0,0 @@ -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: - name = "L_self_modules_stem_modules_conv_parameters_weight_" - shape = [96, 3, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: - name = "L_self_modules_stem_modules_conv_parameters_bias_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_s1: - name = "s1" - shape = [] - dtype = "torch.int64" - device = "cpu" - mean = None - std = None - data = [4] - - -class Program_weight_tensor_meta_L_x_: - name = "L_x_" - shape = [1, 3, 224, 224] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.230 - std = 1.288 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: - name = "L_self_modules_stem_modules_norm_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [192, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [192, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [96, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [384, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [96, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [192, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [192, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [96, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [384, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [96, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [192, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [192, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [96, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [384, 96, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [96, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" - shape = [96] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" - shape = [192, 96, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.021 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm1_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [384, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [384, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [192, 384, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_norm2_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" - shape = [768, 192, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" - shape = [192, 768, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" - shape = [192] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" - shape = [384, 192, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.021 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.021 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale1_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm1_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1152, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_token_mixer_modules_proj_parameters_weight_" - shape = [384, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_res_scale2_parameters_scale_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm2_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" - shape = [1536, 384] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" - shape = [384, 1536] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" - shape = [384] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" - shape = [576, 384, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1728, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [576, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [2304, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [576, 2304] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1728, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [576, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [2304, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [576, 2304] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1728, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [576, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [2304, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [576, 2304] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_norm_parameters_weight_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_norm_parameters_bias_" - shape = [576] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" - shape = [2304, 576] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" - shape = [2304] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" - shape = [2304] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" - shape = [2304] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" - shape = [1000, 2304] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" - shape = [1000] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None diff --git a/samples/timm/caformer_s18.sail_in1k/graph_net.json b/samples/timm/caformer_s18.sail_in1k/graph_net.json deleted file mode 100644 index 1373fe3b58..0000000000 --- a/samples/timm/caformer_s18.sail_in1k/graph_net.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "framework": "torch", - "num_devices_required": 1, - "num_nodes_required": 1 -} \ No newline at end of file diff --git a/samples/timm/caformer_s18.sail_in1k/input_meta.py b/samples/timm/caformer_s18.sail_in1k/input_meta.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_s18.sail_in1k/input_tensor_constraints.py b/samples/timm/caformer_s18.sail_in1k/input_tensor_constraints.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_s18.sail_in1k/model.py b/samples/timm/caformer_s18.sail_in1k/model.py deleted file mode 100644 index ba7a8ecdba..0000000000 --- a/samples/timm/caformer_s18.sail_in1k/model.py +++ /dev/null @@ -1,2468 +0,0 @@ -import torch - - -class GraphModule(torch.nn.Module): - def forward( - self, - L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - s1: torch.SymInt, - L_x_: torch.Tensor, - L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - ): - l_self_modules_stem_modules_conv_parameters_weight_ = ( - L_self_modules_stem_modules_conv_parameters_weight_ - ) - l_self_modules_stem_modules_conv_parameters_bias_ = ( - L_self_modules_stem_modules_conv_parameters_bias_ - ) - l_x_ = L_x_ - l_self_modules_stem_modules_norm_parameters_weight_ = ( - L_self_modules_stem_modules_norm_parameters_weight_ - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_head_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ - ) - x = torch.conv2d( - l_x_, - l_self_modules_stem_modules_conv_parameters_weight_, - l_self_modules_stem_modules_conv_parameters_bias_, - (4, 4), - (2, 2), - (1, 1), - 1, - ) - l_x_ = ( - l_self_modules_stem_modules_conv_parameters_weight_ - ) = l_self_modules_stem_modules_conv_parameters_bias_ = None - x_1 = x.permute(0, 2, 3, 1) - x = None - x_2 = torch.nn.functional.layer_norm( - x_1, (64,), l_self_modules_stem_modules_norm_parameters_weight_, None, 1e-06 - ) - x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None - x_3 = x_2.permute(0, 3, 1, 2) - x_2 = None - sym_sum = torch.sym_sum([-3, s1]) - s1 = None - floordiv = sym_sum // 4 - sym_sum_1 = torch.sym_sum([1, floordiv]) - floordiv = sym_sum_1 = None - x_4 = x_3.permute(0, 2, 3, 1) - x_5 = torch.nn.functional.layer_norm( - x_4, - (64,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_6 = x_5.permute(0, 3, 1, 2) - x_5 = None - x_7 = torch.conv2d( - x_6, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu = torch.nn.functional.relu(x_7, inplace=False) - x_7 = None - pow_1 = relu**2 - relu = None - mul = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_1 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_1 - ) = None - x_8 = ( - mul - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_9 = torch.conv2d( - x_8, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 128, - ) - x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_10 = torch.conv2d( - x_9, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_11 = x_3 + x_10 - x_3 = x_10 = None - x_12 = x_11.permute(0, 2, 3, 1) - x_13 = torch.nn.functional.layer_norm( - x_12, - (64,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_14 = x_13.permute(0, 3, 1, 2) - x_13 = None - x_15 = torch.conv2d( - x_14, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_1 = torch.nn.functional.relu(x_15, inplace=False) - x_15 = None - pow_2 = relu_1**2 - relu_1 = None - mul_1 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_2 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_2 - ) = None - x_16 = ( - mul_1 - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) - x_16 = None - x_18 = torch.conv2d( - x_17, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) - x_18 = None - x_20 = x_11 + x_19 - x_11 = x_19 = None - x_21 = x_20.permute(0, 2, 3, 1) - x_22 = torch.nn.functional.layer_norm( - x_21, - (64,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_23 = x_22.permute(0, 3, 1, 2) - x_22 = None - x_24 = torch.conv2d( - x_23, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_2 = torch.nn.functional.relu(x_24, inplace=False) - x_24 = None - pow_3 = relu_2**2 - relu_2 = None - mul_2 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_3 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_3 - ) = None - x_25 = ( - mul_2 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_26 = torch.conv2d( - x_25, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 128, - ) - x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_27 = torch.conv2d( - x_26, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_28 = x_20 + x_27 - x_20 = x_27 = None - x_29 = x_28.permute(0, 2, 3, 1) - x_30 = torch.nn.functional.layer_norm( - x_29, - (64,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_31 = x_30.permute(0, 3, 1, 2) - x_30 = None - x_32 = torch.conv2d( - x_31, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_3 = torch.nn.functional.relu(x_32, inplace=False) - x_32 = None - pow_4 = relu_3**2 - relu_3 = None - mul_3 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_4 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_4 - ) = None - x_33 = ( - mul_3 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) - x_33 = None - x_35 = torch.conv2d( - x_34, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) - x_35 = None - x_37 = x_28 + x_36 - x_28 = x_36 = None - x_38 = x_37.permute(0, 2, 3, 1) - x_39 = torch.nn.functional.layer_norm( - x_38, - (64,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_40 = x_39.permute(0, 3, 1, 2) - x_39 = None - x_41 = torch.conv2d( - x_40, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_4 = torch.nn.functional.relu(x_41, inplace=False) - x_41 = None - pow_5 = relu_4**2 - relu_4 = None - mul_4 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_5 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_5 - ) = None - x_42 = ( - mul_4 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_43 = torch.conv2d( - x_42, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 128, - ) - x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_44 = torch.conv2d( - x_43, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_45 = x_37 + x_44 - x_37 = x_44 = None - x_46 = x_45.permute(0, 2, 3, 1) - x_47 = torch.nn.functional.layer_norm( - x_46, - (64,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_48 = x_47.permute(0, 3, 1, 2) - x_47 = None - x_49 = torch.conv2d( - x_48, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_5 = torch.nn.functional.relu(x_49, inplace=False) - x_49 = None - pow_6 = relu_5**2 - relu_5 = None - mul_5 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_6 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_6 - ) = None - x_50 = ( - mul_5 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) - x_50 = None - x_52 = torch.conv2d( - x_51, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) - x_52 = None - x_54 = x_45 + x_53 - x_45 = x_53 = None - x_55 = x_54.permute(0, 2, 3, 1) - x_54 = None - x_56 = torch.nn.functional.layer_norm( - x_55, - (64,), - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) - x_57 = x_56.permute(0, 3, 1, 2) - x_56 = None - x_58 = torch.conv2d( - x_57, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_1 = sym_sum // 8 - sym_sum_2 = torch.sym_sum([1, floordiv_1]) - floordiv_1 = sym_sum_2 = None - x_59 = x_58.permute(0, 2, 3, 1) - x_60 = torch.nn.functional.layer_norm( - x_59, - (128,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_61 = x_60.permute(0, 3, 1, 2) - x_60 = None - x_62 = torch.conv2d( - x_61, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_6 = torch.nn.functional.relu(x_62, inplace=False) - x_62 = None - pow_7 = relu_6**2 - relu_6 = None - mul_6 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_7 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_7 - ) = None - x_63 = ( - mul_6 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_64 = torch.conv2d( - x_63, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 256, - ) - x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_65 = torch.conv2d( - x_64, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_66 = x_58 + x_65 - x_58 = x_65 = None - x_67 = x_66.permute(0, 2, 3, 1) - x_68 = torch.nn.functional.layer_norm( - x_67, - (128,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_69 = x_68.permute(0, 3, 1, 2) - x_68 = None - x_70 = torch.conv2d( - x_69, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_7 = torch.nn.functional.relu(x_70, inplace=False) - x_70 = None - pow_8 = relu_7**2 - relu_7 = None - mul_7 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_8 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_8 - ) = None - x_71 = ( - mul_7 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) - x_71 = None - x_73 = torch.conv2d( - x_72, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) - x_73 = None - x_75 = x_66 + x_74 - x_66 = x_74 = None - x_76 = x_75.permute(0, 2, 3, 1) - x_77 = torch.nn.functional.layer_norm( - x_76, - (128,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_78 = x_77.permute(0, 3, 1, 2) - x_77 = None - x_79 = torch.conv2d( - x_78, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_8 = torch.nn.functional.relu(x_79, inplace=False) - x_79 = None - pow_9 = relu_8**2 - relu_8 = None - mul_8 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_9 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_9 - ) = None - x_80 = ( - mul_8 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_81 = torch.conv2d( - x_80, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 256, - ) - x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_82 = torch.conv2d( - x_81, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_83 = x_75 + x_82 - x_75 = x_82 = None - x_84 = x_83.permute(0, 2, 3, 1) - x_85 = torch.nn.functional.layer_norm( - x_84, - (128,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_86 = x_85.permute(0, 3, 1, 2) - x_85 = None - x_87 = torch.conv2d( - x_86, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_9 = torch.nn.functional.relu(x_87, inplace=False) - x_87 = None - pow_10 = relu_9**2 - relu_9 = None - mul_9 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_10 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_10 - ) = None - x_88 = ( - mul_9 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) - x_88 = None - x_90 = torch.conv2d( - x_89, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) - x_90 = None - x_92 = x_83 + x_91 - x_83 = x_91 = None - x_93 = x_92.permute(0, 2, 3, 1) - x_94 = torch.nn.functional.layer_norm( - x_93, - (128,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_95 = x_94.permute(0, 3, 1, 2) - x_94 = None - x_96 = torch.conv2d( - x_95, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_10 = torch.nn.functional.relu(x_96, inplace=False) - x_96 = None - pow_11 = relu_10**2 - relu_10 = None - mul_10 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_11 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_11 - ) = None - x_97 = ( - mul_10 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_98 = torch.conv2d( - x_97, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 256, - ) - x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_99 = torch.conv2d( - x_98, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_100 = x_92 + x_99 - x_92 = x_99 = None - x_101 = x_100.permute(0, 2, 3, 1) - x_102 = torch.nn.functional.layer_norm( - x_101, - (128,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_103 = x_102.permute(0, 3, 1, 2) - x_102 = None - x_104 = torch.conv2d( - x_103, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_11 = torch.nn.functional.relu(x_104, inplace=False) - x_104 = None - pow_12 = relu_11**2 - relu_11 = None - mul_11 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_12 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_12 - ) = None - x_105 = ( - mul_11 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) - x_105 = None - x_107 = torch.conv2d( - x_106, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) - x_107 = None - x_109 = x_100 + x_108 - x_100 = x_108 = None - x_110 = x_109.permute(0, 2, 3, 1) - x_109 = None - x_111 = torch.nn.functional.layer_norm( - x_110, - (128,), - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_110 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) - x_112 = x_111.permute(0, 3, 1, 2) - x_111 = None - x_113 = torch.conv2d( - x_112, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_112 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_2 = sym_sum // 16 - sym_sum_3 = torch.sym_sum([1, floordiv_2]) - floordiv_2 = None - reshape = x_113.reshape(1, 320, -1) - x_113 = None - x_114 = reshape.transpose(1, 2) - reshape = None - view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_12 = x_114 * view - view = None - layer_norm_15 = torch.nn.functional.layer_norm( - x_114, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_25 = sym_sum_3**2 - linear = torch._C._nn.linear( - layer_norm_15, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_15 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_1 = linear.reshape(1, pow_25, 3, 10, 32) - linear = None - qkv = reshape_1.permute(2, 0, 3, 1, 4) - reshape_1 = None - unbind = qkv.unbind(0) - qkv = None - q = unbind[0] - k = unbind[1] - v = unbind[2] - unbind = None - x_115 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) - q = k = v = None - transpose_1 = x_115.transpose(1, 2) - x_115 = None - x_116 = transpose_1.reshape(1, pow_25, 320) - transpose_1 = None - x_117 = torch._C._nn.linear( - x_116, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_116 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) - x_117 = None - x_119 = mul_12 + x_118 - mul_12 = x_118 = None - view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_13 = x_119 * view_1 - view_1 = None - layer_norm_16 = torch.nn.functional.layer_norm( - x_119, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_119 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_120 = torch._C._nn.linear( - layer_norm_16, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_16 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_12 = torch.nn.functional.relu(x_120, inplace=False) - x_120 = None - pow_13 = relu_12**2 - relu_12 = None - mul_14 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_13 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_13 - ) = None - x_121 = ( - mul_14 - + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_14 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_122 = torch.nn.functional.dropout(x_121, 0.0, False, False) - x_121 = None - x_123 = torch._C._nn.linear( - x_122, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) - x_123 = None - x_125 = mul_13 + x_124 - mul_13 = x_124 = None - view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_15 = x_125 * view_2 - view_2 = None - layer_norm_17 = torch.nn.functional.layer_norm( - x_125, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_4 = torch._C._nn.linear( - layer_norm_17, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_17 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_3 = linear_4.reshape(1, pow_25, 3, 10, 32) - linear_4 = None - qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) - reshape_3 = None - unbind_1 = qkv_1.unbind(0) - qkv_1 = None - q_1 = unbind_1[0] - k_1 = unbind_1[1] - v_1 = unbind_1[2] - unbind_1 = None - x_126 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) - q_1 = k_1 = v_1 = None - transpose_2 = x_126.transpose(1, 2) - x_126 = None - x_127 = transpose_2.reshape(1, pow_25, 320) - transpose_2 = None - x_128 = torch._C._nn.linear( - x_127, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_127 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_129 = torch.nn.functional.dropout(x_128, 0.0, False, False) - x_128 = None - x_130 = mul_15 + x_129 - mul_15 = x_129 = None - view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_16 = x_130 * view_3 - view_3 = None - layer_norm_18 = torch.nn.functional.layer_norm( - x_130, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_130 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_131 = torch._C._nn.linear( - layer_norm_18, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_18 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_13 = torch.nn.functional.relu(x_131, inplace=False) - x_131 = None - pow_14 = relu_13**2 - relu_13 = None - mul_17 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_14 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_14 - ) = None - x_132 = ( - mul_17 - + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_17 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) - x_132 = None - x_134 = torch._C._nn.linear( - x_133, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_133 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) - x_134 = None - x_136 = mul_16 + x_135 - mul_16 = x_135 = None - view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_18 = x_136 * view_4 - view_4 = None - layer_norm_19 = torch.nn.functional.layer_norm( - x_136, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_8 = torch._C._nn.linear( - layer_norm_19, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_19 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_5 = linear_8.reshape(1, pow_25, 3, 10, 32) - linear_8 = None - qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) - reshape_5 = None - unbind_2 = qkv_2.unbind(0) - qkv_2 = None - q_2 = unbind_2[0] - k_2 = unbind_2[1] - v_2 = unbind_2[2] - unbind_2 = None - x_137 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) - q_2 = k_2 = v_2 = None - transpose_3 = x_137.transpose(1, 2) - x_137 = None - x_138 = transpose_3.reshape(1, pow_25, 320) - transpose_3 = None - x_139 = torch._C._nn.linear( - x_138, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_138 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) - x_139 = None - x_141 = mul_18 + x_140 - mul_18 = x_140 = None - view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_19 = x_141 * view_5 - view_5 = None - layer_norm_20 = torch.nn.functional.layer_norm( - x_141, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_141 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_142 = torch._C._nn.linear( - layer_norm_20, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_20 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_14 = torch.nn.functional.relu(x_142, inplace=False) - x_142 = None - pow_15 = relu_14**2 - relu_14 = None - mul_20 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_15 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_15 - ) = None - x_143 = ( - mul_20 - + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_20 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) - x_143 = None - x_145 = torch._C._nn.linear( - x_144, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_144 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) - x_145 = None - x_147 = mul_19 + x_146 - mul_19 = x_146 = None - view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_21 = x_147 * view_6 - view_6 = None - layer_norm_21 = torch.nn.functional.layer_norm( - x_147, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_147 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) - linear_12 = torch._C._nn.linear( - layer_norm_21, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_21 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_7 = linear_12.reshape(1, pow_25, 3, 10, 32) - linear_12 = None - qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) - reshape_7 = None - unbind_3 = qkv_3.unbind(0) - qkv_3 = None - q_3 = unbind_3[0] - k_3 = unbind_3[1] - v_3 = unbind_3[2] - unbind_3 = None - x_148 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) - q_3 = k_3 = v_3 = None - transpose_4 = x_148.transpose(1, 2) - x_148 = None - x_149 = transpose_4.reshape(1, pow_25, 320) - transpose_4 = None - x_150 = torch._C._nn.linear( - x_149, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_149 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_151 = torch.nn.functional.dropout(x_150, 0.0, False, False) - x_150 = None - x_152 = mul_21 + x_151 - mul_21 = x_151 = None - view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_22 = x_152 * view_7 - view_7 = None - layer_norm_22 = torch.nn.functional.layer_norm( - x_152, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_152 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) - x_153 = torch._C._nn.linear( - layer_norm_22, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_22 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_15 = torch.nn.functional.relu(x_153, inplace=False) - x_153 = None - pow_16 = relu_15**2 - relu_15 = None - mul_23 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - * pow_16 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( - pow_16 - ) = None - x_154 = ( - mul_23 - + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - ) - mul_23 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) - x_155 = torch.nn.functional.dropout(x_154, 0.0, False, False) - x_154 = None - x_156 = torch._C._nn.linear( - x_155, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_155 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) - x_156 = None - x_158 = mul_22 + x_157 - mul_22 = x_157 = None - view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_24 = x_158 * view_8 - view_8 = None - layer_norm_23 = torch.nn.functional.layer_norm( - x_158, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) - linear_16 = torch._C._nn.linear( - layer_norm_23, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_23 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_9 = linear_16.reshape(1, pow_25, 3, 10, 32) - linear_16 = None - qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) - reshape_9 = None - unbind_4 = qkv_4.unbind(0) - qkv_4 = None - q_4 = unbind_4[0] - k_4 = unbind_4[1] - v_4 = unbind_4[2] - unbind_4 = None - x_159 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) - q_4 = k_4 = v_4 = None - transpose_5 = x_159.transpose(1, 2) - x_159 = None - x_160 = transpose_5.reshape(1, pow_25, 320) - transpose_5 = None - x_161 = torch._C._nn.linear( - x_160, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_160 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_162 = torch.nn.functional.dropout(x_161, 0.0, False, False) - x_161 = None - x_163 = mul_24 + x_162 - mul_24 = x_162 = None - view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_25 = x_163 * view_9 - view_9 = None - layer_norm_24 = torch.nn.functional.layer_norm( - x_163, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_163 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) - x_164 = torch._C._nn.linear( - layer_norm_24, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_24 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_16 = torch.nn.functional.relu(x_164, inplace=False) - x_164 = None - pow_17 = relu_16**2 - relu_16 = None - mul_26 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - * pow_17 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( - pow_17 - ) = None - x_165 = ( - mul_26 - + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - ) - mul_26 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) - x_166 = torch.nn.functional.dropout(x_165, 0.0, False, False) - x_165 = None - x_167 = torch._C._nn.linear( - x_166, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_166 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) - x_167 = None - x_169 = mul_25 + x_168 - mul_25 = x_168 = None - view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_27 = x_169 * view_10 - view_10 = None - layer_norm_25 = torch.nn.functional.layer_norm( - x_169, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) - linear_20 = torch._C._nn.linear( - layer_norm_25, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_25 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_11 = linear_20.reshape(1, pow_25, 3, 10, 32) - linear_20 = None - qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) - reshape_11 = None - unbind_5 = qkv_5.unbind(0) - qkv_5 = None - q_5 = unbind_5[0] - k_5 = unbind_5[1] - v_5 = unbind_5[2] - unbind_5 = None - x_170 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) - q_5 = k_5 = v_5 = None - transpose_6 = x_170.transpose(1, 2) - x_170 = None - x_171 = transpose_6.reshape(1, pow_25, 320) - transpose_6 = None - x_172 = torch._C._nn.linear( - x_171, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_171 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_173 = torch.nn.functional.dropout(x_172, 0.0, False, False) - x_172 = None - x_174 = mul_27 + x_173 - mul_27 = x_173 = None - view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_28 = x_174 * view_11 - view_11 = None - layer_norm_26 = torch.nn.functional.layer_norm( - x_174, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_174 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) - x_175 = torch._C._nn.linear( - layer_norm_26, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_26 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_17 = torch.nn.functional.relu(x_175, inplace=False) - x_175 = None - pow_18 = relu_17**2 - relu_17 = None - mul_29 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - * pow_18 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( - pow_18 - ) = None - x_176 = ( - mul_29 - + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - ) - mul_29 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) - x_177 = torch.nn.functional.dropout(x_176, 0.0, False, False) - x_176 = None - x_178 = torch._C._nn.linear( - x_177, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_177 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) - x_178 = None - x_180 = mul_28 + x_179 - mul_28 = x_179 = None - view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_30 = x_180 * view_12 - view_12 = None - layer_norm_27 = torch.nn.functional.layer_norm( - x_180, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_180 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) - linear_24 = torch._C._nn.linear( - layer_norm_27, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_27 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_13 = linear_24.reshape(1, pow_25, 3, 10, 32) - linear_24 = None - qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) - reshape_13 = None - unbind_6 = qkv_6.unbind(0) - qkv_6 = None - q_6 = unbind_6[0] - k_6 = unbind_6[1] - v_6 = unbind_6[2] - unbind_6 = None - x_181 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) - q_6 = k_6 = v_6 = None - transpose_7 = x_181.transpose(1, 2) - x_181 = None - x_182 = transpose_7.reshape(1, pow_25, 320) - transpose_7 = None - x_183 = torch._C._nn.linear( - x_182, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_182 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_184 = torch.nn.functional.dropout(x_183, 0.0, False, False) - x_183 = None - x_185 = mul_30 + x_184 - mul_30 = x_184 = None - view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_31 = x_185 * view_13 - view_13 = None - layer_norm_28 = torch.nn.functional.layer_norm( - x_185, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_185 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) - x_186 = torch._C._nn.linear( - layer_norm_28, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_28 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_18 = torch.nn.functional.relu(x_186, inplace=False) - x_186 = None - pow_19 = relu_18**2 - relu_18 = None - mul_32 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - * pow_19 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( - pow_19 - ) = None - x_187 = ( - mul_32 - + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - ) - mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) - x_188 = torch.nn.functional.dropout(x_187, 0.0, False, False) - x_187 = None - x_189 = torch._C._nn.linear( - x_188, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_188 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) - x_189 = None - x_191 = mul_31 + x_190 - mul_31 = x_190 = None - view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_33 = x_191 * view_14 - view_14 = None - layer_norm_29 = torch.nn.functional.layer_norm( - x_191, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_191 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) - linear_28 = torch._C._nn.linear( - layer_norm_29, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_29 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_15 = linear_28.reshape(1, pow_25, 3, 10, 32) - linear_28 = None - qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) - reshape_15 = None - unbind_7 = qkv_7.unbind(0) - qkv_7 = None - q_7 = unbind_7[0] - k_7 = unbind_7[1] - v_7 = unbind_7[2] - unbind_7 = None - x_192 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) - q_7 = k_7 = v_7 = None - transpose_8 = x_192.transpose(1, 2) - x_192 = None - x_193 = transpose_8.reshape(1, pow_25, 320) - transpose_8 = None - x_194 = torch._C._nn.linear( - x_193, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_193 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_195 = torch.nn.functional.dropout(x_194, 0.0, False, False) - x_194 = None - x_196 = mul_33 + x_195 - mul_33 = x_195 = None - view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_34 = x_196 * view_15 - view_15 = None - layer_norm_30 = torch.nn.functional.layer_norm( - x_196, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_196 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) - x_197 = torch._C._nn.linear( - layer_norm_30, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_30 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_19 = torch.nn.functional.relu(x_197, inplace=False) - x_197 = None - pow_20 = relu_19**2 - relu_19 = None - mul_35 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - * pow_20 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( - pow_20 - ) = None - x_198 = ( - mul_35 - + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - ) - mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) - x_199 = torch.nn.functional.dropout(x_198, 0.0, False, False) - x_198 = None - x_200 = torch._C._nn.linear( - x_199, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_199 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) - x_200 = None - x_202 = mul_34 + x_201 - mul_34 = x_201 = None - view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_36 = x_202 * view_16 - view_16 = None - layer_norm_31 = torch.nn.functional.layer_norm( - x_202, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_202 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) - linear_32 = torch._C._nn.linear( - layer_norm_31, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_31 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_17 = linear_32.reshape(1, pow_25, 3, 10, 32) - linear_32 = None - qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) - reshape_17 = None - unbind_8 = qkv_8.unbind(0) - qkv_8 = None - q_8 = unbind_8[0] - k_8 = unbind_8[1] - v_8 = unbind_8[2] - unbind_8 = None - x_203 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) - q_8 = k_8 = v_8 = None - transpose_9 = x_203.transpose(1, 2) - x_203 = None - x_204 = transpose_9.reshape(1, pow_25, 320) - transpose_9 = pow_25 = None - x_205 = torch._C._nn.linear( - x_204, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_204 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_206 = torch.nn.functional.dropout(x_205, 0.0, False, False) - x_205 = None - x_207 = mul_36 + x_206 - mul_36 = x_206 = None - view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_37 = x_207 * view_17 - view_17 = None - layer_norm_32 = torch.nn.functional.layer_norm( - x_207, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_207 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) - x_208 = torch._C._nn.linear( - layer_norm_32, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_32 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_20 = torch.nn.functional.relu(x_208, inplace=False) - x_208 = None - pow_21 = relu_20**2 - relu_20 = None - mul_38 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - * pow_21 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( - pow_21 - ) = None - x_209 = ( - mul_38 - + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - ) - mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) - x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) - x_209 = None - x_211 = torch._C._nn.linear( - x_210, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_210 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) - x_211 = None - x_213 = mul_37 + x_212 - mul_37 = x_212 = None - transpose_10 = x_213.transpose(1, 2) - x_213 = None - x_214 = transpose_10.reshape(1, 320, sym_sum_3, sym_sum_3) - transpose_10 = sym_sum_3 = None - x_215 = x_214.permute(0, 2, 3, 1) - x_214 = None - x_216 = torch.nn.functional.layer_norm( - x_215, - (320,), - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_215 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) - x_217 = x_216.permute(0, 3, 1, 2) - x_216 = None - x_218 = torch.conv2d( - x_217, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_217 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_3 = sym_sum // 32 - sym_sum = None - sym_sum_4 = torch.sym_sum([1, floordiv_3]) - floordiv_3 = None - reshape_20 = x_218.reshape(1, 512, -1) - x_218 = None - x_219 = reshape_20.transpose(1, 2) - reshape_20 = None - view_18 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_39 = x_219 * view_18 - view_18 = None - layer_norm_34 = torch.nn.functional.layer_norm( - x_219, - (512,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_219 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_26 = sym_sum_4**2 - linear_36 = torch._C._nn.linear( - layer_norm_34, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_34 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_21 = linear_36.reshape(1, pow_26, 3, 16, 32) - linear_36 = None - qkv_9 = reshape_21.permute(2, 0, 3, 1, 4) - reshape_21 = None - unbind_9 = qkv_9.unbind(0) - qkv_9 = None - q_9 = unbind_9[0] - k_9 = unbind_9[1] - v_9 = unbind_9[2] - unbind_9 = None - x_220 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) - q_9 = k_9 = v_9 = None - transpose_12 = x_220.transpose(1, 2) - x_220 = None - x_221 = transpose_12.reshape(1, pow_26, 512) - transpose_12 = None - x_222 = torch._C._nn.linear( - x_221, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_221 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) - x_222 = None - x_224 = mul_39 + x_223 - mul_39 = x_223 = None - view_19 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_40 = x_224 * view_19 - view_19 = None - layer_norm_35 = torch.nn.functional.layer_norm( - x_224, - (512,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_224 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_225 = torch._C._nn.linear( - layer_norm_35, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_35 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_21 = torch.nn.functional.relu(x_225, inplace=False) - x_225 = None - pow_22 = relu_21**2 - relu_21 = None - mul_41 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_22 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_22 - ) = None - x_226 = ( - mul_41 - + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_41 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) - x_226 = None - x_228 = torch._C._nn.linear( - x_227, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_227 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_229 = torch.nn.functional.dropout(x_228, 0.0, False, False) - x_228 = None - x_230 = mul_40 + x_229 - mul_40 = x_229 = None - view_20 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_42 = x_230 * view_20 - view_20 = None - layer_norm_36 = torch.nn.functional.layer_norm( - x_230, - (512,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_230 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_40 = torch._C._nn.linear( - layer_norm_36, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_36 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_23 = linear_40.reshape(1, pow_26, 3, 16, 32) - linear_40 = None - qkv_10 = reshape_23.permute(2, 0, 3, 1, 4) - reshape_23 = None - unbind_10 = qkv_10.unbind(0) - qkv_10 = None - q_10 = unbind_10[0] - k_10 = unbind_10[1] - v_10 = unbind_10[2] - unbind_10 = None - x_231 = torch._C._nn.scaled_dot_product_attention( - q_10, k_10, v_10, dropout_p=0.0 - ) - q_10 = k_10 = v_10 = None - transpose_13 = x_231.transpose(1, 2) - x_231 = None - x_232 = transpose_13.reshape(1, pow_26, 512) - transpose_13 = None - x_233 = torch._C._nn.linear( - x_232, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_232 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) - x_233 = None - x_235 = mul_42 + x_234 - mul_42 = x_234 = None - view_21 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_43 = x_235 * view_21 - view_21 = None - layer_norm_37 = torch.nn.functional.layer_norm( - x_235, - (512,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_235 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_236 = torch._C._nn.linear( - layer_norm_37, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_37 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_22 = torch.nn.functional.relu(x_236, inplace=False) - x_236 = None - pow_23 = relu_22**2 - relu_22 = None - mul_44 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_23 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_23 - ) = None - x_237 = ( - mul_44 - + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_44 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_238 = torch.nn.functional.dropout(x_237, 0.0, False, False) - x_237 = None - x_239 = torch._C._nn.linear( - x_238, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_238 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_240 = torch.nn.functional.dropout(x_239, 0.0, False, False) - x_239 = None - x_241 = mul_43 + x_240 - mul_43 = x_240 = None - view_22 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_45 = x_241 * view_22 - view_22 = None - layer_norm_38 = torch.nn.functional.layer_norm( - x_241, - (512,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_241 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_44 = torch._C._nn.linear( - layer_norm_38, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_38 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_25 = linear_44.reshape(1, pow_26, 3, 16, 32) - linear_44 = None - qkv_11 = reshape_25.permute(2, 0, 3, 1, 4) - reshape_25 = None - unbind_11 = qkv_11.unbind(0) - qkv_11 = None - q_11 = unbind_11[0] - k_11 = unbind_11[1] - v_11 = unbind_11[2] - unbind_11 = None - x_242 = torch._C._nn.scaled_dot_product_attention( - q_11, k_11, v_11, dropout_p=0.0 - ) - q_11 = k_11 = v_11 = None - transpose_14 = x_242.transpose(1, 2) - x_242 = None - x_243 = transpose_14.reshape(1, pow_26, 512) - transpose_14 = pow_26 = None - x_244 = torch._C._nn.linear( - x_243, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_243 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) - x_244 = None - x_246 = mul_45 + x_245 - mul_45 = x_245 = None - view_23 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_46 = x_246 * view_23 - view_23 = None - layer_norm_39 = torch.nn.functional.layer_norm( - x_246, - (512,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_246 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_247 = torch._C._nn.linear( - layer_norm_39, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_39 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_23 = torch.nn.functional.relu(x_247, inplace=False) - x_247 = None - pow_24 = relu_23**2 - relu_23 = None - mul_47 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_24 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_24 - ) = None - x_248 = ( - mul_47 - + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_47 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_249 = torch.nn.functional.dropout(x_248, 0.0, False, False) - x_248 = None - x_250 = torch._C._nn.linear( - x_249, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_249 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_251 = torch.nn.functional.dropout(x_250, 0.0, False, False) - x_250 = None - x_252 = mul_46 + x_251 - mul_46 = x_251 = None - transpose_15 = x_252.transpose(1, 2) - x_252 = None - x_253 = transpose_15.reshape(1, 512, sym_sum_4, sym_sum_4) - transpose_15 = sym_sum_4 = None - x_254 = torch.nn.functional.adaptive_avg_pool2d(x_253, 1) - x_253 = None - x_255 = x_254.permute(0, 2, 3, 1) - x_254 = None - x_256 = torch.nn.functional.layer_norm( - x_255, - (512,), - l_self_modules_head_modules_norm_parameters_weight_, - l_self_modules_head_modules_norm_parameters_bias_, - 1e-06, - ) - x_255 = ( - l_self_modules_head_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_norm_parameters_bias_ = None - x_257 = x_256.permute(0, 3, 1, 2) - x_256 = None - x_258 = x_257.flatten(1, -1) - x_257 = None - x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) - x_258 = None - x_260 = torch._C._nn.linear( - x_259, - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, - ) - x_259 = ( - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None - relu_24 = torch.nn.functional.relu(x_260, inplace=False) - x_260 = None - x_261 = torch.square(relu_24) - relu_24 = None - x_262 = torch.nn.functional.layer_norm( - x_261, - (2048,), - l_self_modules_head_modules_fc_modules_norm_parameters_weight_, - l_self_modules_head_modules_fc_modules_norm_parameters_bias_, - 1e-06, - ) - x_261 = ( - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None - x_263 = torch.nn.functional.dropout(x_262, 0.0, False, False) - x_262 = None - x_264 = torch._C._nn.linear( - x_263, - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, - ) - x_263 = ( - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None - return (x_264,) diff --git a/samples/timm/caformer_s18.sail_in1k/weight_meta.py b/samples/timm/caformer_s18.sail_in1k/weight_meta.py deleted file mode 100644 index 9c99be36a5..0000000000 --- a/samples/timm/caformer_s18.sail_in1k/weight_meta.py +++ /dev/null @@ -1,2078 +0,0 @@ -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: - name = "L_self_modules_stem_modules_conv_parameters_weight_" - shape = [64, 3, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: - name = "L_self_modules_stem_modules_conv_parameters_bias_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_s1: - name = "s1" - shape = [] - dtype = "torch.int64" - device = "cpu" - mean = None - std = None - data = [4] - - -class Program_weight_tensor_meta_L_x_: - name = "L_x_" - shape = [1, 3, 224, 224] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.230 - std = 1.288 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: - name = "L_self_modules_stem_modules_norm_parameters_weight_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [128, 64, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [128, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [64, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [256, 64, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [64, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [128, 64, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [128, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [64, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [256, 64, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [64, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [128, 64, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [128, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [64, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [256, 64, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [64, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" - shape = [128, 64, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [256, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [256, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [128, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [512, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [128, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [256, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [256, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [128, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [512, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [128, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [256, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [256, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [128, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [512, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [128, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" - shape = [320, 128, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" - shape = [512, 320, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_norm_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_norm_parameters_bias_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" - shape = [2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" - shape = [2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" - shape = [2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" - shape = [1000, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" - shape = [1000] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None diff --git a/samples/timm/caformer_s18.sail_in1k_384/graph_net.json b/samples/timm/caformer_s18.sail_in1k_384/graph_net.json deleted file mode 100644 index 1373fe3b58..0000000000 --- a/samples/timm/caformer_s18.sail_in1k_384/graph_net.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "framework": "torch", - "num_devices_required": 1, - "num_nodes_required": 1 -} \ No newline at end of file diff --git a/samples/timm/caformer_s18.sail_in1k_384/input_meta.py b/samples/timm/caformer_s18.sail_in1k_384/input_meta.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_s18.sail_in1k_384/input_tensor_constraints.py b/samples/timm/caformer_s18.sail_in1k_384/input_tensor_constraints.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_s18.sail_in1k_384/model.py b/samples/timm/caformer_s18.sail_in1k_384/model.py deleted file mode 100644 index ba7a8ecdba..0000000000 --- a/samples/timm/caformer_s18.sail_in1k_384/model.py +++ /dev/null @@ -1,2468 +0,0 @@ -import torch - - -class GraphModule(torch.nn.Module): - def forward( - self, - L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - s1: torch.SymInt, - L_x_: torch.Tensor, - L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - ): - l_self_modules_stem_modules_conv_parameters_weight_ = ( - L_self_modules_stem_modules_conv_parameters_weight_ - ) - l_self_modules_stem_modules_conv_parameters_bias_ = ( - L_self_modules_stem_modules_conv_parameters_bias_ - ) - l_x_ = L_x_ - l_self_modules_stem_modules_norm_parameters_weight_ = ( - L_self_modules_stem_modules_norm_parameters_weight_ - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_head_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ - ) - x = torch.conv2d( - l_x_, - l_self_modules_stem_modules_conv_parameters_weight_, - l_self_modules_stem_modules_conv_parameters_bias_, - (4, 4), - (2, 2), - (1, 1), - 1, - ) - l_x_ = ( - l_self_modules_stem_modules_conv_parameters_weight_ - ) = l_self_modules_stem_modules_conv_parameters_bias_ = None - x_1 = x.permute(0, 2, 3, 1) - x = None - x_2 = torch.nn.functional.layer_norm( - x_1, (64,), l_self_modules_stem_modules_norm_parameters_weight_, None, 1e-06 - ) - x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None - x_3 = x_2.permute(0, 3, 1, 2) - x_2 = None - sym_sum = torch.sym_sum([-3, s1]) - s1 = None - floordiv = sym_sum // 4 - sym_sum_1 = torch.sym_sum([1, floordiv]) - floordiv = sym_sum_1 = None - x_4 = x_3.permute(0, 2, 3, 1) - x_5 = torch.nn.functional.layer_norm( - x_4, - (64,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_6 = x_5.permute(0, 3, 1, 2) - x_5 = None - x_7 = torch.conv2d( - x_6, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu = torch.nn.functional.relu(x_7, inplace=False) - x_7 = None - pow_1 = relu**2 - relu = None - mul = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_1 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_1 - ) = None - x_8 = ( - mul - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_9 = torch.conv2d( - x_8, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 128, - ) - x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_10 = torch.conv2d( - x_9, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_11 = x_3 + x_10 - x_3 = x_10 = None - x_12 = x_11.permute(0, 2, 3, 1) - x_13 = torch.nn.functional.layer_norm( - x_12, - (64,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_14 = x_13.permute(0, 3, 1, 2) - x_13 = None - x_15 = torch.conv2d( - x_14, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_1 = torch.nn.functional.relu(x_15, inplace=False) - x_15 = None - pow_2 = relu_1**2 - relu_1 = None - mul_1 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_2 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_2 - ) = None - x_16 = ( - mul_1 - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) - x_16 = None - x_18 = torch.conv2d( - x_17, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) - x_18 = None - x_20 = x_11 + x_19 - x_11 = x_19 = None - x_21 = x_20.permute(0, 2, 3, 1) - x_22 = torch.nn.functional.layer_norm( - x_21, - (64,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_23 = x_22.permute(0, 3, 1, 2) - x_22 = None - x_24 = torch.conv2d( - x_23, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_2 = torch.nn.functional.relu(x_24, inplace=False) - x_24 = None - pow_3 = relu_2**2 - relu_2 = None - mul_2 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_3 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_3 - ) = None - x_25 = ( - mul_2 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_26 = torch.conv2d( - x_25, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 128, - ) - x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_27 = torch.conv2d( - x_26, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_28 = x_20 + x_27 - x_20 = x_27 = None - x_29 = x_28.permute(0, 2, 3, 1) - x_30 = torch.nn.functional.layer_norm( - x_29, - (64,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_31 = x_30.permute(0, 3, 1, 2) - x_30 = None - x_32 = torch.conv2d( - x_31, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_3 = torch.nn.functional.relu(x_32, inplace=False) - x_32 = None - pow_4 = relu_3**2 - relu_3 = None - mul_3 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_4 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_4 - ) = None - x_33 = ( - mul_3 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) - x_33 = None - x_35 = torch.conv2d( - x_34, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) - x_35 = None - x_37 = x_28 + x_36 - x_28 = x_36 = None - x_38 = x_37.permute(0, 2, 3, 1) - x_39 = torch.nn.functional.layer_norm( - x_38, - (64,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_40 = x_39.permute(0, 3, 1, 2) - x_39 = None - x_41 = torch.conv2d( - x_40, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_4 = torch.nn.functional.relu(x_41, inplace=False) - x_41 = None - pow_5 = relu_4**2 - relu_4 = None - mul_4 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_5 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_5 - ) = None - x_42 = ( - mul_4 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_43 = torch.conv2d( - x_42, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 128, - ) - x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_44 = torch.conv2d( - x_43, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_45 = x_37 + x_44 - x_37 = x_44 = None - x_46 = x_45.permute(0, 2, 3, 1) - x_47 = torch.nn.functional.layer_norm( - x_46, - (64,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_48 = x_47.permute(0, 3, 1, 2) - x_47 = None - x_49 = torch.conv2d( - x_48, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_5 = torch.nn.functional.relu(x_49, inplace=False) - x_49 = None - pow_6 = relu_5**2 - relu_5 = None - mul_5 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_6 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_6 - ) = None - x_50 = ( - mul_5 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) - x_50 = None - x_52 = torch.conv2d( - x_51, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) - x_52 = None - x_54 = x_45 + x_53 - x_45 = x_53 = None - x_55 = x_54.permute(0, 2, 3, 1) - x_54 = None - x_56 = torch.nn.functional.layer_norm( - x_55, - (64,), - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) - x_57 = x_56.permute(0, 3, 1, 2) - x_56 = None - x_58 = torch.conv2d( - x_57, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_1 = sym_sum // 8 - sym_sum_2 = torch.sym_sum([1, floordiv_1]) - floordiv_1 = sym_sum_2 = None - x_59 = x_58.permute(0, 2, 3, 1) - x_60 = torch.nn.functional.layer_norm( - x_59, - (128,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_61 = x_60.permute(0, 3, 1, 2) - x_60 = None - x_62 = torch.conv2d( - x_61, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_6 = torch.nn.functional.relu(x_62, inplace=False) - x_62 = None - pow_7 = relu_6**2 - relu_6 = None - mul_6 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_7 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_7 - ) = None - x_63 = ( - mul_6 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_64 = torch.conv2d( - x_63, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 256, - ) - x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_65 = torch.conv2d( - x_64, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_66 = x_58 + x_65 - x_58 = x_65 = None - x_67 = x_66.permute(0, 2, 3, 1) - x_68 = torch.nn.functional.layer_norm( - x_67, - (128,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_69 = x_68.permute(0, 3, 1, 2) - x_68 = None - x_70 = torch.conv2d( - x_69, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_7 = torch.nn.functional.relu(x_70, inplace=False) - x_70 = None - pow_8 = relu_7**2 - relu_7 = None - mul_7 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_8 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_8 - ) = None - x_71 = ( - mul_7 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) - x_71 = None - x_73 = torch.conv2d( - x_72, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) - x_73 = None - x_75 = x_66 + x_74 - x_66 = x_74 = None - x_76 = x_75.permute(0, 2, 3, 1) - x_77 = torch.nn.functional.layer_norm( - x_76, - (128,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_78 = x_77.permute(0, 3, 1, 2) - x_77 = None - x_79 = torch.conv2d( - x_78, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_8 = torch.nn.functional.relu(x_79, inplace=False) - x_79 = None - pow_9 = relu_8**2 - relu_8 = None - mul_8 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_9 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_9 - ) = None - x_80 = ( - mul_8 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_81 = torch.conv2d( - x_80, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 256, - ) - x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_82 = torch.conv2d( - x_81, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_83 = x_75 + x_82 - x_75 = x_82 = None - x_84 = x_83.permute(0, 2, 3, 1) - x_85 = torch.nn.functional.layer_norm( - x_84, - (128,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_86 = x_85.permute(0, 3, 1, 2) - x_85 = None - x_87 = torch.conv2d( - x_86, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_9 = torch.nn.functional.relu(x_87, inplace=False) - x_87 = None - pow_10 = relu_9**2 - relu_9 = None - mul_9 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_10 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_10 - ) = None - x_88 = ( - mul_9 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) - x_88 = None - x_90 = torch.conv2d( - x_89, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) - x_90 = None - x_92 = x_83 + x_91 - x_83 = x_91 = None - x_93 = x_92.permute(0, 2, 3, 1) - x_94 = torch.nn.functional.layer_norm( - x_93, - (128,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_95 = x_94.permute(0, 3, 1, 2) - x_94 = None - x_96 = torch.conv2d( - x_95, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_10 = torch.nn.functional.relu(x_96, inplace=False) - x_96 = None - pow_11 = relu_10**2 - relu_10 = None - mul_10 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_11 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_11 - ) = None - x_97 = ( - mul_10 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_98 = torch.conv2d( - x_97, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 256, - ) - x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_99 = torch.conv2d( - x_98, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_100 = x_92 + x_99 - x_92 = x_99 = None - x_101 = x_100.permute(0, 2, 3, 1) - x_102 = torch.nn.functional.layer_norm( - x_101, - (128,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_103 = x_102.permute(0, 3, 1, 2) - x_102 = None - x_104 = torch.conv2d( - x_103, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_11 = torch.nn.functional.relu(x_104, inplace=False) - x_104 = None - pow_12 = relu_11**2 - relu_11 = None - mul_11 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_12 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_12 - ) = None - x_105 = ( - mul_11 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) - x_105 = None - x_107 = torch.conv2d( - x_106, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) - x_107 = None - x_109 = x_100 + x_108 - x_100 = x_108 = None - x_110 = x_109.permute(0, 2, 3, 1) - x_109 = None - x_111 = torch.nn.functional.layer_norm( - x_110, - (128,), - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_110 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) - x_112 = x_111.permute(0, 3, 1, 2) - x_111 = None - x_113 = torch.conv2d( - x_112, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_112 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_2 = sym_sum // 16 - sym_sum_3 = torch.sym_sum([1, floordiv_2]) - floordiv_2 = None - reshape = x_113.reshape(1, 320, -1) - x_113 = None - x_114 = reshape.transpose(1, 2) - reshape = None - view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_12 = x_114 * view - view = None - layer_norm_15 = torch.nn.functional.layer_norm( - x_114, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_25 = sym_sum_3**2 - linear = torch._C._nn.linear( - layer_norm_15, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_15 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_1 = linear.reshape(1, pow_25, 3, 10, 32) - linear = None - qkv = reshape_1.permute(2, 0, 3, 1, 4) - reshape_1 = None - unbind = qkv.unbind(0) - qkv = None - q = unbind[0] - k = unbind[1] - v = unbind[2] - unbind = None - x_115 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) - q = k = v = None - transpose_1 = x_115.transpose(1, 2) - x_115 = None - x_116 = transpose_1.reshape(1, pow_25, 320) - transpose_1 = None - x_117 = torch._C._nn.linear( - x_116, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_116 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) - x_117 = None - x_119 = mul_12 + x_118 - mul_12 = x_118 = None - view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_13 = x_119 * view_1 - view_1 = None - layer_norm_16 = torch.nn.functional.layer_norm( - x_119, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_119 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_120 = torch._C._nn.linear( - layer_norm_16, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_16 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_12 = torch.nn.functional.relu(x_120, inplace=False) - x_120 = None - pow_13 = relu_12**2 - relu_12 = None - mul_14 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_13 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_13 - ) = None - x_121 = ( - mul_14 - + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_14 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_122 = torch.nn.functional.dropout(x_121, 0.0, False, False) - x_121 = None - x_123 = torch._C._nn.linear( - x_122, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) - x_123 = None - x_125 = mul_13 + x_124 - mul_13 = x_124 = None - view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_15 = x_125 * view_2 - view_2 = None - layer_norm_17 = torch.nn.functional.layer_norm( - x_125, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_4 = torch._C._nn.linear( - layer_norm_17, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_17 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_3 = linear_4.reshape(1, pow_25, 3, 10, 32) - linear_4 = None - qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) - reshape_3 = None - unbind_1 = qkv_1.unbind(0) - qkv_1 = None - q_1 = unbind_1[0] - k_1 = unbind_1[1] - v_1 = unbind_1[2] - unbind_1 = None - x_126 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) - q_1 = k_1 = v_1 = None - transpose_2 = x_126.transpose(1, 2) - x_126 = None - x_127 = transpose_2.reshape(1, pow_25, 320) - transpose_2 = None - x_128 = torch._C._nn.linear( - x_127, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_127 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_129 = torch.nn.functional.dropout(x_128, 0.0, False, False) - x_128 = None - x_130 = mul_15 + x_129 - mul_15 = x_129 = None - view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_16 = x_130 * view_3 - view_3 = None - layer_norm_18 = torch.nn.functional.layer_norm( - x_130, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_130 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_131 = torch._C._nn.linear( - layer_norm_18, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_18 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_13 = torch.nn.functional.relu(x_131, inplace=False) - x_131 = None - pow_14 = relu_13**2 - relu_13 = None - mul_17 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_14 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_14 - ) = None - x_132 = ( - mul_17 - + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_17 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) - x_132 = None - x_134 = torch._C._nn.linear( - x_133, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_133 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) - x_134 = None - x_136 = mul_16 + x_135 - mul_16 = x_135 = None - view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_18 = x_136 * view_4 - view_4 = None - layer_norm_19 = torch.nn.functional.layer_norm( - x_136, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_8 = torch._C._nn.linear( - layer_norm_19, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_19 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_5 = linear_8.reshape(1, pow_25, 3, 10, 32) - linear_8 = None - qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) - reshape_5 = None - unbind_2 = qkv_2.unbind(0) - qkv_2 = None - q_2 = unbind_2[0] - k_2 = unbind_2[1] - v_2 = unbind_2[2] - unbind_2 = None - x_137 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) - q_2 = k_2 = v_2 = None - transpose_3 = x_137.transpose(1, 2) - x_137 = None - x_138 = transpose_3.reshape(1, pow_25, 320) - transpose_3 = None - x_139 = torch._C._nn.linear( - x_138, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_138 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) - x_139 = None - x_141 = mul_18 + x_140 - mul_18 = x_140 = None - view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_19 = x_141 * view_5 - view_5 = None - layer_norm_20 = torch.nn.functional.layer_norm( - x_141, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_141 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_142 = torch._C._nn.linear( - layer_norm_20, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_20 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_14 = torch.nn.functional.relu(x_142, inplace=False) - x_142 = None - pow_15 = relu_14**2 - relu_14 = None - mul_20 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_15 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_15 - ) = None - x_143 = ( - mul_20 - + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_20 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) - x_143 = None - x_145 = torch._C._nn.linear( - x_144, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_144 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) - x_145 = None - x_147 = mul_19 + x_146 - mul_19 = x_146 = None - view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_21 = x_147 * view_6 - view_6 = None - layer_norm_21 = torch.nn.functional.layer_norm( - x_147, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_147 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) - linear_12 = torch._C._nn.linear( - layer_norm_21, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_21 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_7 = linear_12.reshape(1, pow_25, 3, 10, 32) - linear_12 = None - qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) - reshape_7 = None - unbind_3 = qkv_3.unbind(0) - qkv_3 = None - q_3 = unbind_3[0] - k_3 = unbind_3[1] - v_3 = unbind_3[2] - unbind_3 = None - x_148 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) - q_3 = k_3 = v_3 = None - transpose_4 = x_148.transpose(1, 2) - x_148 = None - x_149 = transpose_4.reshape(1, pow_25, 320) - transpose_4 = None - x_150 = torch._C._nn.linear( - x_149, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_149 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_151 = torch.nn.functional.dropout(x_150, 0.0, False, False) - x_150 = None - x_152 = mul_21 + x_151 - mul_21 = x_151 = None - view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_22 = x_152 * view_7 - view_7 = None - layer_norm_22 = torch.nn.functional.layer_norm( - x_152, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_152 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) - x_153 = torch._C._nn.linear( - layer_norm_22, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_22 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_15 = torch.nn.functional.relu(x_153, inplace=False) - x_153 = None - pow_16 = relu_15**2 - relu_15 = None - mul_23 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - * pow_16 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( - pow_16 - ) = None - x_154 = ( - mul_23 - + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - ) - mul_23 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) - x_155 = torch.nn.functional.dropout(x_154, 0.0, False, False) - x_154 = None - x_156 = torch._C._nn.linear( - x_155, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_155 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) - x_156 = None - x_158 = mul_22 + x_157 - mul_22 = x_157 = None - view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_24 = x_158 * view_8 - view_8 = None - layer_norm_23 = torch.nn.functional.layer_norm( - x_158, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) - linear_16 = torch._C._nn.linear( - layer_norm_23, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_23 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_9 = linear_16.reshape(1, pow_25, 3, 10, 32) - linear_16 = None - qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) - reshape_9 = None - unbind_4 = qkv_4.unbind(0) - qkv_4 = None - q_4 = unbind_4[0] - k_4 = unbind_4[1] - v_4 = unbind_4[2] - unbind_4 = None - x_159 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) - q_4 = k_4 = v_4 = None - transpose_5 = x_159.transpose(1, 2) - x_159 = None - x_160 = transpose_5.reshape(1, pow_25, 320) - transpose_5 = None - x_161 = torch._C._nn.linear( - x_160, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_160 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_162 = torch.nn.functional.dropout(x_161, 0.0, False, False) - x_161 = None - x_163 = mul_24 + x_162 - mul_24 = x_162 = None - view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_25 = x_163 * view_9 - view_9 = None - layer_norm_24 = torch.nn.functional.layer_norm( - x_163, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_163 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) - x_164 = torch._C._nn.linear( - layer_norm_24, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_24 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_16 = torch.nn.functional.relu(x_164, inplace=False) - x_164 = None - pow_17 = relu_16**2 - relu_16 = None - mul_26 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - * pow_17 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( - pow_17 - ) = None - x_165 = ( - mul_26 - + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - ) - mul_26 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) - x_166 = torch.nn.functional.dropout(x_165, 0.0, False, False) - x_165 = None - x_167 = torch._C._nn.linear( - x_166, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_166 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) - x_167 = None - x_169 = mul_25 + x_168 - mul_25 = x_168 = None - view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_27 = x_169 * view_10 - view_10 = None - layer_norm_25 = torch.nn.functional.layer_norm( - x_169, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) - linear_20 = torch._C._nn.linear( - layer_norm_25, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_25 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_11 = linear_20.reshape(1, pow_25, 3, 10, 32) - linear_20 = None - qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) - reshape_11 = None - unbind_5 = qkv_5.unbind(0) - qkv_5 = None - q_5 = unbind_5[0] - k_5 = unbind_5[1] - v_5 = unbind_5[2] - unbind_5 = None - x_170 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) - q_5 = k_5 = v_5 = None - transpose_6 = x_170.transpose(1, 2) - x_170 = None - x_171 = transpose_6.reshape(1, pow_25, 320) - transpose_6 = None - x_172 = torch._C._nn.linear( - x_171, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_171 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_173 = torch.nn.functional.dropout(x_172, 0.0, False, False) - x_172 = None - x_174 = mul_27 + x_173 - mul_27 = x_173 = None - view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_28 = x_174 * view_11 - view_11 = None - layer_norm_26 = torch.nn.functional.layer_norm( - x_174, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_174 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) - x_175 = torch._C._nn.linear( - layer_norm_26, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_26 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_17 = torch.nn.functional.relu(x_175, inplace=False) - x_175 = None - pow_18 = relu_17**2 - relu_17 = None - mul_29 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - * pow_18 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( - pow_18 - ) = None - x_176 = ( - mul_29 - + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - ) - mul_29 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) - x_177 = torch.nn.functional.dropout(x_176, 0.0, False, False) - x_176 = None - x_178 = torch._C._nn.linear( - x_177, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_177 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) - x_178 = None - x_180 = mul_28 + x_179 - mul_28 = x_179 = None - view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_30 = x_180 * view_12 - view_12 = None - layer_norm_27 = torch.nn.functional.layer_norm( - x_180, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_180 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) - linear_24 = torch._C._nn.linear( - layer_norm_27, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_27 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_13 = linear_24.reshape(1, pow_25, 3, 10, 32) - linear_24 = None - qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) - reshape_13 = None - unbind_6 = qkv_6.unbind(0) - qkv_6 = None - q_6 = unbind_6[0] - k_6 = unbind_6[1] - v_6 = unbind_6[2] - unbind_6 = None - x_181 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) - q_6 = k_6 = v_6 = None - transpose_7 = x_181.transpose(1, 2) - x_181 = None - x_182 = transpose_7.reshape(1, pow_25, 320) - transpose_7 = None - x_183 = torch._C._nn.linear( - x_182, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_182 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_184 = torch.nn.functional.dropout(x_183, 0.0, False, False) - x_183 = None - x_185 = mul_30 + x_184 - mul_30 = x_184 = None - view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_31 = x_185 * view_13 - view_13 = None - layer_norm_28 = torch.nn.functional.layer_norm( - x_185, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_185 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) - x_186 = torch._C._nn.linear( - layer_norm_28, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_28 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_18 = torch.nn.functional.relu(x_186, inplace=False) - x_186 = None - pow_19 = relu_18**2 - relu_18 = None - mul_32 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - * pow_19 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( - pow_19 - ) = None - x_187 = ( - mul_32 - + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - ) - mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) - x_188 = torch.nn.functional.dropout(x_187, 0.0, False, False) - x_187 = None - x_189 = torch._C._nn.linear( - x_188, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_188 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) - x_189 = None - x_191 = mul_31 + x_190 - mul_31 = x_190 = None - view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_33 = x_191 * view_14 - view_14 = None - layer_norm_29 = torch.nn.functional.layer_norm( - x_191, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_191 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) - linear_28 = torch._C._nn.linear( - layer_norm_29, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_29 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_15 = linear_28.reshape(1, pow_25, 3, 10, 32) - linear_28 = None - qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) - reshape_15 = None - unbind_7 = qkv_7.unbind(0) - qkv_7 = None - q_7 = unbind_7[0] - k_7 = unbind_7[1] - v_7 = unbind_7[2] - unbind_7 = None - x_192 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) - q_7 = k_7 = v_7 = None - transpose_8 = x_192.transpose(1, 2) - x_192 = None - x_193 = transpose_8.reshape(1, pow_25, 320) - transpose_8 = None - x_194 = torch._C._nn.linear( - x_193, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_193 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_195 = torch.nn.functional.dropout(x_194, 0.0, False, False) - x_194 = None - x_196 = mul_33 + x_195 - mul_33 = x_195 = None - view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_34 = x_196 * view_15 - view_15 = None - layer_norm_30 = torch.nn.functional.layer_norm( - x_196, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_196 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) - x_197 = torch._C._nn.linear( - layer_norm_30, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_30 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_19 = torch.nn.functional.relu(x_197, inplace=False) - x_197 = None - pow_20 = relu_19**2 - relu_19 = None - mul_35 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - * pow_20 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( - pow_20 - ) = None - x_198 = ( - mul_35 - + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - ) - mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) - x_199 = torch.nn.functional.dropout(x_198, 0.0, False, False) - x_198 = None - x_200 = torch._C._nn.linear( - x_199, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_199 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) - x_200 = None - x_202 = mul_34 + x_201 - mul_34 = x_201 = None - view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_36 = x_202 * view_16 - view_16 = None - layer_norm_31 = torch.nn.functional.layer_norm( - x_202, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_202 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) - linear_32 = torch._C._nn.linear( - layer_norm_31, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_31 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_17 = linear_32.reshape(1, pow_25, 3, 10, 32) - linear_32 = None - qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) - reshape_17 = None - unbind_8 = qkv_8.unbind(0) - qkv_8 = None - q_8 = unbind_8[0] - k_8 = unbind_8[1] - v_8 = unbind_8[2] - unbind_8 = None - x_203 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) - q_8 = k_8 = v_8 = None - transpose_9 = x_203.transpose(1, 2) - x_203 = None - x_204 = transpose_9.reshape(1, pow_25, 320) - transpose_9 = pow_25 = None - x_205 = torch._C._nn.linear( - x_204, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_204 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_206 = torch.nn.functional.dropout(x_205, 0.0, False, False) - x_205 = None - x_207 = mul_36 + x_206 - mul_36 = x_206 = None - view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_37 = x_207 * view_17 - view_17 = None - layer_norm_32 = torch.nn.functional.layer_norm( - x_207, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_207 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) - x_208 = torch._C._nn.linear( - layer_norm_32, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_32 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_20 = torch.nn.functional.relu(x_208, inplace=False) - x_208 = None - pow_21 = relu_20**2 - relu_20 = None - mul_38 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - * pow_21 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( - pow_21 - ) = None - x_209 = ( - mul_38 - + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - ) - mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) - x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) - x_209 = None - x_211 = torch._C._nn.linear( - x_210, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_210 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) - x_211 = None - x_213 = mul_37 + x_212 - mul_37 = x_212 = None - transpose_10 = x_213.transpose(1, 2) - x_213 = None - x_214 = transpose_10.reshape(1, 320, sym_sum_3, sym_sum_3) - transpose_10 = sym_sum_3 = None - x_215 = x_214.permute(0, 2, 3, 1) - x_214 = None - x_216 = torch.nn.functional.layer_norm( - x_215, - (320,), - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_215 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) - x_217 = x_216.permute(0, 3, 1, 2) - x_216 = None - x_218 = torch.conv2d( - x_217, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_217 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_3 = sym_sum // 32 - sym_sum = None - sym_sum_4 = torch.sym_sum([1, floordiv_3]) - floordiv_3 = None - reshape_20 = x_218.reshape(1, 512, -1) - x_218 = None - x_219 = reshape_20.transpose(1, 2) - reshape_20 = None - view_18 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_39 = x_219 * view_18 - view_18 = None - layer_norm_34 = torch.nn.functional.layer_norm( - x_219, - (512,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_219 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_26 = sym_sum_4**2 - linear_36 = torch._C._nn.linear( - layer_norm_34, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_34 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_21 = linear_36.reshape(1, pow_26, 3, 16, 32) - linear_36 = None - qkv_9 = reshape_21.permute(2, 0, 3, 1, 4) - reshape_21 = None - unbind_9 = qkv_9.unbind(0) - qkv_9 = None - q_9 = unbind_9[0] - k_9 = unbind_9[1] - v_9 = unbind_9[2] - unbind_9 = None - x_220 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) - q_9 = k_9 = v_9 = None - transpose_12 = x_220.transpose(1, 2) - x_220 = None - x_221 = transpose_12.reshape(1, pow_26, 512) - transpose_12 = None - x_222 = torch._C._nn.linear( - x_221, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_221 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) - x_222 = None - x_224 = mul_39 + x_223 - mul_39 = x_223 = None - view_19 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_40 = x_224 * view_19 - view_19 = None - layer_norm_35 = torch.nn.functional.layer_norm( - x_224, - (512,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_224 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_225 = torch._C._nn.linear( - layer_norm_35, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_35 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_21 = torch.nn.functional.relu(x_225, inplace=False) - x_225 = None - pow_22 = relu_21**2 - relu_21 = None - mul_41 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_22 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_22 - ) = None - x_226 = ( - mul_41 - + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_41 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) - x_226 = None - x_228 = torch._C._nn.linear( - x_227, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_227 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_229 = torch.nn.functional.dropout(x_228, 0.0, False, False) - x_228 = None - x_230 = mul_40 + x_229 - mul_40 = x_229 = None - view_20 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_42 = x_230 * view_20 - view_20 = None - layer_norm_36 = torch.nn.functional.layer_norm( - x_230, - (512,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_230 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_40 = torch._C._nn.linear( - layer_norm_36, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_36 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_23 = linear_40.reshape(1, pow_26, 3, 16, 32) - linear_40 = None - qkv_10 = reshape_23.permute(2, 0, 3, 1, 4) - reshape_23 = None - unbind_10 = qkv_10.unbind(0) - qkv_10 = None - q_10 = unbind_10[0] - k_10 = unbind_10[1] - v_10 = unbind_10[2] - unbind_10 = None - x_231 = torch._C._nn.scaled_dot_product_attention( - q_10, k_10, v_10, dropout_p=0.0 - ) - q_10 = k_10 = v_10 = None - transpose_13 = x_231.transpose(1, 2) - x_231 = None - x_232 = transpose_13.reshape(1, pow_26, 512) - transpose_13 = None - x_233 = torch._C._nn.linear( - x_232, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_232 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) - x_233 = None - x_235 = mul_42 + x_234 - mul_42 = x_234 = None - view_21 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_43 = x_235 * view_21 - view_21 = None - layer_norm_37 = torch.nn.functional.layer_norm( - x_235, - (512,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_235 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_236 = torch._C._nn.linear( - layer_norm_37, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_37 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_22 = torch.nn.functional.relu(x_236, inplace=False) - x_236 = None - pow_23 = relu_22**2 - relu_22 = None - mul_44 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_23 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_23 - ) = None - x_237 = ( - mul_44 - + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_44 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_238 = torch.nn.functional.dropout(x_237, 0.0, False, False) - x_237 = None - x_239 = torch._C._nn.linear( - x_238, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_238 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_240 = torch.nn.functional.dropout(x_239, 0.0, False, False) - x_239 = None - x_241 = mul_43 + x_240 - mul_43 = x_240 = None - view_22 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_45 = x_241 * view_22 - view_22 = None - layer_norm_38 = torch.nn.functional.layer_norm( - x_241, - (512,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_241 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_44 = torch._C._nn.linear( - layer_norm_38, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_38 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_25 = linear_44.reshape(1, pow_26, 3, 16, 32) - linear_44 = None - qkv_11 = reshape_25.permute(2, 0, 3, 1, 4) - reshape_25 = None - unbind_11 = qkv_11.unbind(0) - qkv_11 = None - q_11 = unbind_11[0] - k_11 = unbind_11[1] - v_11 = unbind_11[2] - unbind_11 = None - x_242 = torch._C._nn.scaled_dot_product_attention( - q_11, k_11, v_11, dropout_p=0.0 - ) - q_11 = k_11 = v_11 = None - transpose_14 = x_242.transpose(1, 2) - x_242 = None - x_243 = transpose_14.reshape(1, pow_26, 512) - transpose_14 = pow_26 = None - x_244 = torch._C._nn.linear( - x_243, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_243 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) - x_244 = None - x_246 = mul_45 + x_245 - mul_45 = x_245 = None - view_23 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_46 = x_246 * view_23 - view_23 = None - layer_norm_39 = torch.nn.functional.layer_norm( - x_246, - (512,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_246 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_247 = torch._C._nn.linear( - layer_norm_39, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_39 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_23 = torch.nn.functional.relu(x_247, inplace=False) - x_247 = None - pow_24 = relu_23**2 - relu_23 = None - mul_47 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_24 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_24 - ) = None - x_248 = ( - mul_47 - + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_47 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_249 = torch.nn.functional.dropout(x_248, 0.0, False, False) - x_248 = None - x_250 = torch._C._nn.linear( - x_249, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_249 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_251 = torch.nn.functional.dropout(x_250, 0.0, False, False) - x_250 = None - x_252 = mul_46 + x_251 - mul_46 = x_251 = None - transpose_15 = x_252.transpose(1, 2) - x_252 = None - x_253 = transpose_15.reshape(1, 512, sym_sum_4, sym_sum_4) - transpose_15 = sym_sum_4 = None - x_254 = torch.nn.functional.adaptive_avg_pool2d(x_253, 1) - x_253 = None - x_255 = x_254.permute(0, 2, 3, 1) - x_254 = None - x_256 = torch.nn.functional.layer_norm( - x_255, - (512,), - l_self_modules_head_modules_norm_parameters_weight_, - l_self_modules_head_modules_norm_parameters_bias_, - 1e-06, - ) - x_255 = ( - l_self_modules_head_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_norm_parameters_bias_ = None - x_257 = x_256.permute(0, 3, 1, 2) - x_256 = None - x_258 = x_257.flatten(1, -1) - x_257 = None - x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) - x_258 = None - x_260 = torch._C._nn.linear( - x_259, - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, - ) - x_259 = ( - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None - relu_24 = torch.nn.functional.relu(x_260, inplace=False) - x_260 = None - x_261 = torch.square(relu_24) - relu_24 = None - x_262 = torch.nn.functional.layer_norm( - x_261, - (2048,), - l_self_modules_head_modules_fc_modules_norm_parameters_weight_, - l_self_modules_head_modules_fc_modules_norm_parameters_bias_, - 1e-06, - ) - x_261 = ( - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None - x_263 = torch.nn.functional.dropout(x_262, 0.0, False, False) - x_262 = None - x_264 = torch._C._nn.linear( - x_263, - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, - ) - x_263 = ( - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None - return (x_264,) diff --git a/samples/timm/caformer_s18.sail_in1k_384/weight_meta.py b/samples/timm/caformer_s18.sail_in1k_384/weight_meta.py deleted file mode 100644 index 51484b5182..0000000000 --- a/samples/timm/caformer_s18.sail_in1k_384/weight_meta.py +++ /dev/null @@ -1,2078 +0,0 @@ -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: - name = "L_self_modules_stem_modules_conv_parameters_weight_" - shape = [64, 3, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: - name = "L_self_modules_stem_modules_conv_parameters_bias_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_s1: - name = "s1" - shape = [] - dtype = "torch.int64" - device = "cpu" - mean = None - std = None - data = [4] - - -class Program_weight_tensor_meta_L_x_: - name = "L_x_" - shape = [1, 3, 224, 224] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.227 - std = 1.288 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: - name = "L_self_modules_stem_modules_norm_parameters_weight_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [128, 64, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [128, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [64, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [256, 64, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [64, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [128, 64, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [128, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [64, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [256, 64, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [64, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [128, 64, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [128, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.001 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [64, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [256, 64, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [64, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" - shape = [128, 64, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [256, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [256, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [128, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [512, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [128, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [256, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [256, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [128, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [512, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [128, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [256, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [256, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [128, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [512, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [128, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" - shape = [320, 128, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" - shape = [512, 320, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_norm_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_norm_parameters_bias_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" - shape = [2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" - shape = [2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" - shape = [2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" - shape = [1000, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" - shape = [1000] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None diff --git a/samples/timm/caformer_s18.sail_in22k/graph_net.json b/samples/timm/caformer_s18.sail_in22k/graph_net.json deleted file mode 100644 index 1373fe3b58..0000000000 --- a/samples/timm/caformer_s18.sail_in22k/graph_net.json +++ /dev/null @@ -1,5 +0,0 @@ -{ - "framework": "torch", - "num_devices_required": 1, - "num_nodes_required": 1 -} \ No newline at end of file diff --git a/samples/timm/caformer_s18.sail_in22k/input_meta.py b/samples/timm/caformer_s18.sail_in22k/input_meta.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_s18.sail_in22k/input_tensor_constraints.py b/samples/timm/caformer_s18.sail_in22k/input_tensor_constraints.py deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/samples/timm/caformer_s18.sail_in22k/model.py b/samples/timm/caformer_s18.sail_in22k/model.py deleted file mode 100644 index ba7a8ecdba..0000000000 --- a/samples/timm/caformer_s18.sail_in22k/model.py +++ /dev/null @@ -1,2468 +0,0 @@ -import torch - - -class GraphModule(torch.nn.Module): - def forward( - self, - L_self_modules_stem_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stem_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - s1: torch.SymInt, - L_x_: torch.Tensor, - L_self_modules_stem_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - ): - l_self_modules_stem_modules_conv_parameters_weight_ = ( - L_self_modules_stem_modules_conv_parameters_weight_ - ) - l_self_modules_stem_modules_conv_parameters_bias_ = ( - L_self_modules_stem_modules_conv_parameters_bias_ - ) - l_x_ = L_x_ - l_self_modules_stem_modules_norm_parameters_weight_ = ( - L_self_modules_stem_modules_norm_parameters_weight_ - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - l_self_modules_head_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc1_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_norm_parameters_bias_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = ( - L_self_modules_head_modules_fc_modules_fc2_parameters_bias_ - ) - x = torch.conv2d( - l_x_, - l_self_modules_stem_modules_conv_parameters_weight_, - l_self_modules_stem_modules_conv_parameters_bias_, - (4, 4), - (2, 2), - (1, 1), - 1, - ) - l_x_ = ( - l_self_modules_stem_modules_conv_parameters_weight_ - ) = l_self_modules_stem_modules_conv_parameters_bias_ = None - x_1 = x.permute(0, 2, 3, 1) - x = None - x_2 = torch.nn.functional.layer_norm( - x_1, (64,), l_self_modules_stem_modules_norm_parameters_weight_, None, 1e-06 - ) - x_1 = l_self_modules_stem_modules_norm_parameters_weight_ = None - x_3 = x_2.permute(0, 3, 1, 2) - x_2 = None - sym_sum = torch.sym_sum([-3, s1]) - s1 = None - floordiv = sym_sum // 4 - sym_sum_1 = torch.sym_sum([1, floordiv]) - floordiv = sym_sum_1 = None - x_4 = x_3.permute(0, 2, 3, 1) - x_5 = torch.nn.functional.layer_norm( - x_4, - (64,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_6 = x_5.permute(0, 3, 1, 2) - x_5 = None - x_7 = torch.conv2d( - x_6, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu = torch.nn.functional.relu(x_7, inplace=False) - x_7 = None - pow_1 = relu**2 - relu = None - mul = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_1 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_1 - ) = None - x_8 = ( - mul - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_9 = torch.conv2d( - x_8, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 128, - ) - x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_10 = torch.conv2d( - x_9, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_11 = x_3 + x_10 - x_3 = x_10 = None - x_12 = x_11.permute(0, 2, 3, 1) - x_13 = torch.nn.functional.layer_norm( - x_12, - (64,), - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_14 = x_13.permute(0, 3, 1, 2) - x_13 = None - x_15 = torch.conv2d( - x_14, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_1 = torch.nn.functional.relu(x_15, inplace=False) - x_15 = None - pow_2 = relu_1**2 - relu_1 = None - mul_1 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_2 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_2 - ) = None - x_16 = ( - mul_1 - + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) - x_16 = None - x_18 = torch.conv2d( - x_17, - l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) - x_18 = None - x_20 = x_11 + x_19 - x_11 = x_19 = None - x_21 = x_20.permute(0, 2, 3, 1) - x_22 = torch.nn.functional.layer_norm( - x_21, - (64,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_23 = x_22.permute(0, 3, 1, 2) - x_22 = None - x_24 = torch.conv2d( - x_23, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_2 = torch.nn.functional.relu(x_24, inplace=False) - x_24 = None - pow_3 = relu_2**2 - relu_2 = None - mul_2 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_3 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_3 - ) = None - x_25 = ( - mul_2 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_26 = torch.conv2d( - x_25, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 128, - ) - x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_27 = torch.conv2d( - x_26, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_28 = x_20 + x_27 - x_20 = x_27 = None - x_29 = x_28.permute(0, 2, 3, 1) - x_30 = torch.nn.functional.layer_norm( - x_29, - (64,), - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_31 = x_30.permute(0, 3, 1, 2) - x_30 = None - x_32 = torch.conv2d( - x_31, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_3 = torch.nn.functional.relu(x_32, inplace=False) - x_32 = None - pow_4 = relu_3**2 - relu_3 = None - mul_3 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_4 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_4 - ) = None - x_33 = ( - mul_3 - + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) - x_33 = None - x_35 = torch.conv2d( - x_34, - l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_34 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) - x_35 = None - x_37 = x_28 + x_36 - x_28 = x_36 = None - x_38 = x_37.permute(0, 2, 3, 1) - x_39 = torch.nn.functional.layer_norm( - x_38, - (64,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_38 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_40 = x_39.permute(0, 3, 1, 2) - x_39 = None - x_41 = torch.conv2d( - x_40, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_40 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_4 = torch.nn.functional.relu(x_41, inplace=False) - x_41 = None - pow_5 = relu_4**2 - relu_4 = None - mul_4 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_5 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_5 - ) = None - x_42 = ( - mul_4 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_43 = torch.conv2d( - x_42, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 128, - ) - x_42 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_44 = torch.conv2d( - x_43, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_43 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_45 = x_37 + x_44 - x_37 = x_44 = None - x_46 = x_45.permute(0, 2, 3, 1) - x_47 = torch.nn.functional.layer_norm( - x_46, - (64,), - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_46 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_48 = x_47.permute(0, 3, 1, 2) - x_47 = None - x_49 = torch.conv2d( - x_48, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_48 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_5 = torch.nn.functional.relu(x_49, inplace=False) - x_49 = None - pow_6 = relu_5**2 - relu_5 = None - mul_5 = ( - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_6 - ) - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_6 - ) = None - x_50 = ( - mul_5 - + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) - x_50 = None - x_52 = torch.conv2d( - x_51, - l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_51 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) - x_52 = None - x_54 = x_45 + x_53 - x_45 = x_53 = None - x_55 = x_54.permute(0, 2, 3, 1) - x_54 = None - x_56 = torch.nn.functional.layer_norm( - x_55, - (64,), - l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_55 = l_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_ = (None) - x_57 = x_56.permute(0, 3, 1, 2) - x_56 = None - x_58 = torch.conv2d( - x_57, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_57 = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_1 = sym_sum // 8 - sym_sum_2 = torch.sym_sum([1, floordiv_1]) - floordiv_1 = sym_sum_2 = None - x_59 = x_58.permute(0, 2, 3, 1) - x_60 = torch.nn.functional.layer_norm( - x_59, - (128,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - x_61 = x_60.permute(0, 3, 1, 2) - x_60 = None - x_62 = torch.conv2d( - x_61, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_6 = torch.nn.functional.relu(x_62, inplace=False) - x_62 = None - pow_7 = relu_6**2 - relu_6 = None - mul_6 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ - * pow_7 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_7 - ) = None - x_63 = ( - mul_6 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_64 = torch.conv2d( - x_63, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 256, - ) - x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_65 = torch.conv2d( - x_64, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_66 = x_58 + x_65 - x_58 = x_65 = None - x_67 = x_66.permute(0, 2, 3, 1) - x_68 = torch.nn.functional.layer_norm( - x_67, - (128,), - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_69 = x_68.permute(0, 3, 1, 2) - x_68 = None - x_70 = torch.conv2d( - x_69, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_7 = torch.nn.functional.relu(x_70, inplace=False) - x_70 = None - pow_8 = relu_7**2 - relu_7 = None - mul_7 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_8 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_8 - ) = None - x_71 = ( - mul_7 - + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) - x_71 = None - x_73 = torch.conv2d( - x_72, - l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) - x_73 = None - x_75 = x_66 + x_74 - x_66 = x_74 = None - x_76 = x_75.permute(0, 2, 3, 1) - x_77 = torch.nn.functional.layer_norm( - x_76, - (128,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - x_78 = x_77.permute(0, 3, 1, 2) - x_77 = None - x_79 = torch.conv2d( - x_78, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_78 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_8 = torch.nn.functional.relu(x_79, inplace=False) - x_79 = None - pow_9 = relu_8**2 - relu_8 = None - mul_8 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ - * pow_9 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_9 - ) = None - x_80 = ( - mul_8 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_81 = torch.conv2d( - x_80, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 256, - ) - x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_82 = torch.conv2d( - x_81, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_83 = x_75 + x_82 - x_75 = x_82 = None - x_84 = x_83.permute(0, 2, 3, 1) - x_85 = torch.nn.functional.layer_norm( - x_84, - (128,), - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_86 = x_85.permute(0, 3, 1, 2) - x_85 = None - x_87 = torch.conv2d( - x_86, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_9 = torch.nn.functional.relu(x_87, inplace=False) - x_87 = None - pow_10 = relu_9**2 - relu_9 = None - mul_9 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_10 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_10 - ) = None - x_88 = ( - mul_9 - + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_89 = torch.nn.functional.dropout(x_88, 0.0, False, False) - x_88 = None - x_90 = torch.conv2d( - x_89, - l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) - x_90 = None - x_92 = x_83 + x_91 - x_83 = x_91 = None - x_93 = x_92.permute(0, 2, 3, 1) - x_94 = torch.nn.functional.layer_norm( - x_93, - (128,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - x_95 = x_94.permute(0, 3, 1, 2) - x_94 = None - x_96 = torch.conv2d( - x_95, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_95 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_ = (None) - relu_10 = torch.nn.functional.relu(x_96, inplace=False) - x_96 = None - pow_11 = relu_10**2 - relu_10 = None - mul_10 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ - * pow_11 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_ = ( - pow_11 - ) = None - x_97 = ( - mul_10 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ - ) - mul_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_ = (None) - x_98 = torch.conv2d( - x_97, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_, - None, - (1, 1), - (3, 3), - (1, 1), - 256, - ) - x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_ = (None) - x_99 = torch.conv2d( - x_98, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_98 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_ = (None) - x_100 = x_92 + x_99 - x_92 = x_99 = None - x_101 = x_100.permute(0, 2, 3, 1) - x_102 = torch.nn.functional.layer_norm( - x_101, - (128,), - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_101 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_103 = x_102.permute(0, 3, 1, 2) - x_102 = None - x_104 = torch.conv2d( - x_103, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_103 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_11 = torch.nn.functional.relu(x_104, inplace=False) - x_104 = None - pow_12 = relu_11**2 - relu_11 = None - mul_11 = ( - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_12 - ) - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_12 - ) = None - x_105 = ( - mul_11 - + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) - x_105 = None - x_107 = torch.conv2d( - x_106, - l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - (1, 1), - (0, 0), - (1, 1), - 1, - ) - x_106 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) - x_107 = None - x_109 = x_100 + x_108 - x_100 = x_108 = None - x_110 = x_109.permute(0, 2, 3, 1) - x_109 = None - x_111 = torch.nn.functional.layer_norm( - x_110, - (128,), - l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_110 = l_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_ = (None) - x_112 = x_111.permute(0, 3, 1, 2) - x_111 = None - x_113 = torch.conv2d( - x_112, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_112 = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_2 = sym_sum // 16 - sym_sum_3 = torch.sym_sum([1, floordiv_2]) - floordiv_2 = None - reshape = x_113.reshape(1, 320, -1) - x_113 = None - x_114 = reshape.transpose(1, 2) - reshape = None - view = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_12 = x_114 * view - view = None - layer_norm_15 = torch.nn.functional.layer_norm( - x_114, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_25 = sym_sum_3**2 - linear = torch._C._nn.linear( - layer_norm_15, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_15 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_1 = linear.reshape(1, pow_25, 3, 10, 32) - linear = None - qkv = reshape_1.permute(2, 0, 3, 1, 4) - reshape_1 = None - unbind = qkv.unbind(0) - qkv = None - q = unbind[0] - k = unbind[1] - v = unbind[2] - unbind = None - x_115 = torch._C._nn.scaled_dot_product_attention(q, k, v, dropout_p=0.0) - q = k = v = None - transpose_1 = x_115.transpose(1, 2) - x_115 = None - x_116 = transpose_1.reshape(1, pow_25, 320) - transpose_1 = None - x_117 = torch._C._nn.linear( - x_116, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_116 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) - x_117 = None - x_119 = mul_12 + x_118 - mul_12 = x_118 = None - view_1 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_13 = x_119 * view_1 - view_1 = None - layer_norm_16 = torch.nn.functional.layer_norm( - x_119, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_119 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_120 = torch._C._nn.linear( - layer_norm_16, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_16 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_12 = torch.nn.functional.relu(x_120, inplace=False) - x_120 = None - pow_13 = relu_12**2 - relu_12 = None - mul_14 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_13 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_13 - ) = None - x_121 = ( - mul_14 - + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_14 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_122 = torch.nn.functional.dropout(x_121, 0.0, False, False) - x_121 = None - x_123 = torch._C._nn.linear( - x_122, - l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) - x_123 = None - x_125 = mul_13 + x_124 - mul_13 = x_124 = None - view_2 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_15 = x_125 * view_2 - view_2 = None - layer_norm_17 = torch.nn.functional.layer_norm( - x_125, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_4 = torch._C._nn.linear( - layer_norm_17, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_17 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_3 = linear_4.reshape(1, pow_25, 3, 10, 32) - linear_4 = None - qkv_1 = reshape_3.permute(2, 0, 3, 1, 4) - reshape_3 = None - unbind_1 = qkv_1.unbind(0) - qkv_1 = None - q_1 = unbind_1[0] - k_1 = unbind_1[1] - v_1 = unbind_1[2] - unbind_1 = None - x_126 = torch._C._nn.scaled_dot_product_attention(q_1, k_1, v_1, dropout_p=0.0) - q_1 = k_1 = v_1 = None - transpose_2 = x_126.transpose(1, 2) - x_126 = None - x_127 = transpose_2.reshape(1, pow_25, 320) - transpose_2 = None - x_128 = torch._C._nn.linear( - x_127, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_127 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_129 = torch.nn.functional.dropout(x_128, 0.0, False, False) - x_128 = None - x_130 = mul_15 + x_129 - mul_15 = x_129 = None - view_3 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_16 = x_130 * view_3 - view_3 = None - layer_norm_18 = torch.nn.functional.layer_norm( - x_130, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_130 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_131 = torch._C._nn.linear( - layer_norm_18, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_18 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_13 = torch.nn.functional.relu(x_131, inplace=False) - x_131 = None - pow_14 = relu_13**2 - relu_13 = None - mul_17 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_14 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_14 - ) = None - x_132 = ( - mul_17 - + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_17 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) - x_132 = None - x_134 = torch._C._nn.linear( - x_133, - l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_133 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) - x_134 = None - x_136 = mul_16 + x_135 - mul_16 = x_135 = None - view_4 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_18 = x_136 * view_4 - view_4 = None - layer_norm_19 = torch.nn.functional.layer_norm( - x_136, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_8 = torch._C._nn.linear( - layer_norm_19, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_19 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_5 = linear_8.reshape(1, pow_25, 3, 10, 32) - linear_8 = None - qkv_2 = reshape_5.permute(2, 0, 3, 1, 4) - reshape_5 = None - unbind_2 = qkv_2.unbind(0) - qkv_2 = None - q_2 = unbind_2[0] - k_2 = unbind_2[1] - v_2 = unbind_2[2] - unbind_2 = None - x_137 = torch._C._nn.scaled_dot_product_attention(q_2, k_2, v_2, dropout_p=0.0) - q_2 = k_2 = v_2 = None - transpose_3 = x_137.transpose(1, 2) - x_137 = None - x_138 = transpose_3.reshape(1, pow_25, 320) - transpose_3 = None - x_139 = torch._C._nn.linear( - x_138, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_138 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) - x_139 = None - x_141 = mul_18 + x_140 - mul_18 = x_140 = None - view_5 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_19 = x_141 * view_5 - view_5 = None - layer_norm_20 = torch.nn.functional.layer_norm( - x_141, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_141 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_142 = torch._C._nn.linear( - layer_norm_20, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_20 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_14 = torch.nn.functional.relu(x_142, inplace=False) - x_142 = None - pow_15 = relu_14**2 - relu_14 = None - mul_20 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_15 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_15 - ) = None - x_143 = ( - mul_20 - + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_20 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) - x_143 = None - x_145 = torch._C._nn.linear( - x_144, - l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_144 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) - x_145 = None - x_147 = mul_19 + x_146 - mul_19 = x_146 = None - view_6 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_21 = x_147 * view_6 - view_6 = None - layer_norm_21 = torch.nn.functional.layer_norm( - x_147, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_147 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_ = (None) - linear_12 = torch._C._nn.linear( - layer_norm_21, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_21 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_7 = linear_12.reshape(1, pow_25, 3, 10, 32) - linear_12 = None - qkv_3 = reshape_7.permute(2, 0, 3, 1, 4) - reshape_7 = None - unbind_3 = qkv_3.unbind(0) - qkv_3 = None - q_3 = unbind_3[0] - k_3 = unbind_3[1] - v_3 = unbind_3[2] - unbind_3 = None - x_148 = torch._C._nn.scaled_dot_product_attention(q_3, k_3, v_3, dropout_p=0.0) - q_3 = k_3 = v_3 = None - transpose_4 = x_148.transpose(1, 2) - x_148 = None - x_149 = transpose_4.reshape(1, pow_25, 320) - transpose_4 = None - x_150 = torch._C._nn.linear( - x_149, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_149 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_151 = torch.nn.functional.dropout(x_150, 0.0, False, False) - x_150 = None - x_152 = mul_21 + x_151 - mul_21 = x_151 = None - view_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_22 = x_152 * view_7 - view_7 = None - layer_norm_22 = torch.nn.functional.layer_norm( - x_152, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_152 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_ = (None) - x_153 = torch._C._nn.linear( - layer_norm_22, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_22 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_15 = torch.nn.functional.relu(x_153, inplace=False) - x_153 = None - pow_16 = relu_15**2 - relu_15 = None - mul_23 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ - * pow_16 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_ = ( - pow_16 - ) = None - x_154 = ( - mul_23 - + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ - ) - mul_23 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_ = (None) - x_155 = torch.nn.functional.dropout(x_154, 0.0, False, False) - x_154 = None - x_156 = torch._C._nn.linear( - x_155, - l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_155 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) - x_156 = None - x_158 = mul_22 + x_157 - mul_22 = x_157 = None - view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_24 = x_158 * view_8 - view_8 = None - layer_norm_23 = torch.nn.functional.layer_norm( - x_158, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_ = (None) - linear_16 = torch._C._nn.linear( - layer_norm_23, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_23 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_9 = linear_16.reshape(1, pow_25, 3, 10, 32) - linear_16 = None - qkv_4 = reshape_9.permute(2, 0, 3, 1, 4) - reshape_9 = None - unbind_4 = qkv_4.unbind(0) - qkv_4 = None - q_4 = unbind_4[0] - k_4 = unbind_4[1] - v_4 = unbind_4[2] - unbind_4 = None - x_159 = torch._C._nn.scaled_dot_product_attention(q_4, k_4, v_4, dropout_p=0.0) - q_4 = k_4 = v_4 = None - transpose_5 = x_159.transpose(1, 2) - x_159 = None - x_160 = transpose_5.reshape(1, pow_25, 320) - transpose_5 = None - x_161 = torch._C._nn.linear( - x_160, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_160 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_162 = torch.nn.functional.dropout(x_161, 0.0, False, False) - x_161 = None - x_163 = mul_24 + x_162 - mul_24 = x_162 = None - view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_25 = x_163 * view_9 - view_9 = None - layer_norm_24 = torch.nn.functional.layer_norm( - x_163, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_163 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_ = (None) - x_164 = torch._C._nn.linear( - layer_norm_24, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_24 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_16 = torch.nn.functional.relu(x_164, inplace=False) - x_164 = None - pow_17 = relu_16**2 - relu_16 = None - mul_26 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ - * pow_17 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_ = ( - pow_17 - ) = None - x_165 = ( - mul_26 - + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ - ) - mul_26 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_ = (None) - x_166 = torch.nn.functional.dropout(x_165, 0.0, False, False) - x_165 = None - x_167 = torch._C._nn.linear( - x_166, - l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_166 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) - x_167 = None - x_169 = mul_25 + x_168 - mul_25 = x_168 = None - view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_27 = x_169 * view_10 - view_10 = None - layer_norm_25 = torch.nn.functional.layer_norm( - x_169, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_ = (None) - linear_20 = torch._C._nn.linear( - layer_norm_25, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_25 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_11 = linear_20.reshape(1, pow_25, 3, 10, 32) - linear_20 = None - qkv_5 = reshape_11.permute(2, 0, 3, 1, 4) - reshape_11 = None - unbind_5 = qkv_5.unbind(0) - qkv_5 = None - q_5 = unbind_5[0] - k_5 = unbind_5[1] - v_5 = unbind_5[2] - unbind_5 = None - x_170 = torch._C._nn.scaled_dot_product_attention(q_5, k_5, v_5, dropout_p=0.0) - q_5 = k_5 = v_5 = None - transpose_6 = x_170.transpose(1, 2) - x_170 = None - x_171 = transpose_6.reshape(1, pow_25, 320) - transpose_6 = None - x_172 = torch._C._nn.linear( - x_171, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_171 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_173 = torch.nn.functional.dropout(x_172, 0.0, False, False) - x_172 = None - x_174 = mul_27 + x_173 - mul_27 = x_173 = None - view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_28 = x_174 * view_11 - view_11 = None - layer_norm_26 = torch.nn.functional.layer_norm( - x_174, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_174 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_ = (None) - x_175 = torch._C._nn.linear( - layer_norm_26, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_26 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_17 = torch.nn.functional.relu(x_175, inplace=False) - x_175 = None - pow_18 = relu_17**2 - relu_17 = None - mul_29 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ - * pow_18 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_ = ( - pow_18 - ) = None - x_176 = ( - mul_29 - + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ - ) - mul_29 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_ = (None) - x_177 = torch.nn.functional.dropout(x_176, 0.0, False, False) - x_176 = None - x_178 = torch._C._nn.linear( - x_177, - l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_177 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) - x_178 = None - x_180 = mul_28 + x_179 - mul_28 = x_179 = None - view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_30 = x_180 * view_12 - view_12 = None - layer_norm_27 = torch.nn.functional.layer_norm( - x_180, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_180 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_ = (None) - linear_24 = torch._C._nn.linear( - layer_norm_27, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_27 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_13 = linear_24.reshape(1, pow_25, 3, 10, 32) - linear_24 = None - qkv_6 = reshape_13.permute(2, 0, 3, 1, 4) - reshape_13 = None - unbind_6 = qkv_6.unbind(0) - qkv_6 = None - q_6 = unbind_6[0] - k_6 = unbind_6[1] - v_6 = unbind_6[2] - unbind_6 = None - x_181 = torch._C._nn.scaled_dot_product_attention(q_6, k_6, v_6, dropout_p=0.0) - q_6 = k_6 = v_6 = None - transpose_7 = x_181.transpose(1, 2) - x_181 = None - x_182 = transpose_7.reshape(1, pow_25, 320) - transpose_7 = None - x_183 = torch._C._nn.linear( - x_182, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_182 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_184 = torch.nn.functional.dropout(x_183, 0.0, False, False) - x_183 = None - x_185 = mul_30 + x_184 - mul_30 = x_184 = None - view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_31 = x_185 * view_13 - view_13 = None - layer_norm_28 = torch.nn.functional.layer_norm( - x_185, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_185 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_ = (None) - x_186 = torch._C._nn.linear( - layer_norm_28, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_28 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_18 = torch.nn.functional.relu(x_186, inplace=False) - x_186 = None - pow_19 = relu_18**2 - relu_18 = None - mul_32 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ - * pow_19 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_ = ( - pow_19 - ) = None - x_187 = ( - mul_32 - + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ - ) - mul_32 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_ = (None) - x_188 = torch.nn.functional.dropout(x_187, 0.0, False, False) - x_187 = None - x_189 = torch._C._nn.linear( - x_188, - l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_188 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) - x_189 = None - x_191 = mul_31 + x_190 - mul_31 = x_190 = None - view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_33 = x_191 * view_14 - view_14 = None - layer_norm_29 = torch.nn.functional.layer_norm( - x_191, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_191 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_ = (None) - linear_28 = torch._C._nn.linear( - layer_norm_29, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_29 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_15 = linear_28.reshape(1, pow_25, 3, 10, 32) - linear_28 = None - qkv_7 = reshape_15.permute(2, 0, 3, 1, 4) - reshape_15 = None - unbind_7 = qkv_7.unbind(0) - qkv_7 = None - q_7 = unbind_7[0] - k_7 = unbind_7[1] - v_7 = unbind_7[2] - unbind_7 = None - x_192 = torch._C._nn.scaled_dot_product_attention(q_7, k_7, v_7, dropout_p=0.0) - q_7 = k_7 = v_7 = None - transpose_8 = x_192.transpose(1, 2) - x_192 = None - x_193 = transpose_8.reshape(1, pow_25, 320) - transpose_8 = None - x_194 = torch._C._nn.linear( - x_193, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_193 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_195 = torch.nn.functional.dropout(x_194, 0.0, False, False) - x_194 = None - x_196 = mul_33 + x_195 - mul_33 = x_195 = None - view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_34 = x_196 * view_15 - view_15 = None - layer_norm_30 = torch.nn.functional.layer_norm( - x_196, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_196 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_ = (None) - x_197 = torch._C._nn.linear( - layer_norm_30, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_30 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_19 = torch.nn.functional.relu(x_197, inplace=False) - x_197 = None - pow_20 = relu_19**2 - relu_19 = None - mul_35 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ - * pow_20 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_ = ( - pow_20 - ) = None - x_198 = ( - mul_35 - + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ - ) - mul_35 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_ = (None) - x_199 = torch.nn.functional.dropout(x_198, 0.0, False, False) - x_198 = None - x_200 = torch._C._nn.linear( - x_199, - l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_199 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) - x_200 = None - x_202 = mul_34 + x_201 - mul_34 = x_201 = None - view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_36 = x_202 * view_16 - view_16 = None - layer_norm_31 = torch.nn.functional.layer_norm( - x_202, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_202 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_ = (None) - linear_32 = torch._C._nn.linear( - layer_norm_31, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_31 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_17 = linear_32.reshape(1, pow_25, 3, 10, 32) - linear_32 = None - qkv_8 = reshape_17.permute(2, 0, 3, 1, 4) - reshape_17 = None - unbind_8 = qkv_8.unbind(0) - qkv_8 = None - q_8 = unbind_8[0] - k_8 = unbind_8[1] - v_8 = unbind_8[2] - unbind_8 = None - x_203 = torch._C._nn.scaled_dot_product_attention(q_8, k_8, v_8, dropout_p=0.0) - q_8 = k_8 = v_8 = None - transpose_9 = x_203.transpose(1, 2) - x_203 = None - x_204 = transpose_9.reshape(1, pow_25, 320) - transpose_9 = pow_25 = None - x_205 = torch._C._nn.linear( - x_204, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_204 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_206 = torch.nn.functional.dropout(x_205, 0.0, False, False) - x_205 = None - x_207 = mul_36 + x_206 - mul_36 = x_206 = None - view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_.view( - (320,) - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_37 = x_207 * view_17 - view_17 = None - layer_norm_32 = torch.nn.functional.layer_norm( - x_207, - (320,), - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_207 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_ = (None) - x_208 = torch._C._nn.linear( - layer_norm_32, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_32 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_20 = torch.nn.functional.relu(x_208, inplace=False) - x_208 = None - pow_21 = relu_20**2 - relu_20 = None - mul_38 = ( - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ - * pow_21 - ) - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_ = ( - pow_21 - ) = None - x_209 = ( - mul_38 - + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ - ) - mul_38 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_ = (None) - x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) - x_209 = None - x_211 = torch._C._nn.linear( - x_210, - l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_210 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) - x_211 = None - x_213 = mul_37 + x_212 - mul_37 = x_212 = None - transpose_10 = x_213.transpose(1, 2) - x_213 = None - x_214 = transpose_10.reshape(1, 320, sym_sum_3, sym_sum_3) - transpose_10 = sym_sum_3 = None - x_215 = x_214.permute(0, 2, 3, 1) - x_214 = None - x_216 = torch.nn.functional.layer_norm( - x_215, - (320,), - l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_, - None, - 1e-06, - ) - x_215 = l_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_ = (None) - x_217 = x_216.permute(0, 3, 1, 2) - x_216 = None - x_218 = torch.conv2d( - x_217, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_, - l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_, - (2, 2), - (1, 1), - (1, 1), - 1, - ) - x_217 = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_ = (None) - floordiv_3 = sym_sum // 32 - sym_sum = None - sym_sum_4 = torch.sym_sum([1, floordiv_3]) - floordiv_3 = None - reshape_20 = x_218.reshape(1, 512, -1) - x_218 = None - x_219 = reshape_20.transpose(1, 2) - reshape_20 = None - view_18 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_39 = x_219 * view_18 - view_18 = None - layer_norm_34 = torch.nn.functional.layer_norm( - x_219, - (512,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_219 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_ = (None) - pow_26 = sym_sum_4**2 - linear_36 = torch._C._nn.linear( - layer_norm_34, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_34 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_21 = linear_36.reshape(1, pow_26, 3, 16, 32) - linear_36 = None - qkv_9 = reshape_21.permute(2, 0, 3, 1, 4) - reshape_21 = None - unbind_9 = qkv_9.unbind(0) - qkv_9 = None - q_9 = unbind_9[0] - k_9 = unbind_9[1] - v_9 = unbind_9[2] - unbind_9 = None - x_220 = torch._C._nn.scaled_dot_product_attention(q_9, k_9, v_9, dropout_p=0.0) - q_9 = k_9 = v_9 = None - transpose_12 = x_220.transpose(1, 2) - x_220 = None - x_221 = transpose_12.reshape(1, pow_26, 512) - transpose_12 = None - x_222 = torch._C._nn.linear( - x_221, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_221 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) - x_222 = None - x_224 = mul_39 + x_223 - mul_39 = x_223 = None - view_19 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_40 = x_224 * view_19 - view_19 = None - layer_norm_35 = torch.nn.functional.layer_norm( - x_224, - (512,), - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_224 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_ = (None) - x_225 = torch._C._nn.linear( - layer_norm_35, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_35 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_21 = torch.nn.functional.relu(x_225, inplace=False) - x_225 = None - pow_22 = relu_21**2 - relu_21 = None - mul_41 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ - * pow_22 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_ = ( - pow_22 - ) = None - x_226 = ( - mul_41 - + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ - ) - mul_41 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_ = (None) - x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) - x_226 = None - x_228 = torch._C._nn.linear( - x_227, - l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_227 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_229 = torch.nn.functional.dropout(x_228, 0.0, False, False) - x_228 = None - x_230 = mul_40 + x_229 - mul_40 = x_229 = None - view_20 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_42 = x_230 * view_20 - view_20 = None - layer_norm_36 = torch.nn.functional.layer_norm( - x_230, - (512,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_230 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_ = (None) - linear_40 = torch._C._nn.linear( - layer_norm_36, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_36 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_23 = linear_40.reshape(1, pow_26, 3, 16, 32) - linear_40 = None - qkv_10 = reshape_23.permute(2, 0, 3, 1, 4) - reshape_23 = None - unbind_10 = qkv_10.unbind(0) - qkv_10 = None - q_10 = unbind_10[0] - k_10 = unbind_10[1] - v_10 = unbind_10[2] - unbind_10 = None - x_231 = torch._C._nn.scaled_dot_product_attention( - q_10, k_10, v_10, dropout_p=0.0 - ) - q_10 = k_10 = v_10 = None - transpose_13 = x_231.transpose(1, 2) - x_231 = None - x_232 = transpose_13.reshape(1, pow_26, 512) - transpose_13 = None - x_233 = torch._C._nn.linear( - x_232, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_232 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) - x_233 = None - x_235 = mul_42 + x_234 - mul_42 = x_234 = None - view_21 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_43 = x_235 * view_21 - view_21 = None - layer_norm_37 = torch.nn.functional.layer_norm( - x_235, - (512,), - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_235 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_ = (None) - x_236 = torch._C._nn.linear( - layer_norm_37, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_37 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_22 = torch.nn.functional.relu(x_236, inplace=False) - x_236 = None - pow_23 = relu_22**2 - relu_22 = None - mul_44 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ - * pow_23 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_ = ( - pow_23 - ) = None - x_237 = ( - mul_44 - + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ - ) - mul_44 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_ = (None) - x_238 = torch.nn.functional.dropout(x_237, 0.0, False, False) - x_237 = None - x_239 = torch._C._nn.linear( - x_238, - l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_238 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_240 = torch.nn.functional.dropout(x_239, 0.0, False, False) - x_239 = None - x_241 = mul_43 + x_240 - mul_43 = x_240 = None - view_22 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_ = ( - None - ) - mul_45 = x_241 * view_22 - view_22 = None - layer_norm_38 = torch.nn.functional.layer_norm( - x_241, - (512,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_, - None, - 1e-06, - ) - x_241 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_ = (None) - linear_44 = torch._C._nn.linear( - layer_norm_38, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_, - None, - ) - layer_norm_38 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_ = (None) - reshape_25 = linear_44.reshape(1, pow_26, 3, 16, 32) - linear_44 = None - qkv_11 = reshape_25.permute(2, 0, 3, 1, 4) - reshape_25 = None - unbind_11 = qkv_11.unbind(0) - qkv_11 = None - q_11 = unbind_11[0] - k_11 = unbind_11[1] - v_11 = unbind_11[2] - unbind_11 = None - x_242 = torch._C._nn.scaled_dot_product_attention( - q_11, k_11, v_11, dropout_p=0.0 - ) - q_11 = k_11 = v_11 = None - transpose_14 = x_242.transpose(1, 2) - x_242 = None - x_243 = transpose_14.reshape(1, pow_26, 512) - transpose_14 = pow_26 = None - x_244 = torch._C._nn.linear( - x_243, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_, - None, - ) - x_243 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_ = (None) - x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) - x_244 = None - x_246 = mul_45 + x_245 - mul_45 = x_245 = None - view_23 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_.view( - (512,) - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_ = ( - None - ) - mul_46 = x_246 * view_23 - view_23 = None - layer_norm_39 = torch.nn.functional.layer_norm( - x_246, - (512,), - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_, - None, - 1e-06, - ) - x_246 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_ = (None) - x_247 = torch._C._nn.linear( - layer_norm_39, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - None, - ) - layer_norm_39 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = (None) - relu_23 = torch.nn.functional.relu(x_247, inplace=False) - x_247 = None - pow_24 = relu_23**2 - relu_23 = None - mul_47 = ( - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ - * pow_24 - ) - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_ = ( - pow_24 - ) = None - x_248 = ( - mul_47 - + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ - ) - mul_47 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_ = (None) - x_249 = torch.nn.functional.dropout(x_248, 0.0, False, False) - x_248 = None - x_250 = torch._C._nn.linear( - x_249, - l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - None, - ) - x_249 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = (None) - x_251 = torch.nn.functional.dropout(x_250, 0.0, False, False) - x_250 = None - x_252 = mul_46 + x_251 - mul_46 = x_251 = None - transpose_15 = x_252.transpose(1, 2) - x_252 = None - x_253 = transpose_15.reshape(1, 512, sym_sum_4, sym_sum_4) - transpose_15 = sym_sum_4 = None - x_254 = torch.nn.functional.adaptive_avg_pool2d(x_253, 1) - x_253 = None - x_255 = x_254.permute(0, 2, 3, 1) - x_254 = None - x_256 = torch.nn.functional.layer_norm( - x_255, - (512,), - l_self_modules_head_modules_norm_parameters_weight_, - l_self_modules_head_modules_norm_parameters_bias_, - 1e-06, - ) - x_255 = ( - l_self_modules_head_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_norm_parameters_bias_ = None - x_257 = x_256.permute(0, 3, 1, 2) - x_256 = None - x_258 = x_257.flatten(1, -1) - x_257 = None - x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) - x_258 = None - x_260 = torch._C._nn.linear( - x_259, - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc1_parameters_bias_, - ) - x_259 = ( - l_self_modules_head_modules_fc_modules_fc1_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc1_parameters_bias_ = None - relu_24 = torch.nn.functional.relu(x_260, inplace=False) - x_260 = None - x_261 = torch.square(relu_24) - relu_24 = None - x_262 = torch.nn.functional.layer_norm( - x_261, - (2048,), - l_self_modules_head_modules_fc_modules_norm_parameters_weight_, - l_self_modules_head_modules_fc_modules_norm_parameters_bias_, - 1e-06, - ) - x_261 = ( - l_self_modules_head_modules_fc_modules_norm_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_norm_parameters_bias_ = None - x_263 = torch.nn.functional.dropout(x_262, 0.0, False, False) - x_262 = None - x_264 = torch._C._nn.linear( - x_263, - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_, - l_self_modules_head_modules_fc_modules_fc2_parameters_bias_, - ) - x_263 = ( - l_self_modules_head_modules_fc_modules_fc2_parameters_weight_ - ) = l_self_modules_head_modules_fc_modules_fc2_parameters_bias_ = None - return (x_264,) diff --git a/samples/timm/caformer_s18.sail_in22k/weight_meta.py b/samples/timm/caformer_s18.sail_in22k/weight_meta.py deleted file mode 100644 index c663bbb64a..0000000000 --- a/samples/timm/caformer_s18.sail_in22k/weight_meta.py +++ /dev/null @@ -1,2078 +0,0 @@ -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_weight_: - name = "L_self_modules_stem_modules_conv_parameters_weight_" - shape = [64, 3, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_conv_parameters_bias_: - name = "L_self_modules_stem_modules_conv_parameters_bias_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_s1: - name = "s1" - shape = [] - dtype = "torch.int64" - device = "cpu" - mean = None - std = None - data = [4] - - -class Program_weight_tensor_meta_L_x_: - name = "L_x_" - shape = [1, 3, 224, 224] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.226 - std = 1.288 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stem_modules_norm_parameters_weight_: - name = "L_self_modules_stem_modules_norm_parameters_weight_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [128, 64, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [128, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [64, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [256, 64, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [64, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [128, 64, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [128, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [64, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [256, 64, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [64, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [128, 64, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [128, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [64, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [256, 64, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [64, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_norm_parameters_weight_" - shape = [64] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_weight_" - shape = [128, 64, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_downsample_modules_conv_parameters_bias_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [256, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [256, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [128, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [512, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [128, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [256, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [256, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [128, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [512, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [128, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv1_parameters_weight_" - shape = [256, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_act1_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_dwconv_parameters_weight_" - shape = [256, 1, 7, 7] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_token_mixer_modules_pwconv2_parameters_weight_" - shape = [128, 256, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [512, 128, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [128, 512, 1, 1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_norm_parameters_weight_" - shape = [128] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_weight_" - shape = [320, 128, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_downsample_modules_conv_parameters_bias_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale1_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [960, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_token_mixer_modules_proj_parameters_weight_" - shape = [320, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_res_scale2_parameters_scale_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [1280, 320] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [320, 1280] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_norm_parameters_weight_" - shape = [320] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_weight_" - shape = [512, 320, 3, 3] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_downsample_modules_conv_parameters_bias_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale1_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_qkv_parameters_weight_" - shape = [1536, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_token_mixer_modules_proj_parameters_weight_" - shape = [512, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_res_scale2_parameters_scale_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_scale_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_act_parameters_bias_" - shape = [1] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = nan - data = None - - -class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [512, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_norm_parameters_weight_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_norm_parameters_bias_" - shape = [512] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_weight_" - shape = [2048, 512] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc1_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc1_parameters_bias_" - shape = [2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_weight_" - shape = [2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_norm_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_norm_parameters_bias_" - shape = [2048] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_weight_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_weight_" - shape = [21841, 2048] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_modules_fc_modules_fc2_parameters_bias_: - name = "L_self_modules_head_modules_fc_modules_fc2_parameters_bias_" - shape = [21841] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None diff --git a/timm_success_list.txt b/timm_success_list.txt deleted file mode 100644 index e69de29bb2..0000000000 From b173d9efda8688ee9c0326a8d536bbd47e9f29ff Mon Sep 17 00:00:00 2001 From: Liyulingyue <852433440@qq.com> Date: Mon, 4 Aug 2025 23:03:04 +0800 Subject: [PATCH 4/7] timm group --- .../cait_s24_224.fb_dist_in1k/graph_hash.txt | 1 + .../cait_s24_224.fb_dist_in1k/graph_net.json | 5 + .../cait_s24_224.fb_dist_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/cait_s24_224.fb_dist_in1k/model.py | 4812 +++++++++++ .../cait_s24_224.fb_dist_in1k/weight_meta.py | 4980 ++++++++++++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../cait_xxs24_224.fb_dist_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/cait_xxs24_224.fb_dist_in1k/model.py | 4812 +++++++++++ .../weight_meta.py | 4980 ++++++++++++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../cait_xxs36_224.fb_dist_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/cait_xxs36_224.fb_dist_in1k/model.py | 6972 ++++++++++++++++ .../weight_meta.py | 7236 +++++++++++++++++ .../timm/coat_lite_medium.in1k/graph_hash.txt | 1 + .../timm/coat_lite_medium.in1k/graph_net.json | 5 + .../timm/coat_lite_medium.in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/coat_lite_medium.in1k/model.py | 5903 ++++++++++++++ .../timm/coat_lite_medium.in1k/weight_meta.py | 3808 +++++++++ .../timm/coat_lite_mini.in1k/graph_hash.txt | 1 + .../timm/coat_lite_mini.in1k/graph_net.json | 5 + .../timm/coat_lite_mini.in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/coat_lite_mini.in1k/model.py | 1989 +++++ .../timm/coat_lite_mini.in1k/weight_meta.py | 1528 ++++ .../timm/coat_lite_small.in1k/graph_hash.txt | 1 + .../timm/coat_lite_small.in1k/graph_net.json | 5 + .../timm/coat_lite_small.in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/coat_lite_small.in1k/model.py | 3637 +++++++++ .../timm/coat_lite_small.in1k/weight_meta.py | 2488 ++++++ .../timm/coat_lite_tiny.in1k/graph_hash.txt | 1 + .../timm/coat_lite_tiny.in1k/graph_net.json | 5 + .../timm/coat_lite_tiny.in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/coat_lite_tiny.in1k/model.py | 1989 +++++ .../timm/coat_lite_tiny.in1k/weight_meta.py | 1528 ++++ 42 files changed, 56704 insertions(+) create mode 100644 samples/timm/cait_s24_224.fb_dist_in1k/graph_hash.txt create mode 100644 samples/timm/cait_s24_224.fb_dist_in1k/graph_net.json create mode 100644 samples/timm/cait_s24_224.fb_dist_in1k/input_meta.py create mode 100644 samples/timm/cait_s24_224.fb_dist_in1k/input_tensor_constraints.py create mode 100644 samples/timm/cait_s24_224.fb_dist_in1k/model.py create mode 100644 samples/timm/cait_s24_224.fb_dist_in1k/weight_meta.py create mode 100644 samples/timm/cait_xxs24_224.fb_dist_in1k/graph_hash.txt create mode 100644 samples/timm/cait_xxs24_224.fb_dist_in1k/graph_net.json create mode 100644 samples/timm/cait_xxs24_224.fb_dist_in1k/input_meta.py create mode 100644 samples/timm/cait_xxs24_224.fb_dist_in1k/input_tensor_constraints.py create mode 100644 samples/timm/cait_xxs24_224.fb_dist_in1k/model.py create mode 100644 samples/timm/cait_xxs24_224.fb_dist_in1k/weight_meta.py create mode 100644 samples/timm/cait_xxs36_224.fb_dist_in1k/graph_hash.txt create mode 100644 samples/timm/cait_xxs36_224.fb_dist_in1k/graph_net.json create mode 100644 samples/timm/cait_xxs36_224.fb_dist_in1k/input_meta.py create mode 100644 samples/timm/cait_xxs36_224.fb_dist_in1k/input_tensor_constraints.py create mode 100644 samples/timm/cait_xxs36_224.fb_dist_in1k/model.py create mode 100644 samples/timm/cait_xxs36_224.fb_dist_in1k/weight_meta.py create mode 100644 samples/timm/coat_lite_medium.in1k/graph_hash.txt create mode 100644 samples/timm/coat_lite_medium.in1k/graph_net.json create mode 100644 samples/timm/coat_lite_medium.in1k/input_meta.py create mode 100644 samples/timm/coat_lite_medium.in1k/input_tensor_constraints.py create mode 100644 samples/timm/coat_lite_medium.in1k/model.py create mode 100644 samples/timm/coat_lite_medium.in1k/weight_meta.py create mode 100644 samples/timm/coat_lite_mini.in1k/graph_hash.txt create mode 100644 samples/timm/coat_lite_mini.in1k/graph_net.json create mode 100644 samples/timm/coat_lite_mini.in1k/input_meta.py create mode 100644 samples/timm/coat_lite_mini.in1k/input_tensor_constraints.py create mode 100644 samples/timm/coat_lite_mini.in1k/model.py create mode 100644 samples/timm/coat_lite_mini.in1k/weight_meta.py create mode 100644 samples/timm/coat_lite_small.in1k/graph_hash.txt create mode 100644 samples/timm/coat_lite_small.in1k/graph_net.json create mode 100644 samples/timm/coat_lite_small.in1k/input_meta.py create mode 100644 samples/timm/coat_lite_small.in1k/input_tensor_constraints.py create mode 100644 samples/timm/coat_lite_small.in1k/model.py create mode 100644 samples/timm/coat_lite_small.in1k/weight_meta.py create mode 100644 samples/timm/coat_lite_tiny.in1k/graph_hash.txt create mode 100644 samples/timm/coat_lite_tiny.in1k/graph_net.json create mode 100644 samples/timm/coat_lite_tiny.in1k/input_meta.py create mode 100644 samples/timm/coat_lite_tiny.in1k/input_tensor_constraints.py create mode 100644 samples/timm/coat_lite_tiny.in1k/model.py create mode 100644 samples/timm/coat_lite_tiny.in1k/weight_meta.py diff --git a/samples/timm/cait_s24_224.fb_dist_in1k/graph_hash.txt b/samples/timm/cait_s24_224.fb_dist_in1k/graph_hash.txt new file mode 100644 index 0000000000..2bea990948 --- /dev/null +++ b/samples/timm/cait_s24_224.fb_dist_in1k/graph_hash.txt @@ -0,0 +1 @@ +1e2932070cfe79b2301a5dcbf8822c085d0b27f1952820dec3236ec25dee50e4 \ No newline at end of file diff --git a/samples/timm/cait_s24_224.fb_dist_in1k/graph_net.json b/samples/timm/cait_s24_224.fb_dist_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/cait_s24_224.fb_dist_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cait_s24_224.fb_dist_in1k/input_meta.py b/samples/timm/cait_s24_224.fb_dist_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cait_s24_224.fb_dist_in1k/input_tensor_constraints.py b/samples/timm/cait_s24_224.fb_dist_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cait_s24_224.fb_dist_in1k/model.py b/samples/timm/cait_s24_224.fb_dist_in1k/model.py new file mode 100644 index 0000000000..99089efdd9 --- /dev/null +++ b/samples/timm/cait_s24_224.fb_dist_in1k/model.py @@ -0,0 +1,4812 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_modules_blocks_modules_0_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_0_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_0_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_1_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_1_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_2_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_2_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_3_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_3_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_4_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_4_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_5_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_5_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_6_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_6_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_7_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_7_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_8_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_8_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_9_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_9_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_10_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_10_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_11_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_11_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_12_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_12_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_12_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_12_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_13_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_13_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_13_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_13_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_14_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_14_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_14_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_14_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_15_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_15_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_15_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_15_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_16_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_16_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_16_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_16_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_17_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_17_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_17_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_17_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_18_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_18_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_18_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_18_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_19_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_19_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_19_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_19_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_20_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_20_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_20_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_20_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_21_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_21_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_21_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_21_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_22_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_22_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_22_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_22_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_23_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_23_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_23_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_23_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_modules_blocks_token_only_modules_0_parameters_gamma_1_ = ( + L_self_modules_blocks_token_only_modules_0_parameters_gamma_1_ + ) + l_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_weight_ = L_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_weight_ + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_bias_ = L_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_bias_ + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_weight_ = L_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_weight_ + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_bias_ = L_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_bias_ + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_weight_ = L_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_weight_ + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_bias_ = L_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_bias_ + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_token_only_modules_0_parameters_gamma_2_ = ( + L_self_modules_blocks_token_only_modules_0_parameters_gamma_2_ + ) + l_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_token_only_modules_1_parameters_gamma_1_ = ( + L_self_modules_blocks_token_only_modules_1_parameters_gamma_1_ + ) + l_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_weight_ = L_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_weight_ + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_bias_ = L_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_bias_ + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_weight_ = L_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_weight_ + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_bias_ = L_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_bias_ + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_weight_ = L_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_weight_ + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_bias_ = L_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_bias_ + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_token_only_modules_1_parameters_gamma_2_ = ( + L_self_modules_blocks_token_only_modules_1_parameters_gamma_2_ + ) + l_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_norm_parameters_weight_ = L_self_modules_norm_parameters_weight_ + l_self_modules_norm_parameters_bias_ = L_self_modules_norm_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + x_2 = x_1 + l_self_parameters_pos_embed_ + x_1 = l_self_parameters_pos_embed_ = None + x_3 = torch.nn.functional.dropout(x_2, 0.0, False, False) + x_2 = None + layer_norm = torch.nn.functional.layer_norm( + x_3, + (384,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + layer_norm, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 196, 3, 8, 48) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + getitem_4 = qkv[0] + q = getitem_4 * 0.14433756729740643 + getitem_4 = None + k = qkv[1] + v = qkv[2] + qkv = None + transpose_1 = k.transpose(-2, -1) + k = None + attn = q @ transpose_1 + q = transpose_1 = None + permute_1 = attn.permute(0, 2, 3, 1) + attn = None + linear_1 = torch._C._nn.linear( + permute_1, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_1 = l_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_1 = linear_1.permute(0, 3, 1, 2) + linear_1 = None + attn_2 = attn_1.softmax(dim=-1) + attn_1 = None + permute_3 = attn_2.permute(0, 2, 3, 1) + attn_2 = None + linear_2 = torch._C._nn.linear( + permute_3, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_3 = l_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_3 = linear_2.permute(0, 3, 1, 2) + linear_2 = None + attn_4 = torch.nn.functional.dropout(attn_3, 0.0, False, False) + attn_3 = None + matmul_1 = attn_4 @ v + attn_4 = v = None + transpose_2 = matmul_1.transpose(1, 2) + matmul_1 = None + x_4 = transpose_2.reshape(1, 196, 384) + transpose_2 = None + x_5 = torch._C._nn.linear( + x_4, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_4 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_6 = torch.nn.functional.dropout(x_5, 0.0, False, False) + x_5 = None + mul_1 = l_self_modules_blocks_modules_0_parameters_gamma_1_ * x_6 + l_self_modules_blocks_modules_0_parameters_gamma_1_ = x_6 = None + x_7 = x_3 + mul_1 + x_3 = mul_1 = None + layer_norm_1 = torch.nn.functional.layer_norm( + x_7, + (384,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_8 = torch._C._nn.linear( + layer_norm_1, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_1 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_9 = torch._C._nn.gelu(x_8, approximate="none") + x_8 = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = torch._C._nn.linear( + x_10, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_10 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + mul_2 = l_self_modules_blocks_modules_0_parameters_gamma_2_ * x_12 + l_self_modules_blocks_modules_0_parameters_gamma_2_ = x_12 = None + x_13 = x_7 + mul_2 + x_7 = mul_2 = None + layer_norm_2 = torch.nn.functional.layer_norm( + x_13, + (384,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_6 = torch._C._nn.linear( + layer_norm_2, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_2 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_6.reshape(1, 196, 3, 8, 48) + linear_6 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + getitem_7 = qkv_1[0] + q_1 = getitem_7 * 0.14433756729740643 + getitem_7 = None + k_1 = qkv_1[1] + v_1 = qkv_1[2] + qkv_1 = None + transpose_3 = k_1.transpose(-2, -1) + k_1 = None + attn_5 = q_1 @ transpose_3 + q_1 = transpose_3 = None + permute_6 = attn_5.permute(0, 2, 3, 1) + attn_5 = None + linear_7 = torch._C._nn.linear( + permute_6, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_6 = l_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_6 = linear_7.permute(0, 3, 1, 2) + linear_7 = None + attn_7 = attn_6.softmax(dim=-1) + attn_6 = None + permute_8 = attn_7.permute(0, 2, 3, 1) + attn_7 = None + linear_8 = torch._C._nn.linear( + permute_8, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_8 = l_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_8 = linear_8.permute(0, 3, 1, 2) + linear_8 = None + attn_9 = torch.nn.functional.dropout(attn_8, 0.0, False, False) + attn_8 = None + matmul_3 = attn_9 @ v_1 + attn_9 = v_1 = None + transpose_4 = matmul_3.transpose(1, 2) + matmul_3 = None + x_14 = transpose_4.reshape(1, 196, 384) + transpose_4 = None + x_15 = torch._C._nn.linear( + x_14, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_14 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + mul_4 = l_self_modules_blocks_modules_1_parameters_gamma_1_ * x_16 + l_self_modules_blocks_modules_1_parameters_gamma_1_ = x_16 = None + x_17 = x_13 + mul_4 + x_13 = mul_4 = None + layer_norm_3 = torch.nn.functional.layer_norm( + x_17, + (384,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_18 = torch._C._nn.linear( + layer_norm_3, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_3 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_19 = torch._C._nn.gelu(x_18, approximate="none") + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + mul_5 = l_self_modules_blocks_modules_1_parameters_gamma_2_ * x_22 + l_self_modules_blocks_modules_1_parameters_gamma_2_ = x_22 = None + x_23 = x_17 + mul_5 + x_17 = mul_5 = None + layer_norm_4 = torch.nn.functional.layer_norm( + x_23, + (384,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + layer_norm_4, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_4 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_12.reshape(1, 196, 3, 8, 48) + linear_12 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + getitem_10 = qkv_2[0] + q_2 = getitem_10 * 0.14433756729740643 + getitem_10 = None + k_2 = qkv_2[1] + v_2 = qkv_2[2] + qkv_2 = None + transpose_5 = k_2.transpose(-2, -1) + k_2 = None + attn_10 = q_2 @ transpose_5 + q_2 = transpose_5 = None + permute_11 = attn_10.permute(0, 2, 3, 1) + attn_10 = None + linear_13 = torch._C._nn.linear( + permute_11, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_11 = l_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_11 = linear_13.permute(0, 3, 1, 2) + linear_13 = None + attn_12 = attn_11.softmax(dim=-1) + attn_11 = None + permute_13 = attn_12.permute(0, 2, 3, 1) + attn_12 = None + linear_14 = torch._C._nn.linear( + permute_13, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_13 = l_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_13 = linear_14.permute(0, 3, 1, 2) + linear_14 = None + attn_14 = torch.nn.functional.dropout(attn_13, 0.0, False, False) + attn_13 = None + matmul_5 = attn_14 @ v_2 + attn_14 = v_2 = None + transpose_6 = matmul_5.transpose(1, 2) + matmul_5 = None + x_24 = transpose_6.reshape(1, 196, 384) + transpose_6 = None + x_25 = torch._C._nn.linear( + x_24, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_24 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_26 = torch.nn.functional.dropout(x_25, 0.0, False, False) + x_25 = None + mul_7 = l_self_modules_blocks_modules_2_parameters_gamma_1_ * x_26 + l_self_modules_blocks_modules_2_parameters_gamma_1_ = x_26 = None + x_27 = x_23 + mul_7 + x_23 = mul_7 = None + layer_norm_5 = torch.nn.functional.layer_norm( + x_27, + (384,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_28 = torch._C._nn.linear( + layer_norm_5, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_5 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_29 = torch._C._nn.gelu(x_28, approximate="none") + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_31 = torch._C._nn.linear( + x_30, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_30 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + mul_8 = l_self_modules_blocks_modules_2_parameters_gamma_2_ * x_32 + l_self_modules_blocks_modules_2_parameters_gamma_2_ = x_32 = None + x_33 = x_27 + mul_8 + x_27 = mul_8 = None + layer_norm_6 = torch.nn.functional.layer_norm( + x_33, + (384,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_18 = torch._C._nn.linear( + layer_norm_6, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_6 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_18.reshape(1, 196, 3, 8, 48) + linear_18 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + getitem_13 = qkv_3[0] + q_3 = getitem_13 * 0.14433756729740643 + getitem_13 = None + k_3 = qkv_3[1] + v_3 = qkv_3[2] + qkv_3 = None + transpose_7 = k_3.transpose(-2, -1) + k_3 = None + attn_15 = q_3 @ transpose_7 + q_3 = transpose_7 = None + permute_16 = attn_15.permute(0, 2, 3, 1) + attn_15 = None + linear_19 = torch._C._nn.linear( + permute_16, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_16 = l_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_16 = linear_19.permute(0, 3, 1, 2) + linear_19 = None + attn_17 = attn_16.softmax(dim=-1) + attn_16 = None + permute_18 = attn_17.permute(0, 2, 3, 1) + attn_17 = None + linear_20 = torch._C._nn.linear( + permute_18, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_18 = l_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_18 = linear_20.permute(0, 3, 1, 2) + linear_20 = None + attn_19 = torch.nn.functional.dropout(attn_18, 0.0, False, False) + attn_18 = None + matmul_7 = attn_19 @ v_3 + attn_19 = v_3 = None + transpose_8 = matmul_7.transpose(1, 2) + matmul_7 = None + x_34 = transpose_8.reshape(1, 196, 384) + transpose_8 = None + x_35 = torch._C._nn.linear( + x_34, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_34 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + mul_10 = l_self_modules_blocks_modules_3_parameters_gamma_1_ * x_36 + l_self_modules_blocks_modules_3_parameters_gamma_1_ = x_36 = None + x_37 = x_33 + mul_10 + x_33 = mul_10 = None + layer_norm_7 = torch.nn.functional.layer_norm( + x_37, + (384,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_38 = torch._C._nn.linear( + layer_norm_7, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_7 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_39 = torch._C._nn.gelu(x_38, approximate="none") + x_38 = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = torch._C._nn.linear( + x_40, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_40 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_42 = torch.nn.functional.dropout(x_41, 0.0, False, False) + x_41 = None + mul_11 = l_self_modules_blocks_modules_3_parameters_gamma_2_ * x_42 + l_self_modules_blocks_modules_3_parameters_gamma_2_ = x_42 = None + x_43 = x_37 + mul_11 + x_37 = mul_11 = None + layer_norm_8 = torch.nn.functional.layer_norm( + x_43, + (384,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + layer_norm_8, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_8 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_24.reshape(1, 196, 3, 8, 48) + linear_24 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + getitem_16 = qkv_4[0] + q_4 = getitem_16 * 0.14433756729740643 + getitem_16 = None + k_4 = qkv_4[1] + v_4 = qkv_4[2] + qkv_4 = None + transpose_9 = k_4.transpose(-2, -1) + k_4 = None + attn_20 = q_4 @ transpose_9 + q_4 = transpose_9 = None + permute_21 = attn_20.permute(0, 2, 3, 1) + attn_20 = None + linear_25 = torch._C._nn.linear( + permute_21, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_21 = l_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_21 = linear_25.permute(0, 3, 1, 2) + linear_25 = None + attn_22 = attn_21.softmax(dim=-1) + attn_21 = None + permute_23 = attn_22.permute(0, 2, 3, 1) + attn_22 = None + linear_26 = torch._C._nn.linear( + permute_23, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_23 = l_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_23 = linear_26.permute(0, 3, 1, 2) + linear_26 = None + attn_24 = torch.nn.functional.dropout(attn_23, 0.0, False, False) + attn_23 = None + matmul_9 = attn_24 @ v_4 + attn_24 = v_4 = None + transpose_10 = matmul_9.transpose(1, 2) + matmul_9 = None + x_44 = transpose_10.reshape(1, 196, 384) + transpose_10 = None + x_45 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_44 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_46 = torch.nn.functional.dropout(x_45, 0.0, False, False) + x_45 = None + mul_13 = l_self_modules_blocks_modules_4_parameters_gamma_1_ * x_46 + l_self_modules_blocks_modules_4_parameters_gamma_1_ = x_46 = None + x_47 = x_43 + mul_13 + x_43 = mul_13 = None + layer_norm_9 = torch.nn.functional.layer_norm( + x_47, + (384,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_48 = torch._C._nn.linear( + layer_norm_9, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_9 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_49 = torch._C._nn.gelu(x_48, approximate="none") + x_48 = None + x_50 = torch.nn.functional.dropout(x_49, 0.0, False, False) + x_49 = None + x_51 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_52 = torch.nn.functional.dropout(x_51, 0.0, False, False) + x_51 = None + mul_14 = l_self_modules_blocks_modules_4_parameters_gamma_2_ * x_52 + l_self_modules_blocks_modules_4_parameters_gamma_2_ = x_52 = None + x_53 = x_47 + mul_14 + x_47 = mul_14 = None + layer_norm_10 = torch.nn.functional.layer_norm( + x_53, + (384,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_30 = torch._C._nn.linear( + layer_norm_10, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_10 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_30.reshape(1, 196, 3, 8, 48) + linear_30 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + getitem_19 = qkv_5[0] + q_5 = getitem_19 * 0.14433756729740643 + getitem_19 = None + k_5 = qkv_5[1] + v_5 = qkv_5[2] + qkv_5 = None + transpose_11 = k_5.transpose(-2, -1) + k_5 = None + attn_25 = q_5 @ transpose_11 + q_5 = transpose_11 = None + permute_26 = attn_25.permute(0, 2, 3, 1) + attn_25 = None + linear_31 = torch._C._nn.linear( + permute_26, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_26 = l_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_26 = linear_31.permute(0, 3, 1, 2) + linear_31 = None + attn_27 = attn_26.softmax(dim=-1) + attn_26 = None + permute_28 = attn_27.permute(0, 2, 3, 1) + attn_27 = None + linear_32 = torch._C._nn.linear( + permute_28, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_28 = l_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_28 = linear_32.permute(0, 3, 1, 2) + linear_32 = None + attn_29 = torch.nn.functional.dropout(attn_28, 0.0, False, False) + attn_28 = None + matmul_11 = attn_29 @ v_5 + attn_29 = v_5 = None + transpose_12 = matmul_11.transpose(1, 2) + matmul_11 = None + x_54 = transpose_12.reshape(1, 196, 384) + transpose_12 = None + x_55 = torch._C._nn.linear( + x_54, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_54 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_56 = torch.nn.functional.dropout(x_55, 0.0, False, False) + x_55 = None + mul_16 = l_self_modules_blocks_modules_5_parameters_gamma_1_ * x_56 + l_self_modules_blocks_modules_5_parameters_gamma_1_ = x_56 = None + x_57 = x_53 + mul_16 + x_53 = mul_16 = None + layer_norm_11 = torch.nn.functional.layer_norm( + x_57, + (384,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_58 = torch._C._nn.linear( + layer_norm_11, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_11 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_59 = torch._C._nn.gelu(x_58, approximate="none") + x_58 = None + x_60 = torch.nn.functional.dropout(x_59, 0.0, False, False) + x_59 = None + x_61 = torch._C._nn.linear( + x_60, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_60 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + mul_17 = l_self_modules_blocks_modules_5_parameters_gamma_2_ * x_62 + l_self_modules_blocks_modules_5_parameters_gamma_2_ = x_62 = None + x_63 = x_57 + mul_17 + x_57 = mul_17 = None + layer_norm_12 = torch.nn.functional.layer_norm( + x_63, + (384,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_12 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_36.reshape(1, 196, 3, 8, 48) + linear_36 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + getitem_22 = qkv_6[0] + q_6 = getitem_22 * 0.14433756729740643 + getitem_22 = None + k_6 = qkv_6[1] + v_6 = qkv_6[2] + qkv_6 = None + transpose_13 = k_6.transpose(-2, -1) + k_6 = None + attn_30 = q_6 @ transpose_13 + q_6 = transpose_13 = None + permute_31 = attn_30.permute(0, 2, 3, 1) + attn_30 = None + linear_37 = torch._C._nn.linear( + permute_31, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_31 = l_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_31 = linear_37.permute(0, 3, 1, 2) + linear_37 = None + attn_32 = attn_31.softmax(dim=-1) + attn_31 = None + permute_33 = attn_32.permute(0, 2, 3, 1) + attn_32 = None + linear_38 = torch._C._nn.linear( + permute_33, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_33 = l_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_33 = linear_38.permute(0, 3, 1, 2) + linear_38 = None + attn_34 = torch.nn.functional.dropout(attn_33, 0.0, False, False) + attn_33 = None + matmul_13 = attn_34 @ v_6 + attn_34 = v_6 = None + transpose_14 = matmul_13.transpose(1, 2) + matmul_13 = None + x_64 = transpose_14.reshape(1, 196, 384) + transpose_14 = None + x_65 = torch._C._nn.linear( + x_64, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_64 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + mul_19 = l_self_modules_blocks_modules_6_parameters_gamma_1_ * x_66 + l_self_modules_blocks_modules_6_parameters_gamma_1_ = x_66 = None + x_67 = x_63 + mul_19 + x_63 = mul_19 = None + layer_norm_13 = torch.nn.functional.layer_norm( + x_67, + (384,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_68 = torch._C._nn.linear( + layer_norm_13, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_13 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_69 = torch._C._nn.gelu(x_68, approximate="none") + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_71 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_70 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + mul_20 = l_self_modules_blocks_modules_6_parameters_gamma_2_ * x_72 + l_self_modules_blocks_modules_6_parameters_gamma_2_ = x_72 = None + x_73 = x_67 + mul_20 + x_67 = mul_20 = None + layer_norm_14 = torch.nn.functional.layer_norm( + x_73, + (384,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_42 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_14 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_42.reshape(1, 196, 3, 8, 48) + linear_42 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + getitem_25 = qkv_7[0] + q_7 = getitem_25 * 0.14433756729740643 + getitem_25 = None + k_7 = qkv_7[1] + v_7 = qkv_7[2] + qkv_7 = None + transpose_15 = k_7.transpose(-2, -1) + k_7 = None + attn_35 = q_7 @ transpose_15 + q_7 = transpose_15 = None + permute_36 = attn_35.permute(0, 2, 3, 1) + attn_35 = None + linear_43 = torch._C._nn.linear( + permute_36, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_36 = l_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_36 = linear_43.permute(0, 3, 1, 2) + linear_43 = None + attn_37 = attn_36.softmax(dim=-1) + attn_36 = None + permute_38 = attn_37.permute(0, 2, 3, 1) + attn_37 = None + linear_44 = torch._C._nn.linear( + permute_38, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_38 = l_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_38 = linear_44.permute(0, 3, 1, 2) + linear_44 = None + attn_39 = torch.nn.functional.dropout(attn_38, 0.0, False, False) + attn_38 = None + matmul_15 = attn_39 @ v_7 + attn_39 = v_7 = None + transpose_16 = matmul_15.transpose(1, 2) + matmul_15 = None + x_74 = transpose_16.reshape(1, 196, 384) + transpose_16 = None + x_75 = torch._C._nn.linear( + x_74, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_74 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_76 = torch.nn.functional.dropout(x_75, 0.0, False, False) + x_75 = None + mul_22 = l_self_modules_blocks_modules_7_parameters_gamma_1_ * x_76 + l_self_modules_blocks_modules_7_parameters_gamma_1_ = x_76 = None + x_77 = x_73 + mul_22 + x_73 = mul_22 = None + layer_norm_15 = torch.nn.functional.layer_norm( + x_77, + (384,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_78 = torch._C._nn.linear( + layer_norm_15, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_15 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_79 = torch._C._nn.gelu(x_78, approximate="none") + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = torch._C._nn.linear( + x_80, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_80 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_82 = torch.nn.functional.dropout(x_81, 0.0, False, False) + x_81 = None + mul_23 = l_self_modules_blocks_modules_7_parameters_gamma_2_ * x_82 + l_self_modules_blocks_modules_7_parameters_gamma_2_ = x_82 = None + x_83 = x_77 + mul_23 + x_77 = mul_23 = None + layer_norm_16 = torch.nn.functional.layer_norm( + x_83, + (384,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_48 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_16 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_48.reshape(1, 196, 3, 8, 48) + linear_48 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + getitem_28 = qkv_8[0] + q_8 = getitem_28 * 0.14433756729740643 + getitem_28 = None + k_8 = qkv_8[1] + v_8 = qkv_8[2] + qkv_8 = None + transpose_17 = k_8.transpose(-2, -1) + k_8 = None + attn_40 = q_8 @ transpose_17 + q_8 = transpose_17 = None + permute_41 = attn_40.permute(0, 2, 3, 1) + attn_40 = None + linear_49 = torch._C._nn.linear( + permute_41, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_41 = l_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_41 = linear_49.permute(0, 3, 1, 2) + linear_49 = None + attn_42 = attn_41.softmax(dim=-1) + attn_41 = None + permute_43 = attn_42.permute(0, 2, 3, 1) + attn_42 = None + linear_50 = torch._C._nn.linear( + permute_43, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_43 = l_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_43 = linear_50.permute(0, 3, 1, 2) + linear_50 = None + attn_44 = torch.nn.functional.dropout(attn_43, 0.0, False, False) + attn_43 = None + matmul_17 = attn_44 @ v_8 + attn_44 = v_8 = None + transpose_18 = matmul_17.transpose(1, 2) + matmul_17 = None + x_84 = transpose_18.reshape(1, 196, 384) + transpose_18 = None + x_85 = torch._C._nn.linear( + x_84, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_84 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_86 = torch.nn.functional.dropout(x_85, 0.0, False, False) + x_85 = None + mul_25 = l_self_modules_blocks_modules_8_parameters_gamma_1_ * x_86 + l_self_modules_blocks_modules_8_parameters_gamma_1_ = x_86 = None + x_87 = x_83 + mul_25 + x_83 = mul_25 = None + layer_norm_17 = torch.nn.functional.layer_norm( + x_87, + (384,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_88 = torch._C._nn.linear( + layer_norm_17, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_17 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_89 = torch._C._nn.gelu(x_88, approximate="none") + x_88 = None + x_90 = torch.nn.functional.dropout(x_89, 0.0, False, False) + x_89 = None + x_91 = torch._C._nn.linear( + x_90, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_90 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + mul_26 = l_self_modules_blocks_modules_8_parameters_gamma_2_ * x_92 + l_self_modules_blocks_modules_8_parameters_gamma_2_ = x_92 = None + x_93 = x_87 + mul_26 + x_87 = mul_26 = None + layer_norm_18 = torch.nn.functional.layer_norm( + x_93, + (384,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_54 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_18 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_54.reshape(1, 196, 3, 8, 48) + linear_54 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + getitem_31 = qkv_9[0] + q_9 = getitem_31 * 0.14433756729740643 + getitem_31 = None + k_9 = qkv_9[1] + v_9 = qkv_9[2] + qkv_9 = None + transpose_19 = k_9.transpose(-2, -1) + k_9 = None + attn_45 = q_9 @ transpose_19 + q_9 = transpose_19 = None + permute_46 = attn_45.permute(0, 2, 3, 1) + attn_45 = None + linear_55 = torch._C._nn.linear( + permute_46, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_46 = l_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_46 = linear_55.permute(0, 3, 1, 2) + linear_55 = None + attn_47 = attn_46.softmax(dim=-1) + attn_46 = None + permute_48 = attn_47.permute(0, 2, 3, 1) + attn_47 = None + linear_56 = torch._C._nn.linear( + permute_48, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_48 = l_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_48 = linear_56.permute(0, 3, 1, 2) + linear_56 = None + attn_49 = torch.nn.functional.dropout(attn_48, 0.0, False, False) + attn_48 = None + matmul_19 = attn_49 @ v_9 + attn_49 = v_9 = None + transpose_20 = matmul_19.transpose(1, 2) + matmul_19 = None + x_94 = transpose_20.reshape(1, 196, 384) + transpose_20 = None + x_95 = torch._C._nn.linear( + x_94, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_94 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_96 = torch.nn.functional.dropout(x_95, 0.0, False, False) + x_95 = None + mul_28 = l_self_modules_blocks_modules_9_parameters_gamma_1_ * x_96 + l_self_modules_blocks_modules_9_parameters_gamma_1_ = x_96 = None + x_97 = x_93 + mul_28 + x_93 = mul_28 = None + layer_norm_19 = torch.nn.functional.layer_norm( + x_97, + (384,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_98 = torch._C._nn.linear( + layer_norm_19, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_19 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_99 = torch._C._nn.gelu(x_98, approximate="none") + x_98 = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + x_101 = torch._C._nn.linear( + x_100, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_100 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + mul_29 = l_self_modules_blocks_modules_9_parameters_gamma_2_ * x_102 + l_self_modules_blocks_modules_9_parameters_gamma_2_ = x_102 = None + x_103 = x_97 + mul_29 + x_97 = mul_29 = None + layer_norm_20 = torch.nn.functional.layer_norm( + x_103, + (384,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_60 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_20 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_60.reshape(1, 196, 3, 8, 48) + linear_60 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + getitem_34 = qkv_10[0] + q_10 = getitem_34 * 0.14433756729740643 + getitem_34 = None + k_10 = qkv_10[1] + v_10 = qkv_10[2] + qkv_10 = None + transpose_21 = k_10.transpose(-2, -1) + k_10 = None + attn_50 = q_10 @ transpose_21 + q_10 = transpose_21 = None + permute_51 = attn_50.permute(0, 2, 3, 1) + attn_50 = None + linear_61 = torch._C._nn.linear( + permute_51, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_51 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_51 = linear_61.permute(0, 3, 1, 2) + linear_61 = None + attn_52 = attn_51.softmax(dim=-1) + attn_51 = None + permute_53 = attn_52.permute(0, 2, 3, 1) + attn_52 = None + linear_62 = torch._C._nn.linear( + permute_53, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_53 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_53 = linear_62.permute(0, 3, 1, 2) + linear_62 = None + attn_54 = torch.nn.functional.dropout(attn_53, 0.0, False, False) + attn_53 = None + matmul_21 = attn_54 @ v_10 + attn_54 = v_10 = None + transpose_22 = matmul_21.transpose(1, 2) + matmul_21 = None + x_104 = transpose_22.reshape(1, 196, 384) + transpose_22 = None + x_105 = torch._C._nn.linear( + x_104, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_104 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + mul_31 = l_self_modules_blocks_modules_10_parameters_gamma_1_ * x_106 + l_self_modules_blocks_modules_10_parameters_gamma_1_ = x_106 = None + x_107 = x_103 + mul_31 + x_103 = mul_31 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_107, + (384,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_108 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_21 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_109 = torch._C._nn.gelu(x_108, approximate="none") + x_108 = None + x_110 = torch.nn.functional.dropout(x_109, 0.0, False, False) + x_109 = None + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_110 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_112 = torch.nn.functional.dropout(x_111, 0.0, False, False) + x_111 = None + mul_32 = l_self_modules_blocks_modules_10_parameters_gamma_2_ * x_112 + l_self_modules_blocks_modules_10_parameters_gamma_2_ = x_112 = None + x_113 = x_107 + mul_32 + x_107 = mul_32 = None + layer_norm_22 = torch.nn.functional.layer_norm( + x_113, + (384,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_66 = torch._C._nn.linear( + layer_norm_22, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_22 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_66.reshape(1, 196, 3, 8, 48) + linear_66 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + getitem_37 = qkv_11[0] + q_11 = getitem_37 * 0.14433756729740643 + getitem_37 = None + k_11 = qkv_11[1] + v_11 = qkv_11[2] + qkv_11 = None + transpose_23 = k_11.transpose(-2, -1) + k_11 = None + attn_55 = q_11 @ transpose_23 + q_11 = transpose_23 = None + permute_56 = attn_55.permute(0, 2, 3, 1) + attn_55 = None + linear_67 = torch._C._nn.linear( + permute_56, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_56 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_56 = linear_67.permute(0, 3, 1, 2) + linear_67 = None + attn_57 = attn_56.softmax(dim=-1) + attn_56 = None + permute_58 = attn_57.permute(0, 2, 3, 1) + attn_57 = None + linear_68 = torch._C._nn.linear( + permute_58, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_58 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_58 = linear_68.permute(0, 3, 1, 2) + linear_68 = None + attn_59 = torch.nn.functional.dropout(attn_58, 0.0, False, False) + attn_58 = None + matmul_23 = attn_59 @ v_11 + attn_59 = v_11 = None + transpose_24 = matmul_23.transpose(1, 2) + matmul_23 = None + x_114 = transpose_24.reshape(1, 196, 384) + transpose_24 = None + x_115 = torch._C._nn.linear( + x_114, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_114 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + mul_34 = l_self_modules_blocks_modules_11_parameters_gamma_1_ * x_116 + l_self_modules_blocks_modules_11_parameters_gamma_1_ = x_116 = None + x_117 = x_113 + mul_34 + x_113 = mul_34 = None + layer_norm_23 = torch.nn.functional.layer_norm( + x_117, + (384,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_118 = torch._C._nn.linear( + layer_norm_23, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_23 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_119 = torch._C._nn.gelu(x_118, approximate="none") + x_118 = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + x_121 = torch._C._nn.linear( + x_120, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_120 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_122 = torch.nn.functional.dropout(x_121, 0.0, False, False) + x_121 = None + mul_35 = l_self_modules_blocks_modules_11_parameters_gamma_2_ * x_122 + l_self_modules_blocks_modules_11_parameters_gamma_2_ = x_122 = None + x_123 = x_117 + mul_35 + x_117 = mul_35 = None + layer_norm_24 = torch.nn.functional.layer_norm( + x_123, + (384,), + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ + ) = None + linear_72 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_24 = ( + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_24 = linear_72.reshape(1, 196, 3, 8, 48) + linear_72 = None + qkv_12 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + getitem_40 = qkv_12[0] + q_12 = getitem_40 * 0.14433756729740643 + getitem_40 = None + k_12 = qkv_12[1] + v_12 = qkv_12[2] + qkv_12 = None + transpose_25 = k_12.transpose(-2, -1) + k_12 = None + attn_60 = q_12 @ transpose_25 + q_12 = transpose_25 = None + permute_61 = attn_60.permute(0, 2, 3, 1) + attn_60 = None + linear_73 = torch._C._nn.linear( + permute_61, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_61 = l_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_61 = linear_73.permute(0, 3, 1, 2) + linear_73 = None + attn_62 = attn_61.softmax(dim=-1) + attn_61 = None + permute_63 = attn_62.permute(0, 2, 3, 1) + attn_62 = None + linear_74 = torch._C._nn.linear( + permute_63, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_63 = l_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_63 = linear_74.permute(0, 3, 1, 2) + linear_74 = None + attn_64 = torch.nn.functional.dropout(attn_63, 0.0, False, False) + attn_63 = None + matmul_25 = attn_64 @ v_12 + attn_64 = v_12 = None + transpose_26 = matmul_25.transpose(1, 2) + matmul_25 = None + x_124 = transpose_26.reshape(1, 196, 384) + transpose_26 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_, + ) + x_124 = l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ + ) = None + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + mul_37 = l_self_modules_blocks_modules_12_parameters_gamma_1_ * x_126 + l_self_modules_blocks_modules_12_parameters_gamma_1_ = x_126 = None + x_127 = x_123 + mul_37 + x_123 = mul_37 = None + layer_norm_25 = torch.nn.functional.layer_norm( + x_127, + (384,), + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ + ) = None + x_128 = torch._C._nn.linear( + layer_norm_25, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_25 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_129 = torch._C._nn.gelu(x_128, approximate="none") + x_128 = None + x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) + x_129 = None + x_131 = torch._C._nn.linear( + x_130, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_130 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_132 = torch.nn.functional.dropout(x_131, 0.0, False, False) + x_131 = None + mul_38 = l_self_modules_blocks_modules_12_parameters_gamma_2_ * x_132 + l_self_modules_blocks_modules_12_parameters_gamma_2_ = x_132 = None + x_133 = x_127 + mul_38 + x_127 = mul_38 = None + layer_norm_26 = torch.nn.functional.layer_norm( + x_133, + (384,), + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ + ) = None + linear_78 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_26 = ( + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_26 = linear_78.reshape(1, 196, 3, 8, 48) + linear_78 = None + qkv_13 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + getitem_43 = qkv_13[0] + q_13 = getitem_43 * 0.14433756729740643 + getitem_43 = None + k_13 = qkv_13[1] + v_13 = qkv_13[2] + qkv_13 = None + transpose_27 = k_13.transpose(-2, -1) + k_13 = None + attn_65 = q_13 @ transpose_27 + q_13 = transpose_27 = None + permute_66 = attn_65.permute(0, 2, 3, 1) + attn_65 = None + linear_79 = torch._C._nn.linear( + permute_66, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_66 = l_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_66 = linear_79.permute(0, 3, 1, 2) + linear_79 = None + attn_67 = attn_66.softmax(dim=-1) + attn_66 = None + permute_68 = attn_67.permute(0, 2, 3, 1) + attn_67 = None + linear_80 = torch._C._nn.linear( + permute_68, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_68 = l_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_68 = linear_80.permute(0, 3, 1, 2) + linear_80 = None + attn_69 = torch.nn.functional.dropout(attn_68, 0.0, False, False) + attn_68 = None + matmul_27 = attn_69 @ v_13 + attn_69 = v_13 = None + transpose_28 = matmul_27.transpose(1, 2) + matmul_27 = None + x_134 = transpose_28.reshape(1, 196, 384) + transpose_28 = None + x_135 = torch._C._nn.linear( + x_134, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_, + ) + x_134 = l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ + ) = None + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + mul_40 = l_self_modules_blocks_modules_13_parameters_gamma_1_ * x_136 + l_self_modules_blocks_modules_13_parameters_gamma_1_ = x_136 = None + x_137 = x_133 + mul_40 + x_133 = mul_40 = None + layer_norm_27 = torch.nn.functional.layer_norm( + x_137, + (384,), + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ + ) = None + x_138 = torch._C._nn.linear( + layer_norm_27, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_27 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_139 = torch._C._nn.gelu(x_138, approximate="none") + x_138 = None + x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) + x_139 = None + x_141 = torch._C._nn.linear( + x_140, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_140 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + mul_41 = l_self_modules_blocks_modules_13_parameters_gamma_2_ * x_142 + l_self_modules_blocks_modules_13_parameters_gamma_2_ = x_142 = None + x_143 = x_137 + mul_41 + x_137 = mul_41 = None + layer_norm_28 = torch.nn.functional.layer_norm( + x_143, + (384,), + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ + ) = None + linear_84 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_28 = ( + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_28 = linear_84.reshape(1, 196, 3, 8, 48) + linear_84 = None + qkv_14 = reshape_28.permute(2, 0, 3, 1, 4) + reshape_28 = None + getitem_46 = qkv_14[0] + q_14 = getitem_46 * 0.14433756729740643 + getitem_46 = None + k_14 = qkv_14[1] + v_14 = qkv_14[2] + qkv_14 = None + transpose_29 = k_14.transpose(-2, -1) + k_14 = None + attn_70 = q_14 @ transpose_29 + q_14 = transpose_29 = None + permute_71 = attn_70.permute(0, 2, 3, 1) + attn_70 = None + linear_85 = torch._C._nn.linear( + permute_71, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_71 = l_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_71 = linear_85.permute(0, 3, 1, 2) + linear_85 = None + attn_72 = attn_71.softmax(dim=-1) + attn_71 = None + permute_73 = attn_72.permute(0, 2, 3, 1) + attn_72 = None + linear_86 = torch._C._nn.linear( + permute_73, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_73 = l_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_73 = linear_86.permute(0, 3, 1, 2) + linear_86 = None + attn_74 = torch.nn.functional.dropout(attn_73, 0.0, False, False) + attn_73 = None + matmul_29 = attn_74 @ v_14 + attn_74 = v_14 = None + transpose_30 = matmul_29.transpose(1, 2) + matmul_29 = None + x_144 = transpose_30.reshape(1, 196, 384) + transpose_30 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_, + ) + x_144 = l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ + ) = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + mul_43 = l_self_modules_blocks_modules_14_parameters_gamma_1_ * x_146 + l_self_modules_blocks_modules_14_parameters_gamma_1_ = x_146 = None + x_147 = x_143 + mul_43 + x_143 = mul_43 = None + layer_norm_29 = torch.nn.functional.layer_norm( + x_147, + (384,), + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ + ) = None + x_148 = torch._C._nn.linear( + layer_norm_29, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_29 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_149 = torch._C._nn.gelu(x_148, approximate="none") + x_148 = None + x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) + x_149 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_150 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + mul_44 = l_self_modules_blocks_modules_14_parameters_gamma_2_ * x_152 + l_self_modules_blocks_modules_14_parameters_gamma_2_ = x_152 = None + x_153 = x_147 + mul_44 + x_147 = mul_44 = None + layer_norm_30 = torch.nn.functional.layer_norm( + x_153, + (384,), + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ + ) = None + linear_90 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_30 = ( + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_30 = linear_90.reshape(1, 196, 3, 8, 48) + linear_90 = None + qkv_15 = reshape_30.permute(2, 0, 3, 1, 4) + reshape_30 = None + getitem_49 = qkv_15[0] + q_15 = getitem_49 * 0.14433756729740643 + getitem_49 = None + k_15 = qkv_15[1] + v_15 = qkv_15[2] + qkv_15 = None + transpose_31 = k_15.transpose(-2, -1) + k_15 = None + attn_75 = q_15 @ transpose_31 + q_15 = transpose_31 = None + permute_76 = attn_75.permute(0, 2, 3, 1) + attn_75 = None + linear_91 = torch._C._nn.linear( + permute_76, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_76 = l_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_76 = linear_91.permute(0, 3, 1, 2) + linear_91 = None + attn_77 = attn_76.softmax(dim=-1) + attn_76 = None + permute_78 = attn_77.permute(0, 2, 3, 1) + attn_77 = None + linear_92 = torch._C._nn.linear( + permute_78, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_78 = l_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_78 = linear_92.permute(0, 3, 1, 2) + linear_92 = None + attn_79 = torch.nn.functional.dropout(attn_78, 0.0, False, False) + attn_78 = None + matmul_31 = attn_79 @ v_15 + attn_79 = v_15 = None + transpose_32 = matmul_31.transpose(1, 2) + matmul_31 = None + x_154 = transpose_32.reshape(1, 196, 384) + transpose_32 = None + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_, + ) + x_154 = l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ + ) = None + x_156 = torch.nn.functional.dropout(x_155, 0.0, False, False) + x_155 = None + mul_46 = l_self_modules_blocks_modules_15_parameters_gamma_1_ * x_156 + l_self_modules_blocks_modules_15_parameters_gamma_1_ = x_156 = None + x_157 = x_153 + mul_46 + x_153 = mul_46 = None + layer_norm_31 = torch.nn.functional.layer_norm( + x_157, + (384,), + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ + ) = None + x_158 = torch._C._nn.linear( + layer_norm_31, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_31 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_159 = torch._C._nn.gelu(x_158, approximate="none") + x_158 = None + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = torch._C._nn.linear( + x_160, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_160 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_162 = torch.nn.functional.dropout(x_161, 0.0, False, False) + x_161 = None + mul_47 = l_self_modules_blocks_modules_15_parameters_gamma_2_ * x_162 + l_self_modules_blocks_modules_15_parameters_gamma_2_ = x_162 = None + x_163 = x_157 + mul_47 + x_157 = mul_47 = None + layer_norm_32 = torch.nn.functional.layer_norm( + x_163, + (384,), + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ + ) = None + linear_96 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_32 = ( + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_32 = linear_96.reshape(1, 196, 3, 8, 48) + linear_96 = None + qkv_16 = reshape_32.permute(2, 0, 3, 1, 4) + reshape_32 = None + getitem_52 = qkv_16[0] + q_16 = getitem_52 * 0.14433756729740643 + getitem_52 = None + k_16 = qkv_16[1] + v_16 = qkv_16[2] + qkv_16 = None + transpose_33 = k_16.transpose(-2, -1) + k_16 = None + attn_80 = q_16 @ transpose_33 + q_16 = transpose_33 = None + permute_81 = attn_80.permute(0, 2, 3, 1) + attn_80 = None + linear_97 = torch._C._nn.linear( + permute_81, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_81 = l_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_81 = linear_97.permute(0, 3, 1, 2) + linear_97 = None + attn_82 = attn_81.softmax(dim=-1) + attn_81 = None + permute_83 = attn_82.permute(0, 2, 3, 1) + attn_82 = None + linear_98 = torch._C._nn.linear( + permute_83, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_83 = l_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_83 = linear_98.permute(0, 3, 1, 2) + linear_98 = None + attn_84 = torch.nn.functional.dropout(attn_83, 0.0, False, False) + attn_83 = None + matmul_33 = attn_84 @ v_16 + attn_84 = v_16 = None + transpose_34 = matmul_33.transpose(1, 2) + matmul_33 = None + x_164 = transpose_34.reshape(1, 196, 384) + transpose_34 = None + x_165 = torch._C._nn.linear( + x_164, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_, + ) + x_164 = l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ + ) = None + x_166 = torch.nn.functional.dropout(x_165, 0.0, False, False) + x_165 = None + mul_49 = l_self_modules_blocks_modules_16_parameters_gamma_1_ * x_166 + l_self_modules_blocks_modules_16_parameters_gamma_1_ = x_166 = None + x_167 = x_163 + mul_49 + x_163 = mul_49 = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_167, + (384,), + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ + ) = None + x_168 = torch._C._nn.linear( + layer_norm_33, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_33 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_169 = torch._C._nn.gelu(x_168, approximate="none") + x_168 = None + x_170 = torch.nn.functional.dropout(x_169, 0.0, False, False) + x_169 = None + x_171 = torch._C._nn.linear( + x_170, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_170 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_172 = torch.nn.functional.dropout(x_171, 0.0, False, False) + x_171 = None + mul_50 = l_self_modules_blocks_modules_16_parameters_gamma_2_ * x_172 + l_self_modules_blocks_modules_16_parameters_gamma_2_ = x_172 = None + x_173 = x_167 + mul_50 + x_167 = mul_50 = None + layer_norm_34 = torch.nn.functional.layer_norm( + x_173, + (384,), + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ + ) = None + linear_102 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_34 = ( + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_34 = linear_102.reshape(1, 196, 3, 8, 48) + linear_102 = None + qkv_17 = reshape_34.permute(2, 0, 3, 1, 4) + reshape_34 = None + getitem_55 = qkv_17[0] + q_17 = getitem_55 * 0.14433756729740643 + getitem_55 = None + k_17 = qkv_17[1] + v_17 = qkv_17[2] + qkv_17 = None + transpose_35 = k_17.transpose(-2, -1) + k_17 = None + attn_85 = q_17 @ transpose_35 + q_17 = transpose_35 = None + permute_86 = attn_85.permute(0, 2, 3, 1) + attn_85 = None + linear_103 = torch._C._nn.linear( + permute_86, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_86 = l_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_86 = linear_103.permute(0, 3, 1, 2) + linear_103 = None + attn_87 = attn_86.softmax(dim=-1) + attn_86 = None + permute_88 = attn_87.permute(0, 2, 3, 1) + attn_87 = None + linear_104 = torch._C._nn.linear( + permute_88, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_88 = l_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_88 = linear_104.permute(0, 3, 1, 2) + linear_104 = None + attn_89 = torch.nn.functional.dropout(attn_88, 0.0, False, False) + attn_88 = None + matmul_35 = attn_89 @ v_17 + attn_89 = v_17 = None + transpose_36 = matmul_35.transpose(1, 2) + matmul_35 = None + x_174 = transpose_36.reshape(1, 196, 384) + transpose_36 = None + x_175 = torch._C._nn.linear( + x_174, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_, + ) + x_174 = l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ + ) = None + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + mul_52 = l_self_modules_blocks_modules_17_parameters_gamma_1_ * x_176 + l_self_modules_blocks_modules_17_parameters_gamma_1_ = x_176 = None + x_177 = x_173 + mul_52 + x_173 = mul_52 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_177, + (384,), + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ + ) = None + x_178 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_35 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_179 = torch._C._nn.gelu(x_178, approximate="none") + x_178 = None + x_180 = torch.nn.functional.dropout(x_179, 0.0, False, False) + x_179 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + mul_53 = l_self_modules_blocks_modules_17_parameters_gamma_2_ * x_182 + l_self_modules_blocks_modules_17_parameters_gamma_2_ = x_182 = None + x_183 = x_177 + mul_53 + x_177 = mul_53 = None + layer_norm_36 = torch.nn.functional.layer_norm( + x_183, + (384,), + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ + ) = None + linear_108 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_36 = ( + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_36 = linear_108.reshape(1, 196, 3, 8, 48) + linear_108 = None + qkv_18 = reshape_36.permute(2, 0, 3, 1, 4) + reshape_36 = None + getitem_58 = qkv_18[0] + q_18 = getitem_58 * 0.14433756729740643 + getitem_58 = None + k_18 = qkv_18[1] + v_18 = qkv_18[2] + qkv_18 = None + transpose_37 = k_18.transpose(-2, -1) + k_18 = None + attn_90 = q_18 @ transpose_37 + q_18 = transpose_37 = None + permute_91 = attn_90.permute(0, 2, 3, 1) + attn_90 = None + linear_109 = torch._C._nn.linear( + permute_91, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_91 = l_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_91 = linear_109.permute(0, 3, 1, 2) + linear_109 = None + attn_92 = attn_91.softmax(dim=-1) + attn_91 = None + permute_93 = attn_92.permute(0, 2, 3, 1) + attn_92 = None + linear_110 = torch._C._nn.linear( + permute_93, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_93 = l_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_93 = linear_110.permute(0, 3, 1, 2) + linear_110 = None + attn_94 = torch.nn.functional.dropout(attn_93, 0.0, False, False) + attn_93 = None + matmul_37 = attn_94 @ v_18 + attn_94 = v_18 = None + transpose_38 = matmul_37.transpose(1, 2) + matmul_37 = None + x_184 = transpose_38.reshape(1, 196, 384) + transpose_38 = None + x_185 = torch._C._nn.linear( + x_184, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_, + ) + x_184 = l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ + ) = None + x_186 = torch.nn.functional.dropout(x_185, 0.0, False, False) + x_185 = None + mul_55 = l_self_modules_blocks_modules_18_parameters_gamma_1_ * x_186 + l_self_modules_blocks_modules_18_parameters_gamma_1_ = x_186 = None + x_187 = x_183 + mul_55 + x_183 = mul_55 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_187, + (384,), + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ + ) = None + x_188 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_37 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_189 = torch._C._nn.gelu(x_188, approximate="none") + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = torch._C._nn.linear( + x_190, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_190 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_192 = torch.nn.functional.dropout(x_191, 0.0, False, False) + x_191 = None + mul_56 = l_self_modules_blocks_modules_18_parameters_gamma_2_ * x_192 + l_self_modules_blocks_modules_18_parameters_gamma_2_ = x_192 = None + x_193 = x_187 + mul_56 + x_187 = mul_56 = None + layer_norm_38 = torch.nn.functional.layer_norm( + x_193, + (384,), + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ + ) = None + linear_114 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_38 = ( + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_38 = linear_114.reshape(1, 196, 3, 8, 48) + linear_114 = None + qkv_19 = reshape_38.permute(2, 0, 3, 1, 4) + reshape_38 = None + getitem_61 = qkv_19[0] + q_19 = getitem_61 * 0.14433756729740643 + getitem_61 = None + k_19 = qkv_19[1] + v_19 = qkv_19[2] + qkv_19 = None + transpose_39 = k_19.transpose(-2, -1) + k_19 = None + attn_95 = q_19 @ transpose_39 + q_19 = transpose_39 = None + permute_96 = attn_95.permute(0, 2, 3, 1) + attn_95 = None + linear_115 = torch._C._nn.linear( + permute_96, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_96 = l_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_96 = linear_115.permute(0, 3, 1, 2) + linear_115 = None + attn_97 = attn_96.softmax(dim=-1) + attn_96 = None + permute_98 = attn_97.permute(0, 2, 3, 1) + attn_97 = None + linear_116 = torch._C._nn.linear( + permute_98, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_98 = l_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_98 = linear_116.permute(0, 3, 1, 2) + linear_116 = None + attn_99 = torch.nn.functional.dropout(attn_98, 0.0, False, False) + attn_98 = None + matmul_39 = attn_99 @ v_19 + attn_99 = v_19 = None + transpose_40 = matmul_39.transpose(1, 2) + matmul_39 = None + x_194 = transpose_40.reshape(1, 196, 384) + transpose_40 = None + x_195 = torch._C._nn.linear( + x_194, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_, + ) + x_194 = l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ + ) = None + x_196 = torch.nn.functional.dropout(x_195, 0.0, False, False) + x_195 = None + mul_58 = l_self_modules_blocks_modules_19_parameters_gamma_1_ * x_196 + l_self_modules_blocks_modules_19_parameters_gamma_1_ = x_196 = None + x_197 = x_193 + mul_58 + x_193 = mul_58 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_197, + (384,), + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ + ) = None + x_198 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_39 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_199 = torch._C._nn.gelu(x_198, approximate="none") + x_198 = None + x_200 = torch.nn.functional.dropout(x_199, 0.0, False, False) + x_199 = None + x_201 = torch._C._nn.linear( + x_200, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_200 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_202 = torch.nn.functional.dropout(x_201, 0.0, False, False) + x_201 = None + mul_59 = l_self_modules_blocks_modules_19_parameters_gamma_2_ * x_202 + l_self_modules_blocks_modules_19_parameters_gamma_2_ = x_202 = None + x_203 = x_197 + mul_59 + x_197 = mul_59 = None + layer_norm_40 = torch.nn.functional.layer_norm( + x_203, + (384,), + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ + ) = None + linear_120 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_40 = ( + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_40 = linear_120.reshape(1, 196, 3, 8, 48) + linear_120 = None + qkv_20 = reshape_40.permute(2, 0, 3, 1, 4) + reshape_40 = None + getitem_64 = qkv_20[0] + q_20 = getitem_64 * 0.14433756729740643 + getitem_64 = None + k_20 = qkv_20[1] + v_20 = qkv_20[2] + qkv_20 = None + transpose_41 = k_20.transpose(-2, -1) + k_20 = None + attn_100 = q_20 @ transpose_41 + q_20 = transpose_41 = None + permute_101 = attn_100.permute(0, 2, 3, 1) + attn_100 = None + linear_121 = torch._C._nn.linear( + permute_101, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_101 = l_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_101 = linear_121.permute(0, 3, 1, 2) + linear_121 = None + attn_102 = attn_101.softmax(dim=-1) + attn_101 = None + permute_103 = attn_102.permute(0, 2, 3, 1) + attn_102 = None + linear_122 = torch._C._nn.linear( + permute_103, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_103 = l_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_103 = linear_122.permute(0, 3, 1, 2) + linear_122 = None + attn_104 = torch.nn.functional.dropout(attn_103, 0.0, False, False) + attn_103 = None + matmul_41 = attn_104 @ v_20 + attn_104 = v_20 = None + transpose_42 = matmul_41.transpose(1, 2) + matmul_41 = None + x_204 = transpose_42.reshape(1, 196, 384) + transpose_42 = None + x_205 = torch._C._nn.linear( + x_204, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_, + ) + x_204 = l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ + ) = None + x_206 = torch.nn.functional.dropout(x_205, 0.0, False, False) + x_205 = None + mul_61 = l_self_modules_blocks_modules_20_parameters_gamma_1_ * x_206 + l_self_modules_blocks_modules_20_parameters_gamma_1_ = x_206 = None + x_207 = x_203 + mul_61 + x_203 = mul_61 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_207, + (384,), + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ + ) = None + x_208 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_41 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_209 = torch._C._nn.gelu(x_208, approximate="none") + x_208 = None + x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) + x_209 = None + x_211 = torch._C._nn.linear( + x_210, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_210 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + mul_62 = l_self_modules_blocks_modules_20_parameters_gamma_2_ * x_212 + l_self_modules_blocks_modules_20_parameters_gamma_2_ = x_212 = None + x_213 = x_207 + mul_62 + x_207 = mul_62 = None + layer_norm_42 = torch.nn.functional.layer_norm( + x_213, + (384,), + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ + ) = None + linear_126 = torch._C._nn.linear( + layer_norm_42, + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_42 = ( + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_42 = linear_126.reshape(1, 196, 3, 8, 48) + linear_126 = None + qkv_21 = reshape_42.permute(2, 0, 3, 1, 4) + reshape_42 = None + getitem_67 = qkv_21[0] + q_21 = getitem_67 * 0.14433756729740643 + getitem_67 = None + k_21 = qkv_21[1] + v_21 = qkv_21[2] + qkv_21 = None + transpose_43 = k_21.transpose(-2, -1) + k_21 = None + attn_105 = q_21 @ transpose_43 + q_21 = transpose_43 = None + permute_106 = attn_105.permute(0, 2, 3, 1) + attn_105 = None + linear_127 = torch._C._nn.linear( + permute_106, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_106 = l_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_106 = linear_127.permute(0, 3, 1, 2) + linear_127 = None + attn_107 = attn_106.softmax(dim=-1) + attn_106 = None + permute_108 = attn_107.permute(0, 2, 3, 1) + attn_107 = None + linear_128 = torch._C._nn.linear( + permute_108, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_108 = l_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_108 = linear_128.permute(0, 3, 1, 2) + linear_128 = None + attn_109 = torch.nn.functional.dropout(attn_108, 0.0, False, False) + attn_108 = None + matmul_43 = attn_109 @ v_21 + attn_109 = v_21 = None + transpose_44 = matmul_43.transpose(1, 2) + matmul_43 = None + x_214 = transpose_44.reshape(1, 196, 384) + transpose_44 = None + x_215 = torch._C._nn.linear( + x_214, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_, + ) + x_214 = l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ + ) = None + x_216 = torch.nn.functional.dropout(x_215, 0.0, False, False) + x_215 = None + mul_64 = l_self_modules_blocks_modules_21_parameters_gamma_1_ * x_216 + l_self_modules_blocks_modules_21_parameters_gamma_1_ = x_216 = None + x_217 = x_213 + mul_64 + x_213 = mul_64 = None + layer_norm_43 = torch.nn.functional.layer_norm( + x_217, + (384,), + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ + ) = None + x_218 = torch._C._nn.linear( + layer_norm_43, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_43 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_219 = torch._C._nn.gelu(x_218, approximate="none") + x_218 = None + x_220 = torch.nn.functional.dropout(x_219, 0.0, False, False) + x_219 = None + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_220 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_222 = torch.nn.functional.dropout(x_221, 0.0, False, False) + x_221 = None + mul_65 = l_self_modules_blocks_modules_21_parameters_gamma_2_ * x_222 + l_self_modules_blocks_modules_21_parameters_gamma_2_ = x_222 = None + x_223 = x_217 + mul_65 + x_217 = mul_65 = None + layer_norm_44 = torch.nn.functional.layer_norm( + x_223, + (384,), + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ + ) = None + linear_132 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_44 = ( + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_44 = linear_132.reshape(1, 196, 3, 8, 48) + linear_132 = None + qkv_22 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + getitem_70 = qkv_22[0] + q_22 = getitem_70 * 0.14433756729740643 + getitem_70 = None + k_22 = qkv_22[1] + v_22 = qkv_22[2] + qkv_22 = None + transpose_45 = k_22.transpose(-2, -1) + k_22 = None + attn_110 = q_22 @ transpose_45 + q_22 = transpose_45 = None + permute_111 = attn_110.permute(0, 2, 3, 1) + attn_110 = None + linear_133 = torch._C._nn.linear( + permute_111, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_111 = l_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_111 = linear_133.permute(0, 3, 1, 2) + linear_133 = None + attn_112 = attn_111.softmax(dim=-1) + attn_111 = None + permute_113 = attn_112.permute(0, 2, 3, 1) + attn_112 = None + linear_134 = torch._C._nn.linear( + permute_113, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_113 = l_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_113 = linear_134.permute(0, 3, 1, 2) + linear_134 = None + attn_114 = torch.nn.functional.dropout(attn_113, 0.0, False, False) + attn_113 = None + matmul_45 = attn_114 @ v_22 + attn_114 = v_22 = None + transpose_46 = matmul_45.transpose(1, 2) + matmul_45 = None + x_224 = transpose_46.reshape(1, 196, 384) + transpose_46 = None + x_225 = torch._C._nn.linear( + x_224, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_, + ) + x_224 = l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ + ) = None + x_226 = torch.nn.functional.dropout(x_225, 0.0, False, False) + x_225 = None + mul_67 = l_self_modules_blocks_modules_22_parameters_gamma_1_ * x_226 + l_self_modules_blocks_modules_22_parameters_gamma_1_ = x_226 = None + x_227 = x_223 + mul_67 + x_223 = mul_67 = None + layer_norm_45 = torch.nn.functional.layer_norm( + x_227, + (384,), + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ + ) = None + x_228 = torch._C._nn.linear( + layer_norm_45, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_45 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_229 = torch._C._nn.gelu(x_228, approximate="none") + x_228 = None + x_230 = torch.nn.functional.dropout(x_229, 0.0, False, False) + x_229 = None + x_231 = torch._C._nn.linear( + x_230, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_230 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_232 = torch.nn.functional.dropout(x_231, 0.0, False, False) + x_231 = None + mul_68 = l_self_modules_blocks_modules_22_parameters_gamma_2_ * x_232 + l_self_modules_blocks_modules_22_parameters_gamma_2_ = x_232 = None + x_233 = x_227 + mul_68 + x_227 = mul_68 = None + layer_norm_46 = torch.nn.functional.layer_norm( + x_233, + (384,), + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ + ) = None + linear_138 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_46 = ( + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_46 = linear_138.reshape(1, 196, 3, 8, 48) + linear_138 = None + qkv_23 = reshape_46.permute(2, 0, 3, 1, 4) + reshape_46 = None + getitem_73 = qkv_23[0] + q_23 = getitem_73 * 0.14433756729740643 + getitem_73 = None + k_23 = qkv_23[1] + v_23 = qkv_23[2] + qkv_23 = None + transpose_47 = k_23.transpose(-2, -1) + k_23 = None + attn_115 = q_23 @ transpose_47 + q_23 = transpose_47 = None + permute_116 = attn_115.permute(0, 2, 3, 1) + attn_115 = None + linear_139 = torch._C._nn.linear( + permute_116, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_116 = l_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_116 = linear_139.permute(0, 3, 1, 2) + linear_139 = None + attn_117 = attn_116.softmax(dim=-1) + attn_116 = None + permute_118 = attn_117.permute(0, 2, 3, 1) + attn_117 = None + linear_140 = torch._C._nn.linear( + permute_118, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_118 = l_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_118 = linear_140.permute(0, 3, 1, 2) + linear_140 = None + attn_119 = torch.nn.functional.dropout(attn_118, 0.0, False, False) + attn_118 = None + matmul_47 = attn_119 @ v_23 + attn_119 = v_23 = None + transpose_48 = matmul_47.transpose(1, 2) + matmul_47 = None + x_234 = transpose_48.reshape(1, 196, 384) + transpose_48 = None + x_235 = torch._C._nn.linear( + x_234, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_, + ) + x_234 = l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ + ) = None + x_236 = torch.nn.functional.dropout(x_235, 0.0, False, False) + x_235 = None + mul_70 = l_self_modules_blocks_modules_23_parameters_gamma_1_ * x_236 + l_self_modules_blocks_modules_23_parameters_gamma_1_ = x_236 = None + x_237 = x_233 + mul_70 + x_233 = mul_70 = None + layer_norm_47 = torch.nn.functional.layer_norm( + x_237, + (384,), + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ + ) = None + x_238 = torch._C._nn.linear( + layer_norm_47, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_47 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_239 = torch._C._nn.gelu(x_238, approximate="none") + x_238 = None + x_240 = torch.nn.functional.dropout(x_239, 0.0, False, False) + x_239 = None + x_241 = torch._C._nn.linear( + x_240, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_240 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) + x_241 = None + mul_71 = l_self_modules_blocks_modules_23_parameters_gamma_2_ * x_242 + l_self_modules_blocks_modules_23_parameters_gamma_2_ = x_242 = None + x_243 = x_237 + mul_71 + x_237 = mul_71 = None + cls_tokens = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + u = torch.cat((cls_tokens, x_243), dim=1) + layer_norm_48 = torch.nn.functional.layer_norm( + u, + (384,), + l_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + u = ( + l_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_weight_ + ) = ( + l_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_bias_ + ) = None + getitem_76 = layer_norm_48[(slice(None, None, None), 0)] + linear_144 = torch._C._nn.linear( + getitem_76, + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_weight_, + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_bias_, + ) + getitem_76 = l_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_weight_ = l_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_bias_ = (None) + unsqueeze = linear_144.unsqueeze(1) + linear_144 = None + reshape_48 = unsqueeze.reshape(1, 1, 8, 48) + unsqueeze = None + q_24 = reshape_48.permute(0, 2, 1, 3) + reshape_48 = None + linear_145 = torch._C._nn.linear( + layer_norm_48, + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_weight_, + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_bias_, + ) + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_weight_ = l_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_bias_ = (None) + reshape_49 = linear_145.reshape(1, 197, 8, 48) + linear_145 = None + k_24 = reshape_49.permute(0, 2, 1, 3) + reshape_49 = None + linear_146 = torch._C._nn.linear( + layer_norm_48, + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_weight_, + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_bias_, + ) + layer_norm_48 = l_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_weight_ = l_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_bias_ = (None) + reshape_50 = linear_146.reshape(1, 197, 8, 48) + linear_146 = None + v_24 = reshape_50.permute(0, 2, 1, 3) + reshape_50 = None + x_cls = torch._C._nn.scaled_dot_product_attention( + q_24, k_24, v_24, dropout_p=0.0 + ) + q_24 = k_24 = v_24 = None + transpose_49 = x_cls.transpose(1, 2) + x_cls = None + x_cls_1 = transpose_49.reshape(1, 1, 384) + transpose_49 = None + x_cls_2 = torch._C._nn.linear( + x_cls_1, + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_cls_1 = l_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_cls_3 = torch.nn.functional.dropout(x_cls_2, 0.0, False, False) + x_cls_2 = None + mul_72 = ( + l_self_modules_blocks_token_only_modules_0_parameters_gamma_1_ * x_cls_3 + ) + l_self_modules_blocks_token_only_modules_0_parameters_gamma_1_ = x_cls_3 = None + x_cls_4 = cls_tokens + mul_72 + cls_tokens = mul_72 = None + layer_norm_49 = torch.nn.functional.layer_norm( + x_cls_4, + (384,), + l_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_bias_ + ) = None + x_244 = torch._C._nn.linear( + layer_norm_49, + l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_49 = l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_245 = torch._C._nn.gelu(x_244, approximate="none") + x_244 = None + x_246 = torch.nn.functional.dropout(x_245, 0.0, False, False) + x_245 = None + x_247 = torch._C._nn.linear( + x_246, + l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_246 = l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_248 = torch.nn.functional.dropout(x_247, 0.0, False, False) + x_247 = None + mul_73 = l_self_modules_blocks_token_only_modules_0_parameters_gamma_2_ * x_248 + l_self_modules_blocks_token_only_modules_0_parameters_gamma_2_ = x_248 = None + x_cls_5 = x_cls_4 + mul_73 + x_cls_4 = mul_73 = None + u_1 = torch.cat((x_cls_5, x_243), dim=1) + layer_norm_50 = torch.nn.functional.layer_norm( + u_1, + (384,), + l_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + u_1 = ( + l_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_weight_ + ) = ( + l_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_bias_ + ) = None + getitem_77 = layer_norm_50[(slice(None, None, None), 0)] + linear_150 = torch._C._nn.linear( + getitem_77, + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_weight_, + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_bias_, + ) + getitem_77 = l_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_weight_ = l_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_bias_ = (None) + unsqueeze_1 = linear_150.unsqueeze(1) + linear_150 = None + reshape_52 = unsqueeze_1.reshape(1, 1, 8, 48) + unsqueeze_1 = None + q_25 = reshape_52.permute(0, 2, 1, 3) + reshape_52 = None + linear_151 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_weight_, + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_bias_, + ) + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_weight_ = l_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_bias_ = (None) + reshape_53 = linear_151.reshape(1, 197, 8, 48) + linear_151 = None + k_25 = reshape_53.permute(0, 2, 1, 3) + reshape_53 = None + linear_152 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_weight_, + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_bias_, + ) + layer_norm_50 = l_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_weight_ = l_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_bias_ = (None) + reshape_54 = linear_152.reshape(1, 197, 8, 48) + linear_152 = None + v_25 = reshape_54.permute(0, 2, 1, 3) + reshape_54 = None + x_cls_6 = torch._C._nn.scaled_dot_product_attention( + q_25, k_25, v_25, dropout_p=0.0 + ) + q_25 = k_25 = v_25 = None + transpose_50 = x_cls_6.transpose(1, 2) + x_cls_6 = None + x_cls_7 = transpose_50.reshape(1, 1, 384) + transpose_50 = None + x_cls_8 = torch._C._nn.linear( + x_cls_7, + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_cls_7 = l_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_cls_9 = torch.nn.functional.dropout(x_cls_8, 0.0, False, False) + x_cls_8 = None + mul_74 = ( + l_self_modules_blocks_token_only_modules_1_parameters_gamma_1_ * x_cls_9 + ) + l_self_modules_blocks_token_only_modules_1_parameters_gamma_1_ = x_cls_9 = None + x_cls_10 = x_cls_5 + mul_74 + x_cls_5 = mul_74 = None + layer_norm_51 = torch.nn.functional.layer_norm( + x_cls_10, + (384,), + l_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_bias_ + ) = None + x_249 = torch._C._nn.linear( + layer_norm_51, + l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_51 = l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_250 = torch._C._nn.gelu(x_249, approximate="none") + x_249 = None + x_251 = torch.nn.functional.dropout(x_250, 0.0, False, False) + x_250 = None + x_252 = torch._C._nn.linear( + x_251, + l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_251 = l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_253 = torch.nn.functional.dropout(x_252, 0.0, False, False) + x_252 = None + mul_75 = l_self_modules_blocks_token_only_modules_1_parameters_gamma_2_ * x_253 + l_self_modules_blocks_token_only_modules_1_parameters_gamma_2_ = x_253 = None + x_cls_11 = x_cls_10 + mul_75 + x_cls_10 = mul_75 = None + x_254 = torch.cat((x_cls_11, x_243), dim=1) + x_cls_11 = x_243 = None + x_255 = torch.nn.functional.layer_norm( + x_254, + (384,), + l_self_modules_norm_parameters_weight_, + l_self_modules_norm_parameters_bias_, + 1e-06, + ) + x_254 = ( + l_self_modules_norm_parameters_weight_ + ) = l_self_modules_norm_parameters_bias_ = None + x_256 = x_255[(slice(None, None, None), 0)] + x_255 = None + x_257 = torch.nn.functional.dropout(x_256, 0.0, False, False) + x_256 = None + x_258 = torch._C._nn.linear( + x_257, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_257 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_258,) diff --git a/samples/timm/cait_s24_224.fb_dist_in1k/weight_meta.py b/samples/timm/cait_s24_224.fb_dist_in1k/weight_meta.py new file mode 100644 index 0000000000..51562a3d0d --- /dev/null +++ b/samples/timm/cait_s24_224.fb_dist_in1k/weight_meta.py @@ -0,0 +1,4980 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.230 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [384, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 196, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_0_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_0_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_1_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_1_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_2_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_2_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_3_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_3_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_4_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.023 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_4_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_5_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.004 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_5_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_6_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_6_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_7_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.004 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_7_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_8_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.017 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.004 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_8_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_9_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.005 + std = 0.017 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_9_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_10_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.005 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_10_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_11_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_11_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_12_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_12_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_13_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_13_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_14_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_14_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_15_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.023 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_15_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_16_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.008 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.005 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_16_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_17_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_17_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_18_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.007 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_18_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_19_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_19_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_20_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_20_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_21_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_21_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_22_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_22_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_23_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.023 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_weight_" + shape = [8, 8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_23_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_parameters_gamma_1_: + name = "L_self_modules_blocks_token_only_modules_0_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_parameters_gamma_2_: + name = "L_self_modules_blocks_token_only_modules_0_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_parameters_gamma_1_: + name = "L_self_modules_blocks_token_only_modules_1_parameters_gamma_1_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_parameters_gamma_2_: + name = "L_self_modules_blocks_token_only_modules_1_parameters_gamma_2_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_weight_: + name = "L_self_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_bias_: + name = "L_self_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cait_xxs24_224.fb_dist_in1k/graph_hash.txt b/samples/timm/cait_xxs24_224.fb_dist_in1k/graph_hash.txt new file mode 100644 index 0000000000..2810dae274 --- /dev/null +++ b/samples/timm/cait_xxs24_224.fb_dist_in1k/graph_hash.txt @@ -0,0 +1 @@ +5e480860975c00b127e98aacc223c6966c3aa932ab913bbc461632d495e346df \ No newline at end of file diff --git a/samples/timm/cait_xxs24_224.fb_dist_in1k/graph_net.json b/samples/timm/cait_xxs24_224.fb_dist_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/cait_xxs24_224.fb_dist_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cait_xxs24_224.fb_dist_in1k/input_meta.py b/samples/timm/cait_xxs24_224.fb_dist_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cait_xxs24_224.fb_dist_in1k/input_tensor_constraints.py b/samples/timm/cait_xxs24_224.fb_dist_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cait_xxs24_224.fb_dist_in1k/model.py b/samples/timm/cait_xxs24_224.fb_dist_in1k/model.py new file mode 100644 index 0000000000..f0af5f4a19 --- /dev/null +++ b/samples/timm/cait_xxs24_224.fb_dist_in1k/model.py @@ -0,0 +1,4812 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_modules_blocks_modules_0_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_0_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_0_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_1_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_1_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_2_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_2_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_3_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_3_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_4_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_4_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_5_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_5_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_6_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_6_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_7_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_7_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_8_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_8_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_9_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_9_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_10_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_10_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_11_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_11_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_12_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_12_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_12_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_12_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_13_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_13_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_13_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_13_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_14_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_14_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_14_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_14_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_15_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_15_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_15_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_15_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_16_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_16_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_16_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_16_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_17_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_17_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_17_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_17_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_18_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_18_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_18_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_18_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_19_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_19_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_19_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_19_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_20_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_20_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_20_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_20_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_21_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_21_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_21_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_21_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_22_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_22_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_22_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_22_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_23_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_23_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_23_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_23_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_modules_blocks_token_only_modules_0_parameters_gamma_1_ = ( + L_self_modules_blocks_token_only_modules_0_parameters_gamma_1_ + ) + l_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_weight_ = L_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_weight_ + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_bias_ = L_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_bias_ + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_weight_ = L_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_weight_ + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_bias_ = L_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_bias_ + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_weight_ = L_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_weight_ + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_bias_ = L_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_bias_ + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_token_only_modules_0_parameters_gamma_2_ = ( + L_self_modules_blocks_token_only_modules_0_parameters_gamma_2_ + ) + l_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_token_only_modules_1_parameters_gamma_1_ = ( + L_self_modules_blocks_token_only_modules_1_parameters_gamma_1_ + ) + l_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_weight_ = L_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_weight_ + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_bias_ = L_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_bias_ + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_weight_ = L_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_weight_ + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_bias_ = L_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_bias_ + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_weight_ = L_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_weight_ + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_bias_ = L_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_bias_ + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_token_only_modules_1_parameters_gamma_2_ = ( + L_self_modules_blocks_token_only_modules_1_parameters_gamma_2_ + ) + l_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_norm_parameters_weight_ = L_self_modules_norm_parameters_weight_ + l_self_modules_norm_parameters_bias_ = L_self_modules_norm_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + x_2 = x_1 + l_self_parameters_pos_embed_ + x_1 = l_self_parameters_pos_embed_ = None + x_3 = torch.nn.functional.dropout(x_2, 0.0, False, False) + x_2 = None + layer_norm = torch.nn.functional.layer_norm( + x_3, + (192,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + layer_norm, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 196, 3, 4, 48) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + getitem_4 = qkv[0] + q = getitem_4 * 0.14433756729740643 + getitem_4 = None + k = qkv[1] + v = qkv[2] + qkv = None + transpose_1 = k.transpose(-2, -1) + k = None + attn = q @ transpose_1 + q = transpose_1 = None + permute_1 = attn.permute(0, 2, 3, 1) + attn = None + linear_1 = torch._C._nn.linear( + permute_1, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_1 = l_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_1 = linear_1.permute(0, 3, 1, 2) + linear_1 = None + attn_2 = attn_1.softmax(dim=-1) + attn_1 = None + permute_3 = attn_2.permute(0, 2, 3, 1) + attn_2 = None + linear_2 = torch._C._nn.linear( + permute_3, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_3 = l_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_3 = linear_2.permute(0, 3, 1, 2) + linear_2 = None + attn_4 = torch.nn.functional.dropout(attn_3, 0.0, False, False) + attn_3 = None + matmul_1 = attn_4 @ v + attn_4 = v = None + transpose_2 = matmul_1.transpose(1, 2) + matmul_1 = None + x_4 = transpose_2.reshape(1, 196, 192) + transpose_2 = None + x_5 = torch._C._nn.linear( + x_4, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_4 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_6 = torch.nn.functional.dropout(x_5, 0.0, False, False) + x_5 = None + mul_1 = l_self_modules_blocks_modules_0_parameters_gamma_1_ * x_6 + l_self_modules_blocks_modules_0_parameters_gamma_1_ = x_6 = None + x_7 = x_3 + mul_1 + x_3 = mul_1 = None + layer_norm_1 = torch.nn.functional.layer_norm( + x_7, + (192,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_8 = torch._C._nn.linear( + layer_norm_1, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_1 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_9 = torch._C._nn.gelu(x_8, approximate="none") + x_8 = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = torch._C._nn.linear( + x_10, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_10 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + mul_2 = l_self_modules_blocks_modules_0_parameters_gamma_2_ * x_12 + l_self_modules_blocks_modules_0_parameters_gamma_2_ = x_12 = None + x_13 = x_7 + mul_2 + x_7 = mul_2 = None + layer_norm_2 = torch.nn.functional.layer_norm( + x_13, + (192,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_6 = torch._C._nn.linear( + layer_norm_2, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_2 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_6.reshape(1, 196, 3, 4, 48) + linear_6 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + getitem_7 = qkv_1[0] + q_1 = getitem_7 * 0.14433756729740643 + getitem_7 = None + k_1 = qkv_1[1] + v_1 = qkv_1[2] + qkv_1 = None + transpose_3 = k_1.transpose(-2, -1) + k_1 = None + attn_5 = q_1 @ transpose_3 + q_1 = transpose_3 = None + permute_6 = attn_5.permute(0, 2, 3, 1) + attn_5 = None + linear_7 = torch._C._nn.linear( + permute_6, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_6 = l_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_6 = linear_7.permute(0, 3, 1, 2) + linear_7 = None + attn_7 = attn_6.softmax(dim=-1) + attn_6 = None + permute_8 = attn_7.permute(0, 2, 3, 1) + attn_7 = None + linear_8 = torch._C._nn.linear( + permute_8, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_8 = l_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_8 = linear_8.permute(0, 3, 1, 2) + linear_8 = None + attn_9 = torch.nn.functional.dropout(attn_8, 0.0, False, False) + attn_8 = None + matmul_3 = attn_9 @ v_1 + attn_9 = v_1 = None + transpose_4 = matmul_3.transpose(1, 2) + matmul_3 = None + x_14 = transpose_4.reshape(1, 196, 192) + transpose_4 = None + x_15 = torch._C._nn.linear( + x_14, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_14 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + mul_4 = l_self_modules_blocks_modules_1_parameters_gamma_1_ * x_16 + l_self_modules_blocks_modules_1_parameters_gamma_1_ = x_16 = None + x_17 = x_13 + mul_4 + x_13 = mul_4 = None + layer_norm_3 = torch.nn.functional.layer_norm( + x_17, + (192,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_18 = torch._C._nn.linear( + layer_norm_3, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_3 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_19 = torch._C._nn.gelu(x_18, approximate="none") + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + mul_5 = l_self_modules_blocks_modules_1_parameters_gamma_2_ * x_22 + l_self_modules_blocks_modules_1_parameters_gamma_2_ = x_22 = None + x_23 = x_17 + mul_5 + x_17 = mul_5 = None + layer_norm_4 = torch.nn.functional.layer_norm( + x_23, + (192,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + layer_norm_4, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_4 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_12.reshape(1, 196, 3, 4, 48) + linear_12 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + getitem_10 = qkv_2[0] + q_2 = getitem_10 * 0.14433756729740643 + getitem_10 = None + k_2 = qkv_2[1] + v_2 = qkv_2[2] + qkv_2 = None + transpose_5 = k_2.transpose(-2, -1) + k_2 = None + attn_10 = q_2 @ transpose_5 + q_2 = transpose_5 = None + permute_11 = attn_10.permute(0, 2, 3, 1) + attn_10 = None + linear_13 = torch._C._nn.linear( + permute_11, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_11 = l_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_11 = linear_13.permute(0, 3, 1, 2) + linear_13 = None + attn_12 = attn_11.softmax(dim=-1) + attn_11 = None + permute_13 = attn_12.permute(0, 2, 3, 1) + attn_12 = None + linear_14 = torch._C._nn.linear( + permute_13, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_13 = l_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_13 = linear_14.permute(0, 3, 1, 2) + linear_14 = None + attn_14 = torch.nn.functional.dropout(attn_13, 0.0, False, False) + attn_13 = None + matmul_5 = attn_14 @ v_2 + attn_14 = v_2 = None + transpose_6 = matmul_5.transpose(1, 2) + matmul_5 = None + x_24 = transpose_6.reshape(1, 196, 192) + transpose_6 = None + x_25 = torch._C._nn.linear( + x_24, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_24 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_26 = torch.nn.functional.dropout(x_25, 0.0, False, False) + x_25 = None + mul_7 = l_self_modules_blocks_modules_2_parameters_gamma_1_ * x_26 + l_self_modules_blocks_modules_2_parameters_gamma_1_ = x_26 = None + x_27 = x_23 + mul_7 + x_23 = mul_7 = None + layer_norm_5 = torch.nn.functional.layer_norm( + x_27, + (192,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_28 = torch._C._nn.linear( + layer_norm_5, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_5 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_29 = torch._C._nn.gelu(x_28, approximate="none") + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_31 = torch._C._nn.linear( + x_30, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_30 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + mul_8 = l_self_modules_blocks_modules_2_parameters_gamma_2_ * x_32 + l_self_modules_blocks_modules_2_parameters_gamma_2_ = x_32 = None + x_33 = x_27 + mul_8 + x_27 = mul_8 = None + layer_norm_6 = torch.nn.functional.layer_norm( + x_33, + (192,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_18 = torch._C._nn.linear( + layer_norm_6, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_6 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_18.reshape(1, 196, 3, 4, 48) + linear_18 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + getitem_13 = qkv_3[0] + q_3 = getitem_13 * 0.14433756729740643 + getitem_13 = None + k_3 = qkv_3[1] + v_3 = qkv_3[2] + qkv_3 = None + transpose_7 = k_3.transpose(-2, -1) + k_3 = None + attn_15 = q_3 @ transpose_7 + q_3 = transpose_7 = None + permute_16 = attn_15.permute(0, 2, 3, 1) + attn_15 = None + linear_19 = torch._C._nn.linear( + permute_16, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_16 = l_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_16 = linear_19.permute(0, 3, 1, 2) + linear_19 = None + attn_17 = attn_16.softmax(dim=-1) + attn_16 = None + permute_18 = attn_17.permute(0, 2, 3, 1) + attn_17 = None + linear_20 = torch._C._nn.linear( + permute_18, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_18 = l_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_18 = linear_20.permute(0, 3, 1, 2) + linear_20 = None + attn_19 = torch.nn.functional.dropout(attn_18, 0.0, False, False) + attn_18 = None + matmul_7 = attn_19 @ v_3 + attn_19 = v_3 = None + transpose_8 = matmul_7.transpose(1, 2) + matmul_7 = None + x_34 = transpose_8.reshape(1, 196, 192) + transpose_8 = None + x_35 = torch._C._nn.linear( + x_34, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_34 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + mul_10 = l_self_modules_blocks_modules_3_parameters_gamma_1_ * x_36 + l_self_modules_blocks_modules_3_parameters_gamma_1_ = x_36 = None + x_37 = x_33 + mul_10 + x_33 = mul_10 = None + layer_norm_7 = torch.nn.functional.layer_norm( + x_37, + (192,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_38 = torch._C._nn.linear( + layer_norm_7, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_7 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_39 = torch._C._nn.gelu(x_38, approximate="none") + x_38 = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = torch._C._nn.linear( + x_40, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_40 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_42 = torch.nn.functional.dropout(x_41, 0.0, False, False) + x_41 = None + mul_11 = l_self_modules_blocks_modules_3_parameters_gamma_2_ * x_42 + l_self_modules_blocks_modules_3_parameters_gamma_2_ = x_42 = None + x_43 = x_37 + mul_11 + x_37 = mul_11 = None + layer_norm_8 = torch.nn.functional.layer_norm( + x_43, + (192,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + layer_norm_8, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_8 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_24.reshape(1, 196, 3, 4, 48) + linear_24 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + getitem_16 = qkv_4[0] + q_4 = getitem_16 * 0.14433756729740643 + getitem_16 = None + k_4 = qkv_4[1] + v_4 = qkv_4[2] + qkv_4 = None + transpose_9 = k_4.transpose(-2, -1) + k_4 = None + attn_20 = q_4 @ transpose_9 + q_4 = transpose_9 = None + permute_21 = attn_20.permute(0, 2, 3, 1) + attn_20 = None + linear_25 = torch._C._nn.linear( + permute_21, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_21 = l_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_21 = linear_25.permute(0, 3, 1, 2) + linear_25 = None + attn_22 = attn_21.softmax(dim=-1) + attn_21 = None + permute_23 = attn_22.permute(0, 2, 3, 1) + attn_22 = None + linear_26 = torch._C._nn.linear( + permute_23, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_23 = l_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_23 = linear_26.permute(0, 3, 1, 2) + linear_26 = None + attn_24 = torch.nn.functional.dropout(attn_23, 0.0, False, False) + attn_23 = None + matmul_9 = attn_24 @ v_4 + attn_24 = v_4 = None + transpose_10 = matmul_9.transpose(1, 2) + matmul_9 = None + x_44 = transpose_10.reshape(1, 196, 192) + transpose_10 = None + x_45 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_44 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_46 = torch.nn.functional.dropout(x_45, 0.0, False, False) + x_45 = None + mul_13 = l_self_modules_blocks_modules_4_parameters_gamma_1_ * x_46 + l_self_modules_blocks_modules_4_parameters_gamma_1_ = x_46 = None + x_47 = x_43 + mul_13 + x_43 = mul_13 = None + layer_norm_9 = torch.nn.functional.layer_norm( + x_47, + (192,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_48 = torch._C._nn.linear( + layer_norm_9, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_9 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_49 = torch._C._nn.gelu(x_48, approximate="none") + x_48 = None + x_50 = torch.nn.functional.dropout(x_49, 0.0, False, False) + x_49 = None + x_51 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_52 = torch.nn.functional.dropout(x_51, 0.0, False, False) + x_51 = None + mul_14 = l_self_modules_blocks_modules_4_parameters_gamma_2_ * x_52 + l_self_modules_blocks_modules_4_parameters_gamma_2_ = x_52 = None + x_53 = x_47 + mul_14 + x_47 = mul_14 = None + layer_norm_10 = torch.nn.functional.layer_norm( + x_53, + (192,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_30 = torch._C._nn.linear( + layer_norm_10, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_10 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_30.reshape(1, 196, 3, 4, 48) + linear_30 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + getitem_19 = qkv_5[0] + q_5 = getitem_19 * 0.14433756729740643 + getitem_19 = None + k_5 = qkv_5[1] + v_5 = qkv_5[2] + qkv_5 = None + transpose_11 = k_5.transpose(-2, -1) + k_5 = None + attn_25 = q_5 @ transpose_11 + q_5 = transpose_11 = None + permute_26 = attn_25.permute(0, 2, 3, 1) + attn_25 = None + linear_31 = torch._C._nn.linear( + permute_26, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_26 = l_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_26 = linear_31.permute(0, 3, 1, 2) + linear_31 = None + attn_27 = attn_26.softmax(dim=-1) + attn_26 = None + permute_28 = attn_27.permute(0, 2, 3, 1) + attn_27 = None + linear_32 = torch._C._nn.linear( + permute_28, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_28 = l_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_28 = linear_32.permute(0, 3, 1, 2) + linear_32 = None + attn_29 = torch.nn.functional.dropout(attn_28, 0.0, False, False) + attn_28 = None + matmul_11 = attn_29 @ v_5 + attn_29 = v_5 = None + transpose_12 = matmul_11.transpose(1, 2) + matmul_11 = None + x_54 = transpose_12.reshape(1, 196, 192) + transpose_12 = None + x_55 = torch._C._nn.linear( + x_54, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_54 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_56 = torch.nn.functional.dropout(x_55, 0.0, False, False) + x_55 = None + mul_16 = l_self_modules_blocks_modules_5_parameters_gamma_1_ * x_56 + l_self_modules_blocks_modules_5_parameters_gamma_1_ = x_56 = None + x_57 = x_53 + mul_16 + x_53 = mul_16 = None + layer_norm_11 = torch.nn.functional.layer_norm( + x_57, + (192,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_58 = torch._C._nn.linear( + layer_norm_11, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_11 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_59 = torch._C._nn.gelu(x_58, approximate="none") + x_58 = None + x_60 = torch.nn.functional.dropout(x_59, 0.0, False, False) + x_59 = None + x_61 = torch._C._nn.linear( + x_60, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_60 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + mul_17 = l_self_modules_blocks_modules_5_parameters_gamma_2_ * x_62 + l_self_modules_blocks_modules_5_parameters_gamma_2_ = x_62 = None + x_63 = x_57 + mul_17 + x_57 = mul_17 = None + layer_norm_12 = torch.nn.functional.layer_norm( + x_63, + (192,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_12 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_36.reshape(1, 196, 3, 4, 48) + linear_36 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + getitem_22 = qkv_6[0] + q_6 = getitem_22 * 0.14433756729740643 + getitem_22 = None + k_6 = qkv_6[1] + v_6 = qkv_6[2] + qkv_6 = None + transpose_13 = k_6.transpose(-2, -1) + k_6 = None + attn_30 = q_6 @ transpose_13 + q_6 = transpose_13 = None + permute_31 = attn_30.permute(0, 2, 3, 1) + attn_30 = None + linear_37 = torch._C._nn.linear( + permute_31, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_31 = l_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_31 = linear_37.permute(0, 3, 1, 2) + linear_37 = None + attn_32 = attn_31.softmax(dim=-1) + attn_31 = None + permute_33 = attn_32.permute(0, 2, 3, 1) + attn_32 = None + linear_38 = torch._C._nn.linear( + permute_33, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_33 = l_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_33 = linear_38.permute(0, 3, 1, 2) + linear_38 = None + attn_34 = torch.nn.functional.dropout(attn_33, 0.0, False, False) + attn_33 = None + matmul_13 = attn_34 @ v_6 + attn_34 = v_6 = None + transpose_14 = matmul_13.transpose(1, 2) + matmul_13 = None + x_64 = transpose_14.reshape(1, 196, 192) + transpose_14 = None + x_65 = torch._C._nn.linear( + x_64, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_64 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + mul_19 = l_self_modules_blocks_modules_6_parameters_gamma_1_ * x_66 + l_self_modules_blocks_modules_6_parameters_gamma_1_ = x_66 = None + x_67 = x_63 + mul_19 + x_63 = mul_19 = None + layer_norm_13 = torch.nn.functional.layer_norm( + x_67, + (192,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_68 = torch._C._nn.linear( + layer_norm_13, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_13 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_69 = torch._C._nn.gelu(x_68, approximate="none") + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_71 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_70 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + mul_20 = l_self_modules_blocks_modules_6_parameters_gamma_2_ * x_72 + l_self_modules_blocks_modules_6_parameters_gamma_2_ = x_72 = None + x_73 = x_67 + mul_20 + x_67 = mul_20 = None + layer_norm_14 = torch.nn.functional.layer_norm( + x_73, + (192,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_42 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_14 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_42.reshape(1, 196, 3, 4, 48) + linear_42 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + getitem_25 = qkv_7[0] + q_7 = getitem_25 * 0.14433756729740643 + getitem_25 = None + k_7 = qkv_7[1] + v_7 = qkv_7[2] + qkv_7 = None + transpose_15 = k_7.transpose(-2, -1) + k_7 = None + attn_35 = q_7 @ transpose_15 + q_7 = transpose_15 = None + permute_36 = attn_35.permute(0, 2, 3, 1) + attn_35 = None + linear_43 = torch._C._nn.linear( + permute_36, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_36 = l_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_36 = linear_43.permute(0, 3, 1, 2) + linear_43 = None + attn_37 = attn_36.softmax(dim=-1) + attn_36 = None + permute_38 = attn_37.permute(0, 2, 3, 1) + attn_37 = None + linear_44 = torch._C._nn.linear( + permute_38, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_38 = l_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_38 = linear_44.permute(0, 3, 1, 2) + linear_44 = None + attn_39 = torch.nn.functional.dropout(attn_38, 0.0, False, False) + attn_38 = None + matmul_15 = attn_39 @ v_7 + attn_39 = v_7 = None + transpose_16 = matmul_15.transpose(1, 2) + matmul_15 = None + x_74 = transpose_16.reshape(1, 196, 192) + transpose_16 = None + x_75 = torch._C._nn.linear( + x_74, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_74 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_76 = torch.nn.functional.dropout(x_75, 0.0, False, False) + x_75 = None + mul_22 = l_self_modules_blocks_modules_7_parameters_gamma_1_ * x_76 + l_self_modules_blocks_modules_7_parameters_gamma_1_ = x_76 = None + x_77 = x_73 + mul_22 + x_73 = mul_22 = None + layer_norm_15 = torch.nn.functional.layer_norm( + x_77, + (192,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_78 = torch._C._nn.linear( + layer_norm_15, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_15 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_79 = torch._C._nn.gelu(x_78, approximate="none") + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = torch._C._nn.linear( + x_80, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_80 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_82 = torch.nn.functional.dropout(x_81, 0.0, False, False) + x_81 = None + mul_23 = l_self_modules_blocks_modules_7_parameters_gamma_2_ * x_82 + l_self_modules_blocks_modules_7_parameters_gamma_2_ = x_82 = None + x_83 = x_77 + mul_23 + x_77 = mul_23 = None + layer_norm_16 = torch.nn.functional.layer_norm( + x_83, + (192,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_48 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_16 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_48.reshape(1, 196, 3, 4, 48) + linear_48 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + getitem_28 = qkv_8[0] + q_8 = getitem_28 * 0.14433756729740643 + getitem_28 = None + k_8 = qkv_8[1] + v_8 = qkv_8[2] + qkv_8 = None + transpose_17 = k_8.transpose(-2, -1) + k_8 = None + attn_40 = q_8 @ transpose_17 + q_8 = transpose_17 = None + permute_41 = attn_40.permute(0, 2, 3, 1) + attn_40 = None + linear_49 = torch._C._nn.linear( + permute_41, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_41 = l_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_41 = linear_49.permute(0, 3, 1, 2) + linear_49 = None + attn_42 = attn_41.softmax(dim=-1) + attn_41 = None + permute_43 = attn_42.permute(0, 2, 3, 1) + attn_42 = None + linear_50 = torch._C._nn.linear( + permute_43, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_43 = l_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_43 = linear_50.permute(0, 3, 1, 2) + linear_50 = None + attn_44 = torch.nn.functional.dropout(attn_43, 0.0, False, False) + attn_43 = None + matmul_17 = attn_44 @ v_8 + attn_44 = v_8 = None + transpose_18 = matmul_17.transpose(1, 2) + matmul_17 = None + x_84 = transpose_18.reshape(1, 196, 192) + transpose_18 = None + x_85 = torch._C._nn.linear( + x_84, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_84 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_86 = torch.nn.functional.dropout(x_85, 0.0, False, False) + x_85 = None + mul_25 = l_self_modules_blocks_modules_8_parameters_gamma_1_ * x_86 + l_self_modules_blocks_modules_8_parameters_gamma_1_ = x_86 = None + x_87 = x_83 + mul_25 + x_83 = mul_25 = None + layer_norm_17 = torch.nn.functional.layer_norm( + x_87, + (192,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_88 = torch._C._nn.linear( + layer_norm_17, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_17 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_89 = torch._C._nn.gelu(x_88, approximate="none") + x_88 = None + x_90 = torch.nn.functional.dropout(x_89, 0.0, False, False) + x_89 = None + x_91 = torch._C._nn.linear( + x_90, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_90 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + mul_26 = l_self_modules_blocks_modules_8_parameters_gamma_2_ * x_92 + l_self_modules_blocks_modules_8_parameters_gamma_2_ = x_92 = None + x_93 = x_87 + mul_26 + x_87 = mul_26 = None + layer_norm_18 = torch.nn.functional.layer_norm( + x_93, + (192,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_54 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_18 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_54.reshape(1, 196, 3, 4, 48) + linear_54 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + getitem_31 = qkv_9[0] + q_9 = getitem_31 * 0.14433756729740643 + getitem_31 = None + k_9 = qkv_9[1] + v_9 = qkv_9[2] + qkv_9 = None + transpose_19 = k_9.transpose(-2, -1) + k_9 = None + attn_45 = q_9 @ transpose_19 + q_9 = transpose_19 = None + permute_46 = attn_45.permute(0, 2, 3, 1) + attn_45 = None + linear_55 = torch._C._nn.linear( + permute_46, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_46 = l_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_46 = linear_55.permute(0, 3, 1, 2) + linear_55 = None + attn_47 = attn_46.softmax(dim=-1) + attn_46 = None + permute_48 = attn_47.permute(0, 2, 3, 1) + attn_47 = None + linear_56 = torch._C._nn.linear( + permute_48, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_48 = l_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_48 = linear_56.permute(0, 3, 1, 2) + linear_56 = None + attn_49 = torch.nn.functional.dropout(attn_48, 0.0, False, False) + attn_48 = None + matmul_19 = attn_49 @ v_9 + attn_49 = v_9 = None + transpose_20 = matmul_19.transpose(1, 2) + matmul_19 = None + x_94 = transpose_20.reshape(1, 196, 192) + transpose_20 = None + x_95 = torch._C._nn.linear( + x_94, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_94 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_96 = torch.nn.functional.dropout(x_95, 0.0, False, False) + x_95 = None + mul_28 = l_self_modules_blocks_modules_9_parameters_gamma_1_ * x_96 + l_self_modules_blocks_modules_9_parameters_gamma_1_ = x_96 = None + x_97 = x_93 + mul_28 + x_93 = mul_28 = None + layer_norm_19 = torch.nn.functional.layer_norm( + x_97, + (192,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_98 = torch._C._nn.linear( + layer_norm_19, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_19 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_99 = torch._C._nn.gelu(x_98, approximate="none") + x_98 = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + x_101 = torch._C._nn.linear( + x_100, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_100 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + mul_29 = l_self_modules_blocks_modules_9_parameters_gamma_2_ * x_102 + l_self_modules_blocks_modules_9_parameters_gamma_2_ = x_102 = None + x_103 = x_97 + mul_29 + x_97 = mul_29 = None + layer_norm_20 = torch.nn.functional.layer_norm( + x_103, + (192,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_60 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_20 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_60.reshape(1, 196, 3, 4, 48) + linear_60 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + getitem_34 = qkv_10[0] + q_10 = getitem_34 * 0.14433756729740643 + getitem_34 = None + k_10 = qkv_10[1] + v_10 = qkv_10[2] + qkv_10 = None + transpose_21 = k_10.transpose(-2, -1) + k_10 = None + attn_50 = q_10 @ transpose_21 + q_10 = transpose_21 = None + permute_51 = attn_50.permute(0, 2, 3, 1) + attn_50 = None + linear_61 = torch._C._nn.linear( + permute_51, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_51 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_51 = linear_61.permute(0, 3, 1, 2) + linear_61 = None + attn_52 = attn_51.softmax(dim=-1) + attn_51 = None + permute_53 = attn_52.permute(0, 2, 3, 1) + attn_52 = None + linear_62 = torch._C._nn.linear( + permute_53, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_53 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_53 = linear_62.permute(0, 3, 1, 2) + linear_62 = None + attn_54 = torch.nn.functional.dropout(attn_53, 0.0, False, False) + attn_53 = None + matmul_21 = attn_54 @ v_10 + attn_54 = v_10 = None + transpose_22 = matmul_21.transpose(1, 2) + matmul_21 = None + x_104 = transpose_22.reshape(1, 196, 192) + transpose_22 = None + x_105 = torch._C._nn.linear( + x_104, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_104 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + mul_31 = l_self_modules_blocks_modules_10_parameters_gamma_1_ * x_106 + l_self_modules_blocks_modules_10_parameters_gamma_1_ = x_106 = None + x_107 = x_103 + mul_31 + x_103 = mul_31 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_107, + (192,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_108 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_21 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_109 = torch._C._nn.gelu(x_108, approximate="none") + x_108 = None + x_110 = torch.nn.functional.dropout(x_109, 0.0, False, False) + x_109 = None + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_110 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_112 = torch.nn.functional.dropout(x_111, 0.0, False, False) + x_111 = None + mul_32 = l_self_modules_blocks_modules_10_parameters_gamma_2_ * x_112 + l_self_modules_blocks_modules_10_parameters_gamma_2_ = x_112 = None + x_113 = x_107 + mul_32 + x_107 = mul_32 = None + layer_norm_22 = torch.nn.functional.layer_norm( + x_113, + (192,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_66 = torch._C._nn.linear( + layer_norm_22, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_22 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_66.reshape(1, 196, 3, 4, 48) + linear_66 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + getitem_37 = qkv_11[0] + q_11 = getitem_37 * 0.14433756729740643 + getitem_37 = None + k_11 = qkv_11[1] + v_11 = qkv_11[2] + qkv_11 = None + transpose_23 = k_11.transpose(-2, -1) + k_11 = None + attn_55 = q_11 @ transpose_23 + q_11 = transpose_23 = None + permute_56 = attn_55.permute(0, 2, 3, 1) + attn_55 = None + linear_67 = torch._C._nn.linear( + permute_56, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_56 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_56 = linear_67.permute(0, 3, 1, 2) + linear_67 = None + attn_57 = attn_56.softmax(dim=-1) + attn_56 = None + permute_58 = attn_57.permute(0, 2, 3, 1) + attn_57 = None + linear_68 = torch._C._nn.linear( + permute_58, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_58 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_58 = linear_68.permute(0, 3, 1, 2) + linear_68 = None + attn_59 = torch.nn.functional.dropout(attn_58, 0.0, False, False) + attn_58 = None + matmul_23 = attn_59 @ v_11 + attn_59 = v_11 = None + transpose_24 = matmul_23.transpose(1, 2) + matmul_23 = None + x_114 = transpose_24.reshape(1, 196, 192) + transpose_24 = None + x_115 = torch._C._nn.linear( + x_114, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_114 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + mul_34 = l_self_modules_blocks_modules_11_parameters_gamma_1_ * x_116 + l_self_modules_blocks_modules_11_parameters_gamma_1_ = x_116 = None + x_117 = x_113 + mul_34 + x_113 = mul_34 = None + layer_norm_23 = torch.nn.functional.layer_norm( + x_117, + (192,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_118 = torch._C._nn.linear( + layer_norm_23, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_23 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_119 = torch._C._nn.gelu(x_118, approximate="none") + x_118 = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + x_121 = torch._C._nn.linear( + x_120, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_120 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_122 = torch.nn.functional.dropout(x_121, 0.0, False, False) + x_121 = None + mul_35 = l_self_modules_blocks_modules_11_parameters_gamma_2_ * x_122 + l_self_modules_blocks_modules_11_parameters_gamma_2_ = x_122 = None + x_123 = x_117 + mul_35 + x_117 = mul_35 = None + layer_norm_24 = torch.nn.functional.layer_norm( + x_123, + (192,), + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ + ) = None + linear_72 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_24 = ( + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_24 = linear_72.reshape(1, 196, 3, 4, 48) + linear_72 = None + qkv_12 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + getitem_40 = qkv_12[0] + q_12 = getitem_40 * 0.14433756729740643 + getitem_40 = None + k_12 = qkv_12[1] + v_12 = qkv_12[2] + qkv_12 = None + transpose_25 = k_12.transpose(-2, -1) + k_12 = None + attn_60 = q_12 @ transpose_25 + q_12 = transpose_25 = None + permute_61 = attn_60.permute(0, 2, 3, 1) + attn_60 = None + linear_73 = torch._C._nn.linear( + permute_61, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_61 = l_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_61 = linear_73.permute(0, 3, 1, 2) + linear_73 = None + attn_62 = attn_61.softmax(dim=-1) + attn_61 = None + permute_63 = attn_62.permute(0, 2, 3, 1) + attn_62 = None + linear_74 = torch._C._nn.linear( + permute_63, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_63 = l_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_63 = linear_74.permute(0, 3, 1, 2) + linear_74 = None + attn_64 = torch.nn.functional.dropout(attn_63, 0.0, False, False) + attn_63 = None + matmul_25 = attn_64 @ v_12 + attn_64 = v_12 = None + transpose_26 = matmul_25.transpose(1, 2) + matmul_25 = None + x_124 = transpose_26.reshape(1, 196, 192) + transpose_26 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_, + ) + x_124 = l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ + ) = None + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + mul_37 = l_self_modules_blocks_modules_12_parameters_gamma_1_ * x_126 + l_self_modules_blocks_modules_12_parameters_gamma_1_ = x_126 = None + x_127 = x_123 + mul_37 + x_123 = mul_37 = None + layer_norm_25 = torch.nn.functional.layer_norm( + x_127, + (192,), + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ + ) = None + x_128 = torch._C._nn.linear( + layer_norm_25, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_25 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_129 = torch._C._nn.gelu(x_128, approximate="none") + x_128 = None + x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) + x_129 = None + x_131 = torch._C._nn.linear( + x_130, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_130 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_132 = torch.nn.functional.dropout(x_131, 0.0, False, False) + x_131 = None + mul_38 = l_self_modules_blocks_modules_12_parameters_gamma_2_ * x_132 + l_self_modules_blocks_modules_12_parameters_gamma_2_ = x_132 = None + x_133 = x_127 + mul_38 + x_127 = mul_38 = None + layer_norm_26 = torch.nn.functional.layer_norm( + x_133, + (192,), + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ + ) = None + linear_78 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_26 = ( + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_26 = linear_78.reshape(1, 196, 3, 4, 48) + linear_78 = None + qkv_13 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + getitem_43 = qkv_13[0] + q_13 = getitem_43 * 0.14433756729740643 + getitem_43 = None + k_13 = qkv_13[1] + v_13 = qkv_13[2] + qkv_13 = None + transpose_27 = k_13.transpose(-2, -1) + k_13 = None + attn_65 = q_13 @ transpose_27 + q_13 = transpose_27 = None + permute_66 = attn_65.permute(0, 2, 3, 1) + attn_65 = None + linear_79 = torch._C._nn.linear( + permute_66, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_66 = l_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_66 = linear_79.permute(0, 3, 1, 2) + linear_79 = None + attn_67 = attn_66.softmax(dim=-1) + attn_66 = None + permute_68 = attn_67.permute(0, 2, 3, 1) + attn_67 = None + linear_80 = torch._C._nn.linear( + permute_68, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_68 = l_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_68 = linear_80.permute(0, 3, 1, 2) + linear_80 = None + attn_69 = torch.nn.functional.dropout(attn_68, 0.0, False, False) + attn_68 = None + matmul_27 = attn_69 @ v_13 + attn_69 = v_13 = None + transpose_28 = matmul_27.transpose(1, 2) + matmul_27 = None + x_134 = transpose_28.reshape(1, 196, 192) + transpose_28 = None + x_135 = torch._C._nn.linear( + x_134, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_, + ) + x_134 = l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ + ) = None + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + mul_40 = l_self_modules_blocks_modules_13_parameters_gamma_1_ * x_136 + l_self_modules_blocks_modules_13_parameters_gamma_1_ = x_136 = None + x_137 = x_133 + mul_40 + x_133 = mul_40 = None + layer_norm_27 = torch.nn.functional.layer_norm( + x_137, + (192,), + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ + ) = None + x_138 = torch._C._nn.linear( + layer_norm_27, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_27 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_139 = torch._C._nn.gelu(x_138, approximate="none") + x_138 = None + x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) + x_139 = None + x_141 = torch._C._nn.linear( + x_140, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_140 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + mul_41 = l_self_modules_blocks_modules_13_parameters_gamma_2_ * x_142 + l_self_modules_blocks_modules_13_parameters_gamma_2_ = x_142 = None + x_143 = x_137 + mul_41 + x_137 = mul_41 = None + layer_norm_28 = torch.nn.functional.layer_norm( + x_143, + (192,), + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ + ) = None + linear_84 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_28 = ( + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_28 = linear_84.reshape(1, 196, 3, 4, 48) + linear_84 = None + qkv_14 = reshape_28.permute(2, 0, 3, 1, 4) + reshape_28 = None + getitem_46 = qkv_14[0] + q_14 = getitem_46 * 0.14433756729740643 + getitem_46 = None + k_14 = qkv_14[1] + v_14 = qkv_14[2] + qkv_14 = None + transpose_29 = k_14.transpose(-2, -1) + k_14 = None + attn_70 = q_14 @ transpose_29 + q_14 = transpose_29 = None + permute_71 = attn_70.permute(0, 2, 3, 1) + attn_70 = None + linear_85 = torch._C._nn.linear( + permute_71, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_71 = l_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_71 = linear_85.permute(0, 3, 1, 2) + linear_85 = None + attn_72 = attn_71.softmax(dim=-1) + attn_71 = None + permute_73 = attn_72.permute(0, 2, 3, 1) + attn_72 = None + linear_86 = torch._C._nn.linear( + permute_73, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_73 = l_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_73 = linear_86.permute(0, 3, 1, 2) + linear_86 = None + attn_74 = torch.nn.functional.dropout(attn_73, 0.0, False, False) + attn_73 = None + matmul_29 = attn_74 @ v_14 + attn_74 = v_14 = None + transpose_30 = matmul_29.transpose(1, 2) + matmul_29 = None + x_144 = transpose_30.reshape(1, 196, 192) + transpose_30 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_, + ) + x_144 = l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ + ) = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + mul_43 = l_self_modules_blocks_modules_14_parameters_gamma_1_ * x_146 + l_self_modules_blocks_modules_14_parameters_gamma_1_ = x_146 = None + x_147 = x_143 + mul_43 + x_143 = mul_43 = None + layer_norm_29 = torch.nn.functional.layer_norm( + x_147, + (192,), + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ + ) = None + x_148 = torch._C._nn.linear( + layer_norm_29, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_29 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_149 = torch._C._nn.gelu(x_148, approximate="none") + x_148 = None + x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) + x_149 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_150 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + mul_44 = l_self_modules_blocks_modules_14_parameters_gamma_2_ * x_152 + l_self_modules_blocks_modules_14_parameters_gamma_2_ = x_152 = None + x_153 = x_147 + mul_44 + x_147 = mul_44 = None + layer_norm_30 = torch.nn.functional.layer_norm( + x_153, + (192,), + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ + ) = None + linear_90 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_30 = ( + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_30 = linear_90.reshape(1, 196, 3, 4, 48) + linear_90 = None + qkv_15 = reshape_30.permute(2, 0, 3, 1, 4) + reshape_30 = None + getitem_49 = qkv_15[0] + q_15 = getitem_49 * 0.14433756729740643 + getitem_49 = None + k_15 = qkv_15[1] + v_15 = qkv_15[2] + qkv_15 = None + transpose_31 = k_15.transpose(-2, -1) + k_15 = None + attn_75 = q_15 @ transpose_31 + q_15 = transpose_31 = None + permute_76 = attn_75.permute(0, 2, 3, 1) + attn_75 = None + linear_91 = torch._C._nn.linear( + permute_76, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_76 = l_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_76 = linear_91.permute(0, 3, 1, 2) + linear_91 = None + attn_77 = attn_76.softmax(dim=-1) + attn_76 = None + permute_78 = attn_77.permute(0, 2, 3, 1) + attn_77 = None + linear_92 = torch._C._nn.linear( + permute_78, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_78 = l_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_78 = linear_92.permute(0, 3, 1, 2) + linear_92 = None + attn_79 = torch.nn.functional.dropout(attn_78, 0.0, False, False) + attn_78 = None + matmul_31 = attn_79 @ v_15 + attn_79 = v_15 = None + transpose_32 = matmul_31.transpose(1, 2) + matmul_31 = None + x_154 = transpose_32.reshape(1, 196, 192) + transpose_32 = None + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_, + ) + x_154 = l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ + ) = None + x_156 = torch.nn.functional.dropout(x_155, 0.0, False, False) + x_155 = None + mul_46 = l_self_modules_blocks_modules_15_parameters_gamma_1_ * x_156 + l_self_modules_blocks_modules_15_parameters_gamma_1_ = x_156 = None + x_157 = x_153 + mul_46 + x_153 = mul_46 = None + layer_norm_31 = torch.nn.functional.layer_norm( + x_157, + (192,), + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ + ) = None + x_158 = torch._C._nn.linear( + layer_norm_31, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_31 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_159 = torch._C._nn.gelu(x_158, approximate="none") + x_158 = None + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = torch._C._nn.linear( + x_160, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_160 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_162 = torch.nn.functional.dropout(x_161, 0.0, False, False) + x_161 = None + mul_47 = l_self_modules_blocks_modules_15_parameters_gamma_2_ * x_162 + l_self_modules_blocks_modules_15_parameters_gamma_2_ = x_162 = None + x_163 = x_157 + mul_47 + x_157 = mul_47 = None + layer_norm_32 = torch.nn.functional.layer_norm( + x_163, + (192,), + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ + ) = None + linear_96 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_32 = ( + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_32 = linear_96.reshape(1, 196, 3, 4, 48) + linear_96 = None + qkv_16 = reshape_32.permute(2, 0, 3, 1, 4) + reshape_32 = None + getitem_52 = qkv_16[0] + q_16 = getitem_52 * 0.14433756729740643 + getitem_52 = None + k_16 = qkv_16[1] + v_16 = qkv_16[2] + qkv_16 = None + transpose_33 = k_16.transpose(-2, -1) + k_16 = None + attn_80 = q_16 @ transpose_33 + q_16 = transpose_33 = None + permute_81 = attn_80.permute(0, 2, 3, 1) + attn_80 = None + linear_97 = torch._C._nn.linear( + permute_81, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_81 = l_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_81 = linear_97.permute(0, 3, 1, 2) + linear_97 = None + attn_82 = attn_81.softmax(dim=-1) + attn_81 = None + permute_83 = attn_82.permute(0, 2, 3, 1) + attn_82 = None + linear_98 = torch._C._nn.linear( + permute_83, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_83 = l_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_83 = linear_98.permute(0, 3, 1, 2) + linear_98 = None + attn_84 = torch.nn.functional.dropout(attn_83, 0.0, False, False) + attn_83 = None + matmul_33 = attn_84 @ v_16 + attn_84 = v_16 = None + transpose_34 = matmul_33.transpose(1, 2) + matmul_33 = None + x_164 = transpose_34.reshape(1, 196, 192) + transpose_34 = None + x_165 = torch._C._nn.linear( + x_164, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_, + ) + x_164 = l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ + ) = None + x_166 = torch.nn.functional.dropout(x_165, 0.0, False, False) + x_165 = None + mul_49 = l_self_modules_blocks_modules_16_parameters_gamma_1_ * x_166 + l_self_modules_blocks_modules_16_parameters_gamma_1_ = x_166 = None + x_167 = x_163 + mul_49 + x_163 = mul_49 = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_167, + (192,), + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ + ) = None + x_168 = torch._C._nn.linear( + layer_norm_33, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_33 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_169 = torch._C._nn.gelu(x_168, approximate="none") + x_168 = None + x_170 = torch.nn.functional.dropout(x_169, 0.0, False, False) + x_169 = None + x_171 = torch._C._nn.linear( + x_170, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_170 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_172 = torch.nn.functional.dropout(x_171, 0.0, False, False) + x_171 = None + mul_50 = l_self_modules_blocks_modules_16_parameters_gamma_2_ * x_172 + l_self_modules_blocks_modules_16_parameters_gamma_2_ = x_172 = None + x_173 = x_167 + mul_50 + x_167 = mul_50 = None + layer_norm_34 = torch.nn.functional.layer_norm( + x_173, + (192,), + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ + ) = None + linear_102 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_34 = ( + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_34 = linear_102.reshape(1, 196, 3, 4, 48) + linear_102 = None + qkv_17 = reshape_34.permute(2, 0, 3, 1, 4) + reshape_34 = None + getitem_55 = qkv_17[0] + q_17 = getitem_55 * 0.14433756729740643 + getitem_55 = None + k_17 = qkv_17[1] + v_17 = qkv_17[2] + qkv_17 = None + transpose_35 = k_17.transpose(-2, -1) + k_17 = None + attn_85 = q_17 @ transpose_35 + q_17 = transpose_35 = None + permute_86 = attn_85.permute(0, 2, 3, 1) + attn_85 = None + linear_103 = torch._C._nn.linear( + permute_86, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_86 = l_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_86 = linear_103.permute(0, 3, 1, 2) + linear_103 = None + attn_87 = attn_86.softmax(dim=-1) + attn_86 = None + permute_88 = attn_87.permute(0, 2, 3, 1) + attn_87 = None + linear_104 = torch._C._nn.linear( + permute_88, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_88 = l_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_88 = linear_104.permute(0, 3, 1, 2) + linear_104 = None + attn_89 = torch.nn.functional.dropout(attn_88, 0.0, False, False) + attn_88 = None + matmul_35 = attn_89 @ v_17 + attn_89 = v_17 = None + transpose_36 = matmul_35.transpose(1, 2) + matmul_35 = None + x_174 = transpose_36.reshape(1, 196, 192) + transpose_36 = None + x_175 = torch._C._nn.linear( + x_174, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_, + ) + x_174 = l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ + ) = None + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + mul_52 = l_self_modules_blocks_modules_17_parameters_gamma_1_ * x_176 + l_self_modules_blocks_modules_17_parameters_gamma_1_ = x_176 = None + x_177 = x_173 + mul_52 + x_173 = mul_52 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_177, + (192,), + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ + ) = None + x_178 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_35 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_179 = torch._C._nn.gelu(x_178, approximate="none") + x_178 = None + x_180 = torch.nn.functional.dropout(x_179, 0.0, False, False) + x_179 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + mul_53 = l_self_modules_blocks_modules_17_parameters_gamma_2_ * x_182 + l_self_modules_blocks_modules_17_parameters_gamma_2_ = x_182 = None + x_183 = x_177 + mul_53 + x_177 = mul_53 = None + layer_norm_36 = torch.nn.functional.layer_norm( + x_183, + (192,), + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ + ) = None + linear_108 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_36 = ( + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_36 = linear_108.reshape(1, 196, 3, 4, 48) + linear_108 = None + qkv_18 = reshape_36.permute(2, 0, 3, 1, 4) + reshape_36 = None + getitem_58 = qkv_18[0] + q_18 = getitem_58 * 0.14433756729740643 + getitem_58 = None + k_18 = qkv_18[1] + v_18 = qkv_18[2] + qkv_18 = None + transpose_37 = k_18.transpose(-2, -1) + k_18 = None + attn_90 = q_18 @ transpose_37 + q_18 = transpose_37 = None + permute_91 = attn_90.permute(0, 2, 3, 1) + attn_90 = None + linear_109 = torch._C._nn.linear( + permute_91, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_91 = l_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_91 = linear_109.permute(0, 3, 1, 2) + linear_109 = None + attn_92 = attn_91.softmax(dim=-1) + attn_91 = None + permute_93 = attn_92.permute(0, 2, 3, 1) + attn_92 = None + linear_110 = torch._C._nn.linear( + permute_93, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_93 = l_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_93 = linear_110.permute(0, 3, 1, 2) + linear_110 = None + attn_94 = torch.nn.functional.dropout(attn_93, 0.0, False, False) + attn_93 = None + matmul_37 = attn_94 @ v_18 + attn_94 = v_18 = None + transpose_38 = matmul_37.transpose(1, 2) + matmul_37 = None + x_184 = transpose_38.reshape(1, 196, 192) + transpose_38 = None + x_185 = torch._C._nn.linear( + x_184, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_, + ) + x_184 = l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ + ) = None + x_186 = torch.nn.functional.dropout(x_185, 0.0, False, False) + x_185 = None + mul_55 = l_self_modules_blocks_modules_18_parameters_gamma_1_ * x_186 + l_self_modules_blocks_modules_18_parameters_gamma_1_ = x_186 = None + x_187 = x_183 + mul_55 + x_183 = mul_55 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_187, + (192,), + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ + ) = None + x_188 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_37 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_189 = torch._C._nn.gelu(x_188, approximate="none") + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = torch._C._nn.linear( + x_190, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_190 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_192 = torch.nn.functional.dropout(x_191, 0.0, False, False) + x_191 = None + mul_56 = l_self_modules_blocks_modules_18_parameters_gamma_2_ * x_192 + l_self_modules_blocks_modules_18_parameters_gamma_2_ = x_192 = None + x_193 = x_187 + mul_56 + x_187 = mul_56 = None + layer_norm_38 = torch.nn.functional.layer_norm( + x_193, + (192,), + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ + ) = None + linear_114 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_38 = ( + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_38 = linear_114.reshape(1, 196, 3, 4, 48) + linear_114 = None + qkv_19 = reshape_38.permute(2, 0, 3, 1, 4) + reshape_38 = None + getitem_61 = qkv_19[0] + q_19 = getitem_61 * 0.14433756729740643 + getitem_61 = None + k_19 = qkv_19[1] + v_19 = qkv_19[2] + qkv_19 = None + transpose_39 = k_19.transpose(-2, -1) + k_19 = None + attn_95 = q_19 @ transpose_39 + q_19 = transpose_39 = None + permute_96 = attn_95.permute(0, 2, 3, 1) + attn_95 = None + linear_115 = torch._C._nn.linear( + permute_96, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_96 = l_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_96 = linear_115.permute(0, 3, 1, 2) + linear_115 = None + attn_97 = attn_96.softmax(dim=-1) + attn_96 = None + permute_98 = attn_97.permute(0, 2, 3, 1) + attn_97 = None + linear_116 = torch._C._nn.linear( + permute_98, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_98 = l_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_98 = linear_116.permute(0, 3, 1, 2) + linear_116 = None + attn_99 = torch.nn.functional.dropout(attn_98, 0.0, False, False) + attn_98 = None + matmul_39 = attn_99 @ v_19 + attn_99 = v_19 = None + transpose_40 = matmul_39.transpose(1, 2) + matmul_39 = None + x_194 = transpose_40.reshape(1, 196, 192) + transpose_40 = None + x_195 = torch._C._nn.linear( + x_194, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_, + ) + x_194 = l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ + ) = None + x_196 = torch.nn.functional.dropout(x_195, 0.0, False, False) + x_195 = None + mul_58 = l_self_modules_blocks_modules_19_parameters_gamma_1_ * x_196 + l_self_modules_blocks_modules_19_parameters_gamma_1_ = x_196 = None + x_197 = x_193 + mul_58 + x_193 = mul_58 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_197, + (192,), + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ + ) = None + x_198 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_39 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_199 = torch._C._nn.gelu(x_198, approximate="none") + x_198 = None + x_200 = torch.nn.functional.dropout(x_199, 0.0, False, False) + x_199 = None + x_201 = torch._C._nn.linear( + x_200, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_200 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_202 = torch.nn.functional.dropout(x_201, 0.0, False, False) + x_201 = None + mul_59 = l_self_modules_blocks_modules_19_parameters_gamma_2_ * x_202 + l_self_modules_blocks_modules_19_parameters_gamma_2_ = x_202 = None + x_203 = x_197 + mul_59 + x_197 = mul_59 = None + layer_norm_40 = torch.nn.functional.layer_norm( + x_203, + (192,), + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ + ) = None + linear_120 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_40 = ( + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_40 = linear_120.reshape(1, 196, 3, 4, 48) + linear_120 = None + qkv_20 = reshape_40.permute(2, 0, 3, 1, 4) + reshape_40 = None + getitem_64 = qkv_20[0] + q_20 = getitem_64 * 0.14433756729740643 + getitem_64 = None + k_20 = qkv_20[1] + v_20 = qkv_20[2] + qkv_20 = None + transpose_41 = k_20.transpose(-2, -1) + k_20 = None + attn_100 = q_20 @ transpose_41 + q_20 = transpose_41 = None + permute_101 = attn_100.permute(0, 2, 3, 1) + attn_100 = None + linear_121 = torch._C._nn.linear( + permute_101, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_101 = l_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_101 = linear_121.permute(0, 3, 1, 2) + linear_121 = None + attn_102 = attn_101.softmax(dim=-1) + attn_101 = None + permute_103 = attn_102.permute(0, 2, 3, 1) + attn_102 = None + linear_122 = torch._C._nn.linear( + permute_103, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_103 = l_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_103 = linear_122.permute(0, 3, 1, 2) + linear_122 = None + attn_104 = torch.nn.functional.dropout(attn_103, 0.0, False, False) + attn_103 = None + matmul_41 = attn_104 @ v_20 + attn_104 = v_20 = None + transpose_42 = matmul_41.transpose(1, 2) + matmul_41 = None + x_204 = transpose_42.reshape(1, 196, 192) + transpose_42 = None + x_205 = torch._C._nn.linear( + x_204, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_, + ) + x_204 = l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ + ) = None + x_206 = torch.nn.functional.dropout(x_205, 0.0, False, False) + x_205 = None + mul_61 = l_self_modules_blocks_modules_20_parameters_gamma_1_ * x_206 + l_self_modules_blocks_modules_20_parameters_gamma_1_ = x_206 = None + x_207 = x_203 + mul_61 + x_203 = mul_61 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_207, + (192,), + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ + ) = None + x_208 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_41 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_209 = torch._C._nn.gelu(x_208, approximate="none") + x_208 = None + x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) + x_209 = None + x_211 = torch._C._nn.linear( + x_210, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_210 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + mul_62 = l_self_modules_blocks_modules_20_parameters_gamma_2_ * x_212 + l_self_modules_blocks_modules_20_parameters_gamma_2_ = x_212 = None + x_213 = x_207 + mul_62 + x_207 = mul_62 = None + layer_norm_42 = torch.nn.functional.layer_norm( + x_213, + (192,), + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ + ) = None + linear_126 = torch._C._nn.linear( + layer_norm_42, + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_42 = ( + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_42 = linear_126.reshape(1, 196, 3, 4, 48) + linear_126 = None + qkv_21 = reshape_42.permute(2, 0, 3, 1, 4) + reshape_42 = None + getitem_67 = qkv_21[0] + q_21 = getitem_67 * 0.14433756729740643 + getitem_67 = None + k_21 = qkv_21[1] + v_21 = qkv_21[2] + qkv_21 = None + transpose_43 = k_21.transpose(-2, -1) + k_21 = None + attn_105 = q_21 @ transpose_43 + q_21 = transpose_43 = None + permute_106 = attn_105.permute(0, 2, 3, 1) + attn_105 = None + linear_127 = torch._C._nn.linear( + permute_106, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_106 = l_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_106 = linear_127.permute(0, 3, 1, 2) + linear_127 = None + attn_107 = attn_106.softmax(dim=-1) + attn_106 = None + permute_108 = attn_107.permute(0, 2, 3, 1) + attn_107 = None + linear_128 = torch._C._nn.linear( + permute_108, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_108 = l_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_108 = linear_128.permute(0, 3, 1, 2) + linear_128 = None + attn_109 = torch.nn.functional.dropout(attn_108, 0.0, False, False) + attn_108 = None + matmul_43 = attn_109 @ v_21 + attn_109 = v_21 = None + transpose_44 = matmul_43.transpose(1, 2) + matmul_43 = None + x_214 = transpose_44.reshape(1, 196, 192) + transpose_44 = None + x_215 = torch._C._nn.linear( + x_214, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_, + ) + x_214 = l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ + ) = None + x_216 = torch.nn.functional.dropout(x_215, 0.0, False, False) + x_215 = None + mul_64 = l_self_modules_blocks_modules_21_parameters_gamma_1_ * x_216 + l_self_modules_blocks_modules_21_parameters_gamma_1_ = x_216 = None + x_217 = x_213 + mul_64 + x_213 = mul_64 = None + layer_norm_43 = torch.nn.functional.layer_norm( + x_217, + (192,), + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ + ) = None + x_218 = torch._C._nn.linear( + layer_norm_43, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_43 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_219 = torch._C._nn.gelu(x_218, approximate="none") + x_218 = None + x_220 = torch.nn.functional.dropout(x_219, 0.0, False, False) + x_219 = None + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_220 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_222 = torch.nn.functional.dropout(x_221, 0.0, False, False) + x_221 = None + mul_65 = l_self_modules_blocks_modules_21_parameters_gamma_2_ * x_222 + l_self_modules_blocks_modules_21_parameters_gamma_2_ = x_222 = None + x_223 = x_217 + mul_65 + x_217 = mul_65 = None + layer_norm_44 = torch.nn.functional.layer_norm( + x_223, + (192,), + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ + ) = None + linear_132 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_44 = ( + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_44 = linear_132.reshape(1, 196, 3, 4, 48) + linear_132 = None + qkv_22 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + getitem_70 = qkv_22[0] + q_22 = getitem_70 * 0.14433756729740643 + getitem_70 = None + k_22 = qkv_22[1] + v_22 = qkv_22[2] + qkv_22 = None + transpose_45 = k_22.transpose(-2, -1) + k_22 = None + attn_110 = q_22 @ transpose_45 + q_22 = transpose_45 = None + permute_111 = attn_110.permute(0, 2, 3, 1) + attn_110 = None + linear_133 = torch._C._nn.linear( + permute_111, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_111 = l_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_111 = linear_133.permute(0, 3, 1, 2) + linear_133 = None + attn_112 = attn_111.softmax(dim=-1) + attn_111 = None + permute_113 = attn_112.permute(0, 2, 3, 1) + attn_112 = None + linear_134 = torch._C._nn.linear( + permute_113, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_113 = l_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_113 = linear_134.permute(0, 3, 1, 2) + linear_134 = None + attn_114 = torch.nn.functional.dropout(attn_113, 0.0, False, False) + attn_113 = None + matmul_45 = attn_114 @ v_22 + attn_114 = v_22 = None + transpose_46 = matmul_45.transpose(1, 2) + matmul_45 = None + x_224 = transpose_46.reshape(1, 196, 192) + transpose_46 = None + x_225 = torch._C._nn.linear( + x_224, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_, + ) + x_224 = l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ + ) = None + x_226 = torch.nn.functional.dropout(x_225, 0.0, False, False) + x_225 = None + mul_67 = l_self_modules_blocks_modules_22_parameters_gamma_1_ * x_226 + l_self_modules_blocks_modules_22_parameters_gamma_1_ = x_226 = None + x_227 = x_223 + mul_67 + x_223 = mul_67 = None + layer_norm_45 = torch.nn.functional.layer_norm( + x_227, + (192,), + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ + ) = None + x_228 = torch._C._nn.linear( + layer_norm_45, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_45 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_229 = torch._C._nn.gelu(x_228, approximate="none") + x_228 = None + x_230 = torch.nn.functional.dropout(x_229, 0.0, False, False) + x_229 = None + x_231 = torch._C._nn.linear( + x_230, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_230 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_232 = torch.nn.functional.dropout(x_231, 0.0, False, False) + x_231 = None + mul_68 = l_self_modules_blocks_modules_22_parameters_gamma_2_ * x_232 + l_self_modules_blocks_modules_22_parameters_gamma_2_ = x_232 = None + x_233 = x_227 + mul_68 + x_227 = mul_68 = None + layer_norm_46 = torch.nn.functional.layer_norm( + x_233, + (192,), + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ + ) = None + linear_138 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_46 = ( + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_46 = linear_138.reshape(1, 196, 3, 4, 48) + linear_138 = None + qkv_23 = reshape_46.permute(2, 0, 3, 1, 4) + reshape_46 = None + getitem_73 = qkv_23[0] + q_23 = getitem_73 * 0.14433756729740643 + getitem_73 = None + k_23 = qkv_23[1] + v_23 = qkv_23[2] + qkv_23 = None + transpose_47 = k_23.transpose(-2, -1) + k_23 = None + attn_115 = q_23 @ transpose_47 + q_23 = transpose_47 = None + permute_116 = attn_115.permute(0, 2, 3, 1) + attn_115 = None + linear_139 = torch._C._nn.linear( + permute_116, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_116 = l_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_116 = linear_139.permute(0, 3, 1, 2) + linear_139 = None + attn_117 = attn_116.softmax(dim=-1) + attn_116 = None + permute_118 = attn_117.permute(0, 2, 3, 1) + attn_117 = None + linear_140 = torch._C._nn.linear( + permute_118, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_118 = l_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_118 = linear_140.permute(0, 3, 1, 2) + linear_140 = None + attn_119 = torch.nn.functional.dropout(attn_118, 0.0, False, False) + attn_118 = None + matmul_47 = attn_119 @ v_23 + attn_119 = v_23 = None + transpose_48 = matmul_47.transpose(1, 2) + matmul_47 = None + x_234 = transpose_48.reshape(1, 196, 192) + transpose_48 = None + x_235 = torch._C._nn.linear( + x_234, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_, + ) + x_234 = l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ + ) = None + x_236 = torch.nn.functional.dropout(x_235, 0.0, False, False) + x_235 = None + mul_70 = l_self_modules_blocks_modules_23_parameters_gamma_1_ * x_236 + l_self_modules_blocks_modules_23_parameters_gamma_1_ = x_236 = None + x_237 = x_233 + mul_70 + x_233 = mul_70 = None + layer_norm_47 = torch.nn.functional.layer_norm( + x_237, + (192,), + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ + ) = None + x_238 = torch._C._nn.linear( + layer_norm_47, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_47 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_239 = torch._C._nn.gelu(x_238, approximate="none") + x_238 = None + x_240 = torch.nn.functional.dropout(x_239, 0.0, False, False) + x_239 = None + x_241 = torch._C._nn.linear( + x_240, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_240 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) + x_241 = None + mul_71 = l_self_modules_blocks_modules_23_parameters_gamma_2_ * x_242 + l_self_modules_blocks_modules_23_parameters_gamma_2_ = x_242 = None + x_243 = x_237 + mul_71 + x_237 = mul_71 = None + cls_tokens = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + u = torch.cat((cls_tokens, x_243), dim=1) + layer_norm_48 = torch.nn.functional.layer_norm( + u, + (192,), + l_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + u = ( + l_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_weight_ + ) = ( + l_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_bias_ + ) = None + getitem_76 = layer_norm_48[(slice(None, None, None), 0)] + linear_144 = torch._C._nn.linear( + getitem_76, + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_weight_, + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_bias_, + ) + getitem_76 = l_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_weight_ = l_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_bias_ = (None) + unsqueeze = linear_144.unsqueeze(1) + linear_144 = None + reshape_48 = unsqueeze.reshape(1, 1, 4, 48) + unsqueeze = None + q_24 = reshape_48.permute(0, 2, 1, 3) + reshape_48 = None + linear_145 = torch._C._nn.linear( + layer_norm_48, + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_weight_, + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_bias_, + ) + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_weight_ = l_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_bias_ = (None) + reshape_49 = linear_145.reshape(1, 197, 4, 48) + linear_145 = None + k_24 = reshape_49.permute(0, 2, 1, 3) + reshape_49 = None + linear_146 = torch._C._nn.linear( + layer_norm_48, + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_weight_, + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_bias_, + ) + layer_norm_48 = l_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_weight_ = l_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_bias_ = (None) + reshape_50 = linear_146.reshape(1, 197, 4, 48) + linear_146 = None + v_24 = reshape_50.permute(0, 2, 1, 3) + reshape_50 = None + x_cls = torch._C._nn.scaled_dot_product_attention( + q_24, k_24, v_24, dropout_p=0.0 + ) + q_24 = k_24 = v_24 = None + transpose_49 = x_cls.transpose(1, 2) + x_cls = None + x_cls_1 = transpose_49.reshape(1, 1, 192) + transpose_49 = None + x_cls_2 = torch._C._nn.linear( + x_cls_1, + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_cls_1 = l_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_cls_3 = torch.nn.functional.dropout(x_cls_2, 0.0, False, False) + x_cls_2 = None + mul_72 = ( + l_self_modules_blocks_token_only_modules_0_parameters_gamma_1_ * x_cls_3 + ) + l_self_modules_blocks_token_only_modules_0_parameters_gamma_1_ = x_cls_3 = None + x_cls_4 = cls_tokens + mul_72 + cls_tokens = mul_72 = None + layer_norm_49 = torch.nn.functional.layer_norm( + x_cls_4, + (192,), + l_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_bias_ + ) = None + x_244 = torch._C._nn.linear( + layer_norm_49, + l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_49 = l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_245 = torch._C._nn.gelu(x_244, approximate="none") + x_244 = None + x_246 = torch.nn.functional.dropout(x_245, 0.0, False, False) + x_245 = None + x_247 = torch._C._nn.linear( + x_246, + l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_246 = l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_248 = torch.nn.functional.dropout(x_247, 0.0, False, False) + x_247 = None + mul_73 = l_self_modules_blocks_token_only_modules_0_parameters_gamma_2_ * x_248 + l_self_modules_blocks_token_only_modules_0_parameters_gamma_2_ = x_248 = None + x_cls_5 = x_cls_4 + mul_73 + x_cls_4 = mul_73 = None + u_1 = torch.cat((x_cls_5, x_243), dim=1) + layer_norm_50 = torch.nn.functional.layer_norm( + u_1, + (192,), + l_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + u_1 = ( + l_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_weight_ + ) = ( + l_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_bias_ + ) = None + getitem_77 = layer_norm_50[(slice(None, None, None), 0)] + linear_150 = torch._C._nn.linear( + getitem_77, + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_weight_, + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_bias_, + ) + getitem_77 = l_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_weight_ = l_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_bias_ = (None) + unsqueeze_1 = linear_150.unsqueeze(1) + linear_150 = None + reshape_52 = unsqueeze_1.reshape(1, 1, 4, 48) + unsqueeze_1 = None + q_25 = reshape_52.permute(0, 2, 1, 3) + reshape_52 = None + linear_151 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_weight_, + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_bias_, + ) + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_weight_ = l_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_bias_ = (None) + reshape_53 = linear_151.reshape(1, 197, 4, 48) + linear_151 = None + k_25 = reshape_53.permute(0, 2, 1, 3) + reshape_53 = None + linear_152 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_weight_, + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_bias_, + ) + layer_norm_50 = l_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_weight_ = l_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_bias_ = (None) + reshape_54 = linear_152.reshape(1, 197, 4, 48) + linear_152 = None + v_25 = reshape_54.permute(0, 2, 1, 3) + reshape_54 = None + x_cls_6 = torch._C._nn.scaled_dot_product_attention( + q_25, k_25, v_25, dropout_p=0.0 + ) + q_25 = k_25 = v_25 = None + transpose_50 = x_cls_6.transpose(1, 2) + x_cls_6 = None + x_cls_7 = transpose_50.reshape(1, 1, 192) + transpose_50 = None + x_cls_8 = torch._C._nn.linear( + x_cls_7, + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_cls_7 = l_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_cls_9 = torch.nn.functional.dropout(x_cls_8, 0.0, False, False) + x_cls_8 = None + mul_74 = ( + l_self_modules_blocks_token_only_modules_1_parameters_gamma_1_ * x_cls_9 + ) + l_self_modules_blocks_token_only_modules_1_parameters_gamma_1_ = x_cls_9 = None + x_cls_10 = x_cls_5 + mul_74 + x_cls_5 = mul_74 = None + layer_norm_51 = torch.nn.functional.layer_norm( + x_cls_10, + (192,), + l_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_bias_ + ) = None + x_249 = torch._C._nn.linear( + layer_norm_51, + l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_51 = l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_250 = torch._C._nn.gelu(x_249, approximate="none") + x_249 = None + x_251 = torch.nn.functional.dropout(x_250, 0.0, False, False) + x_250 = None + x_252 = torch._C._nn.linear( + x_251, + l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_251 = l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_253 = torch.nn.functional.dropout(x_252, 0.0, False, False) + x_252 = None + mul_75 = l_self_modules_blocks_token_only_modules_1_parameters_gamma_2_ * x_253 + l_self_modules_blocks_token_only_modules_1_parameters_gamma_2_ = x_253 = None + x_cls_11 = x_cls_10 + mul_75 + x_cls_10 = mul_75 = None + x_254 = torch.cat((x_cls_11, x_243), dim=1) + x_cls_11 = x_243 = None + x_255 = torch.nn.functional.layer_norm( + x_254, + (192,), + l_self_modules_norm_parameters_weight_, + l_self_modules_norm_parameters_bias_, + 1e-06, + ) + x_254 = ( + l_self_modules_norm_parameters_weight_ + ) = l_self_modules_norm_parameters_bias_ = None + x_256 = x_255[(slice(None, None, None), 0)] + x_255 = None + x_257 = torch.nn.functional.dropout(x_256, 0.0, False, False) + x_256 = None + x_258 = torch._C._nn.linear( + x_257, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_257 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_258,) diff --git a/samples/timm/cait_xxs24_224.fb_dist_in1k/weight_meta.py b/samples/timm/cait_xxs24_224.fb_dist_in1k/weight_meta.py new file mode 100644 index 0000000000..f0c269b83c --- /dev/null +++ b/samples/timm/cait_xxs24_224.fb_dist_in1k/weight_meta.py @@ -0,0 +1,4980 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.230 + std = 1.284 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [192, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 196, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_0_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.007 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_0_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_1_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.016 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_1_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_2_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.014 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.017 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_2_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_3_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.010 + std = 0.017 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.004 + std = 0.023 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_3_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_4_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.016 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_4_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_5_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.006 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.010 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_5_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_6_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.007 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_6_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_7_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.007 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_7_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_8_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.017 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_8_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_9_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.007 + std = 0.016 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_9_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_10_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.017 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_10_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_11_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.017 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.006 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_11_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_12_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.005 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_12_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_13_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.023 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.013 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_13_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_14_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.023 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_14_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_15_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.025 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.006 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_15_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_16_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.006 + std = 0.017 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_16_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_17_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.010 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.008 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_17_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_18_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.013 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.028 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_18_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_19_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.010 + std = 0.014 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.023 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_19_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_20_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.005 + std = 0.017 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_20_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_21_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.006 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_21_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_22_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.005 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.005 + std = 0.017 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_22_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_23_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.008 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_23_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_parameters_gamma_1_: + name = "L_self_modules_blocks_token_only_modules_0_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_parameters_gamma_2_: + name = "L_self_modules_blocks_token_only_modules_0_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_parameters_gamma_1_: + name = "L_self_modules_blocks_token_only_modules_1_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_parameters_gamma_2_: + name = "L_self_modules_blocks_token_only_modules_1_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_weight_: + name = "L_self_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_bias_: + name = "L_self_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cait_xxs36_224.fb_dist_in1k/graph_hash.txt b/samples/timm/cait_xxs36_224.fb_dist_in1k/graph_hash.txt new file mode 100644 index 0000000000..8348598639 --- /dev/null +++ b/samples/timm/cait_xxs36_224.fb_dist_in1k/graph_hash.txt @@ -0,0 +1 @@ +630243af3229385eb616e6637b54dc563056dc16fd00c309d948e10aec3e8c0e \ No newline at end of file diff --git a/samples/timm/cait_xxs36_224.fb_dist_in1k/graph_net.json b/samples/timm/cait_xxs36_224.fb_dist_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/cait_xxs36_224.fb_dist_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cait_xxs36_224.fb_dist_in1k/input_meta.py b/samples/timm/cait_xxs36_224.fb_dist_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cait_xxs36_224.fb_dist_in1k/input_tensor_constraints.py b/samples/timm/cait_xxs36_224.fb_dist_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cait_xxs36_224.fb_dist_in1k/model.py b/samples/timm/cait_xxs36_224.fb_dist_in1k/model.py new file mode 100644 index 0000000000..9aa5fd29ee --- /dev/null +++ b/samples/timm/cait_xxs36_224.fb_dist_in1k/model.py @@ -0,0 +1,6972 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_32_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_32_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_32_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_32_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_32_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_32_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_32_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_32_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_32_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_32_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_32_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_32_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_32_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_32_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_32_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_32_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_32_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_32_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_33_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_33_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_33_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_33_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_33_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_33_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_33_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_33_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_33_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_33_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_33_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_33_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_33_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_33_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_33_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_33_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_33_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_33_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_34_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_34_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_34_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_34_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_34_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_34_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_34_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_34_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_34_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_34_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_34_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_34_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_34_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_34_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_34_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_34_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_34_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_34_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_35_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_35_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_35_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_35_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_35_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_35_modules_attn_modules_proj_l_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_35_modules_attn_modules_proj_l_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_35_modules_attn_modules_proj_w_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_35_modules_attn_modules_proj_w_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_35_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_35_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_35_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_35_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_35_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_35_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_35_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_35_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_35_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_parameters_gamma_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_parameters_gamma_2_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_modules_blocks_modules_0_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_0_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_0_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_1_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_1_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_2_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_2_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_3_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_3_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_4_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_4_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_5_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_5_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_6_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_6_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_7_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_7_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_8_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_8_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_9_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_9_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_10_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_10_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_11_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_11_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_12_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_12_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_12_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_12_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_13_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_13_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_13_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_13_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_14_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_14_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_14_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_14_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_15_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_15_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_15_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_15_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_16_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_16_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_16_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_16_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_17_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_17_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_17_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_17_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_18_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_18_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_18_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_18_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_19_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_19_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_19_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_19_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_20_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_20_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_20_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_20_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_21_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_21_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_21_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_21_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_22_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_22_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_22_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_22_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_23_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_23_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_23_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_23_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_24_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_24_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_24_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_24_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_24_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_24_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_24_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_24_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_24_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_24_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_24_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_24_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_24_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_24_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_24_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_24_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_24_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_24_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_24_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_24_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_25_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_25_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_25_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_25_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_25_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_25_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_25_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_25_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_25_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_25_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_25_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_25_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_25_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_25_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_25_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_25_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_25_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_25_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_25_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_25_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_26_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_26_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_26_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_26_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_26_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_26_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_26_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_26_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_26_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_26_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_26_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_26_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_26_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_26_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_26_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_26_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_26_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_26_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_26_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_26_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_27_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_27_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_27_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_27_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_27_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_27_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_27_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_27_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_27_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_27_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_27_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_27_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_27_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_27_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_27_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_27_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_27_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_27_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_27_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_27_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_28_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_28_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_28_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_28_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_28_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_28_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_28_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_28_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_28_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_28_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_28_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_28_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_28_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_28_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_28_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_28_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_28_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_28_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_28_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_28_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_29_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_29_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_29_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_29_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_29_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_29_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_29_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_29_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_29_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_29_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_29_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_29_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_29_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_29_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_29_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_29_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_29_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_29_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_29_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_29_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_30_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_30_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_30_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_30_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_30_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_30_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_30_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_30_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_30_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_30_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_30_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_30_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_30_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_30_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_30_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_30_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_30_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_30_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_30_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_30_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_31_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_31_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_31_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_31_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_31_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_31_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_31_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_31_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_31_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_31_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_31_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_31_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_31_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_31_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_31_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_31_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_31_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_31_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_31_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_31_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_32_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_32_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_32_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_32_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_32_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_32_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_32_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_32_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_32_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_32_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_32_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_32_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_32_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_32_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_32_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_32_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_32_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_32_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_32_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_32_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_32_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_32_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_32_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_32_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_32_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_32_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_32_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_32_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_32_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_32_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_32_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_32_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_32_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_32_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_32_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_32_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_33_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_33_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_33_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_33_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_33_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_33_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_33_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_33_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_33_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_33_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_33_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_33_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_33_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_33_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_33_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_33_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_33_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_33_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_33_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_33_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_33_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_33_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_33_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_33_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_33_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_33_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_33_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_33_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_33_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_33_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_33_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_33_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_33_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_33_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_33_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_33_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_34_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_34_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_34_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_34_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_34_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_34_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_34_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_34_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_34_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_34_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_34_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_34_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_34_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_34_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_34_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_34_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_34_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_34_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_34_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_34_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_34_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_34_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_34_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_34_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_34_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_34_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_34_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_34_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_34_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_34_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_34_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_34_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_34_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_34_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_34_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_34_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_35_parameters_gamma_1_ = ( + L_self_modules_blocks_modules_35_parameters_gamma_1_ + ) + l_self_modules_blocks_modules_35_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_35_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_35_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_35_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_35_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_35_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_35_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_35_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_35_modules_attn_modules_proj_l_parameters_weight_ = L_self_modules_blocks_modules_35_modules_attn_modules_proj_l_parameters_weight_ + l_self_modules_blocks_modules_35_modules_attn_modules_proj_l_parameters_bias_ = L_self_modules_blocks_modules_35_modules_attn_modules_proj_l_parameters_bias_ + l_self_modules_blocks_modules_35_modules_attn_modules_proj_w_parameters_weight_ = L_self_modules_blocks_modules_35_modules_attn_modules_proj_w_parameters_weight_ + l_self_modules_blocks_modules_35_modules_attn_modules_proj_w_parameters_bias_ = L_self_modules_blocks_modules_35_modules_attn_modules_proj_w_parameters_bias_ + l_self_modules_blocks_modules_35_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_35_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_35_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_35_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_35_parameters_gamma_2_ = ( + L_self_modules_blocks_modules_35_parameters_gamma_2_ + ) + l_self_modules_blocks_modules_35_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_35_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_35_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_35_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_35_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_35_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_35_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_35_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_35_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_35_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_35_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_35_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_modules_blocks_token_only_modules_0_parameters_gamma_1_ = ( + L_self_modules_blocks_token_only_modules_0_parameters_gamma_1_ + ) + l_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_weight_ = L_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_weight_ + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_bias_ = L_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_bias_ + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_weight_ = L_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_weight_ + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_bias_ = L_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_bias_ + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_weight_ = L_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_weight_ + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_bias_ = L_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_bias_ + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_token_only_modules_0_parameters_gamma_2_ = ( + L_self_modules_blocks_token_only_modules_0_parameters_gamma_2_ + ) + l_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_token_only_modules_1_parameters_gamma_1_ = ( + L_self_modules_blocks_token_only_modules_1_parameters_gamma_1_ + ) + l_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_weight_ = L_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_weight_ + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_bias_ = L_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_bias_ + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_weight_ = L_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_weight_ + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_bias_ = L_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_bias_ + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_weight_ = L_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_weight_ + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_bias_ = L_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_bias_ + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_token_only_modules_1_parameters_gamma_2_ = ( + L_self_modules_blocks_token_only_modules_1_parameters_gamma_2_ + ) + l_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_norm_parameters_weight_ = L_self_modules_norm_parameters_weight_ + l_self_modules_norm_parameters_bias_ = L_self_modules_norm_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + x_2 = x_1 + l_self_parameters_pos_embed_ + x_1 = l_self_parameters_pos_embed_ = None + x_3 = torch.nn.functional.dropout(x_2, 0.0, False, False) + x_2 = None + layer_norm = torch.nn.functional.layer_norm( + x_3, + (192,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + layer_norm, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 196, 3, 4, 48) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + getitem_4 = qkv[0] + q = getitem_4 * 0.14433756729740643 + getitem_4 = None + k = qkv[1] + v = qkv[2] + qkv = None + transpose_1 = k.transpose(-2, -1) + k = None + attn = q @ transpose_1 + q = transpose_1 = None + permute_1 = attn.permute(0, 2, 3, 1) + attn = None + linear_1 = torch._C._nn.linear( + permute_1, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_1 = l_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_1 = linear_1.permute(0, 3, 1, 2) + linear_1 = None + attn_2 = attn_1.softmax(dim=-1) + attn_1 = None + permute_3 = attn_2.permute(0, 2, 3, 1) + attn_2 = None + linear_2 = torch._C._nn.linear( + permute_3, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_3 = l_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_3 = linear_2.permute(0, 3, 1, 2) + linear_2 = None + attn_4 = torch.nn.functional.dropout(attn_3, 0.0, False, False) + attn_3 = None + matmul_1 = attn_4 @ v + attn_4 = v = None + transpose_2 = matmul_1.transpose(1, 2) + matmul_1 = None + x_4 = transpose_2.reshape(1, 196, 192) + transpose_2 = None + x_5 = torch._C._nn.linear( + x_4, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_4 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_6 = torch.nn.functional.dropout(x_5, 0.0, False, False) + x_5 = None + mul_1 = l_self_modules_blocks_modules_0_parameters_gamma_1_ * x_6 + l_self_modules_blocks_modules_0_parameters_gamma_1_ = x_6 = None + x_7 = x_3 + mul_1 + x_3 = mul_1 = None + layer_norm_1 = torch.nn.functional.layer_norm( + x_7, + (192,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_8 = torch._C._nn.linear( + layer_norm_1, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_1 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_9 = torch._C._nn.gelu(x_8, approximate="none") + x_8 = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = torch._C._nn.linear( + x_10, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_10 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + mul_2 = l_self_modules_blocks_modules_0_parameters_gamma_2_ * x_12 + l_self_modules_blocks_modules_0_parameters_gamma_2_ = x_12 = None + x_13 = x_7 + mul_2 + x_7 = mul_2 = None + layer_norm_2 = torch.nn.functional.layer_norm( + x_13, + (192,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_6 = torch._C._nn.linear( + layer_norm_2, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_2 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_6.reshape(1, 196, 3, 4, 48) + linear_6 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + getitem_7 = qkv_1[0] + q_1 = getitem_7 * 0.14433756729740643 + getitem_7 = None + k_1 = qkv_1[1] + v_1 = qkv_1[2] + qkv_1 = None + transpose_3 = k_1.transpose(-2, -1) + k_1 = None + attn_5 = q_1 @ transpose_3 + q_1 = transpose_3 = None + permute_6 = attn_5.permute(0, 2, 3, 1) + attn_5 = None + linear_7 = torch._C._nn.linear( + permute_6, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_6 = l_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_6 = linear_7.permute(0, 3, 1, 2) + linear_7 = None + attn_7 = attn_6.softmax(dim=-1) + attn_6 = None + permute_8 = attn_7.permute(0, 2, 3, 1) + attn_7 = None + linear_8 = torch._C._nn.linear( + permute_8, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_8 = l_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_8 = linear_8.permute(0, 3, 1, 2) + linear_8 = None + attn_9 = torch.nn.functional.dropout(attn_8, 0.0, False, False) + attn_8 = None + matmul_3 = attn_9 @ v_1 + attn_9 = v_1 = None + transpose_4 = matmul_3.transpose(1, 2) + matmul_3 = None + x_14 = transpose_4.reshape(1, 196, 192) + transpose_4 = None + x_15 = torch._C._nn.linear( + x_14, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_14 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + mul_4 = l_self_modules_blocks_modules_1_parameters_gamma_1_ * x_16 + l_self_modules_blocks_modules_1_parameters_gamma_1_ = x_16 = None + x_17 = x_13 + mul_4 + x_13 = mul_4 = None + layer_norm_3 = torch.nn.functional.layer_norm( + x_17, + (192,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_18 = torch._C._nn.linear( + layer_norm_3, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_3 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_19 = torch._C._nn.gelu(x_18, approximate="none") + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + mul_5 = l_self_modules_blocks_modules_1_parameters_gamma_2_ * x_22 + l_self_modules_blocks_modules_1_parameters_gamma_2_ = x_22 = None + x_23 = x_17 + mul_5 + x_17 = mul_5 = None + layer_norm_4 = torch.nn.functional.layer_norm( + x_23, + (192,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + layer_norm_4, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_4 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_12.reshape(1, 196, 3, 4, 48) + linear_12 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + getitem_10 = qkv_2[0] + q_2 = getitem_10 * 0.14433756729740643 + getitem_10 = None + k_2 = qkv_2[1] + v_2 = qkv_2[2] + qkv_2 = None + transpose_5 = k_2.transpose(-2, -1) + k_2 = None + attn_10 = q_2 @ transpose_5 + q_2 = transpose_5 = None + permute_11 = attn_10.permute(0, 2, 3, 1) + attn_10 = None + linear_13 = torch._C._nn.linear( + permute_11, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_11 = l_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_11 = linear_13.permute(0, 3, 1, 2) + linear_13 = None + attn_12 = attn_11.softmax(dim=-1) + attn_11 = None + permute_13 = attn_12.permute(0, 2, 3, 1) + attn_12 = None + linear_14 = torch._C._nn.linear( + permute_13, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_13 = l_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_13 = linear_14.permute(0, 3, 1, 2) + linear_14 = None + attn_14 = torch.nn.functional.dropout(attn_13, 0.0, False, False) + attn_13 = None + matmul_5 = attn_14 @ v_2 + attn_14 = v_2 = None + transpose_6 = matmul_5.transpose(1, 2) + matmul_5 = None + x_24 = transpose_6.reshape(1, 196, 192) + transpose_6 = None + x_25 = torch._C._nn.linear( + x_24, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_24 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_26 = torch.nn.functional.dropout(x_25, 0.0, False, False) + x_25 = None + mul_7 = l_self_modules_blocks_modules_2_parameters_gamma_1_ * x_26 + l_self_modules_blocks_modules_2_parameters_gamma_1_ = x_26 = None + x_27 = x_23 + mul_7 + x_23 = mul_7 = None + layer_norm_5 = torch.nn.functional.layer_norm( + x_27, + (192,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_28 = torch._C._nn.linear( + layer_norm_5, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_5 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_29 = torch._C._nn.gelu(x_28, approximate="none") + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_31 = torch._C._nn.linear( + x_30, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_30 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + mul_8 = l_self_modules_blocks_modules_2_parameters_gamma_2_ * x_32 + l_self_modules_blocks_modules_2_parameters_gamma_2_ = x_32 = None + x_33 = x_27 + mul_8 + x_27 = mul_8 = None + layer_norm_6 = torch.nn.functional.layer_norm( + x_33, + (192,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_18 = torch._C._nn.linear( + layer_norm_6, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_6 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_18.reshape(1, 196, 3, 4, 48) + linear_18 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + getitem_13 = qkv_3[0] + q_3 = getitem_13 * 0.14433756729740643 + getitem_13 = None + k_3 = qkv_3[1] + v_3 = qkv_3[2] + qkv_3 = None + transpose_7 = k_3.transpose(-2, -1) + k_3 = None + attn_15 = q_3 @ transpose_7 + q_3 = transpose_7 = None + permute_16 = attn_15.permute(0, 2, 3, 1) + attn_15 = None + linear_19 = torch._C._nn.linear( + permute_16, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_16 = l_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_16 = linear_19.permute(0, 3, 1, 2) + linear_19 = None + attn_17 = attn_16.softmax(dim=-1) + attn_16 = None + permute_18 = attn_17.permute(0, 2, 3, 1) + attn_17 = None + linear_20 = torch._C._nn.linear( + permute_18, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_18 = l_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_18 = linear_20.permute(0, 3, 1, 2) + linear_20 = None + attn_19 = torch.nn.functional.dropout(attn_18, 0.0, False, False) + attn_18 = None + matmul_7 = attn_19 @ v_3 + attn_19 = v_3 = None + transpose_8 = matmul_7.transpose(1, 2) + matmul_7 = None + x_34 = transpose_8.reshape(1, 196, 192) + transpose_8 = None + x_35 = torch._C._nn.linear( + x_34, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_34 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + mul_10 = l_self_modules_blocks_modules_3_parameters_gamma_1_ * x_36 + l_self_modules_blocks_modules_3_parameters_gamma_1_ = x_36 = None + x_37 = x_33 + mul_10 + x_33 = mul_10 = None + layer_norm_7 = torch.nn.functional.layer_norm( + x_37, + (192,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_38 = torch._C._nn.linear( + layer_norm_7, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_7 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_39 = torch._C._nn.gelu(x_38, approximate="none") + x_38 = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = torch._C._nn.linear( + x_40, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_40 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_42 = torch.nn.functional.dropout(x_41, 0.0, False, False) + x_41 = None + mul_11 = l_self_modules_blocks_modules_3_parameters_gamma_2_ * x_42 + l_self_modules_blocks_modules_3_parameters_gamma_2_ = x_42 = None + x_43 = x_37 + mul_11 + x_37 = mul_11 = None + layer_norm_8 = torch.nn.functional.layer_norm( + x_43, + (192,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + layer_norm_8, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_8 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_24.reshape(1, 196, 3, 4, 48) + linear_24 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + getitem_16 = qkv_4[0] + q_4 = getitem_16 * 0.14433756729740643 + getitem_16 = None + k_4 = qkv_4[1] + v_4 = qkv_4[2] + qkv_4 = None + transpose_9 = k_4.transpose(-2, -1) + k_4 = None + attn_20 = q_4 @ transpose_9 + q_4 = transpose_9 = None + permute_21 = attn_20.permute(0, 2, 3, 1) + attn_20 = None + linear_25 = torch._C._nn.linear( + permute_21, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_21 = l_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_21 = linear_25.permute(0, 3, 1, 2) + linear_25 = None + attn_22 = attn_21.softmax(dim=-1) + attn_21 = None + permute_23 = attn_22.permute(0, 2, 3, 1) + attn_22 = None + linear_26 = torch._C._nn.linear( + permute_23, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_23 = l_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_23 = linear_26.permute(0, 3, 1, 2) + linear_26 = None + attn_24 = torch.nn.functional.dropout(attn_23, 0.0, False, False) + attn_23 = None + matmul_9 = attn_24 @ v_4 + attn_24 = v_4 = None + transpose_10 = matmul_9.transpose(1, 2) + matmul_9 = None + x_44 = transpose_10.reshape(1, 196, 192) + transpose_10 = None + x_45 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_44 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_46 = torch.nn.functional.dropout(x_45, 0.0, False, False) + x_45 = None + mul_13 = l_self_modules_blocks_modules_4_parameters_gamma_1_ * x_46 + l_self_modules_blocks_modules_4_parameters_gamma_1_ = x_46 = None + x_47 = x_43 + mul_13 + x_43 = mul_13 = None + layer_norm_9 = torch.nn.functional.layer_norm( + x_47, + (192,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_48 = torch._C._nn.linear( + layer_norm_9, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_9 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_49 = torch._C._nn.gelu(x_48, approximate="none") + x_48 = None + x_50 = torch.nn.functional.dropout(x_49, 0.0, False, False) + x_49 = None + x_51 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_52 = torch.nn.functional.dropout(x_51, 0.0, False, False) + x_51 = None + mul_14 = l_self_modules_blocks_modules_4_parameters_gamma_2_ * x_52 + l_self_modules_blocks_modules_4_parameters_gamma_2_ = x_52 = None + x_53 = x_47 + mul_14 + x_47 = mul_14 = None + layer_norm_10 = torch.nn.functional.layer_norm( + x_53, + (192,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_30 = torch._C._nn.linear( + layer_norm_10, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_10 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_30.reshape(1, 196, 3, 4, 48) + linear_30 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + getitem_19 = qkv_5[0] + q_5 = getitem_19 * 0.14433756729740643 + getitem_19 = None + k_5 = qkv_5[1] + v_5 = qkv_5[2] + qkv_5 = None + transpose_11 = k_5.transpose(-2, -1) + k_5 = None + attn_25 = q_5 @ transpose_11 + q_5 = transpose_11 = None + permute_26 = attn_25.permute(0, 2, 3, 1) + attn_25 = None + linear_31 = torch._C._nn.linear( + permute_26, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_26 = l_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_26 = linear_31.permute(0, 3, 1, 2) + linear_31 = None + attn_27 = attn_26.softmax(dim=-1) + attn_26 = None + permute_28 = attn_27.permute(0, 2, 3, 1) + attn_27 = None + linear_32 = torch._C._nn.linear( + permute_28, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_28 = l_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_28 = linear_32.permute(0, 3, 1, 2) + linear_32 = None + attn_29 = torch.nn.functional.dropout(attn_28, 0.0, False, False) + attn_28 = None + matmul_11 = attn_29 @ v_5 + attn_29 = v_5 = None + transpose_12 = matmul_11.transpose(1, 2) + matmul_11 = None + x_54 = transpose_12.reshape(1, 196, 192) + transpose_12 = None + x_55 = torch._C._nn.linear( + x_54, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_54 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_56 = torch.nn.functional.dropout(x_55, 0.0, False, False) + x_55 = None + mul_16 = l_self_modules_blocks_modules_5_parameters_gamma_1_ * x_56 + l_self_modules_blocks_modules_5_parameters_gamma_1_ = x_56 = None + x_57 = x_53 + mul_16 + x_53 = mul_16 = None + layer_norm_11 = torch.nn.functional.layer_norm( + x_57, + (192,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_58 = torch._C._nn.linear( + layer_norm_11, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_11 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_59 = torch._C._nn.gelu(x_58, approximate="none") + x_58 = None + x_60 = torch.nn.functional.dropout(x_59, 0.0, False, False) + x_59 = None + x_61 = torch._C._nn.linear( + x_60, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_60 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + mul_17 = l_self_modules_blocks_modules_5_parameters_gamma_2_ * x_62 + l_self_modules_blocks_modules_5_parameters_gamma_2_ = x_62 = None + x_63 = x_57 + mul_17 + x_57 = mul_17 = None + layer_norm_12 = torch.nn.functional.layer_norm( + x_63, + (192,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_12 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_36.reshape(1, 196, 3, 4, 48) + linear_36 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + getitem_22 = qkv_6[0] + q_6 = getitem_22 * 0.14433756729740643 + getitem_22 = None + k_6 = qkv_6[1] + v_6 = qkv_6[2] + qkv_6 = None + transpose_13 = k_6.transpose(-2, -1) + k_6 = None + attn_30 = q_6 @ transpose_13 + q_6 = transpose_13 = None + permute_31 = attn_30.permute(0, 2, 3, 1) + attn_30 = None + linear_37 = torch._C._nn.linear( + permute_31, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_31 = l_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_31 = linear_37.permute(0, 3, 1, 2) + linear_37 = None + attn_32 = attn_31.softmax(dim=-1) + attn_31 = None + permute_33 = attn_32.permute(0, 2, 3, 1) + attn_32 = None + linear_38 = torch._C._nn.linear( + permute_33, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_33 = l_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_33 = linear_38.permute(0, 3, 1, 2) + linear_38 = None + attn_34 = torch.nn.functional.dropout(attn_33, 0.0, False, False) + attn_33 = None + matmul_13 = attn_34 @ v_6 + attn_34 = v_6 = None + transpose_14 = matmul_13.transpose(1, 2) + matmul_13 = None + x_64 = transpose_14.reshape(1, 196, 192) + transpose_14 = None + x_65 = torch._C._nn.linear( + x_64, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_64 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + mul_19 = l_self_modules_blocks_modules_6_parameters_gamma_1_ * x_66 + l_self_modules_blocks_modules_6_parameters_gamma_1_ = x_66 = None + x_67 = x_63 + mul_19 + x_63 = mul_19 = None + layer_norm_13 = torch.nn.functional.layer_norm( + x_67, + (192,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_68 = torch._C._nn.linear( + layer_norm_13, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_13 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_69 = torch._C._nn.gelu(x_68, approximate="none") + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_71 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_70 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + mul_20 = l_self_modules_blocks_modules_6_parameters_gamma_2_ * x_72 + l_self_modules_blocks_modules_6_parameters_gamma_2_ = x_72 = None + x_73 = x_67 + mul_20 + x_67 = mul_20 = None + layer_norm_14 = torch.nn.functional.layer_norm( + x_73, + (192,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_42 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_14 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_42.reshape(1, 196, 3, 4, 48) + linear_42 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + getitem_25 = qkv_7[0] + q_7 = getitem_25 * 0.14433756729740643 + getitem_25 = None + k_7 = qkv_7[1] + v_7 = qkv_7[2] + qkv_7 = None + transpose_15 = k_7.transpose(-2, -1) + k_7 = None + attn_35 = q_7 @ transpose_15 + q_7 = transpose_15 = None + permute_36 = attn_35.permute(0, 2, 3, 1) + attn_35 = None + linear_43 = torch._C._nn.linear( + permute_36, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_36 = l_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_36 = linear_43.permute(0, 3, 1, 2) + linear_43 = None + attn_37 = attn_36.softmax(dim=-1) + attn_36 = None + permute_38 = attn_37.permute(0, 2, 3, 1) + attn_37 = None + linear_44 = torch._C._nn.linear( + permute_38, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_38 = l_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_38 = linear_44.permute(0, 3, 1, 2) + linear_44 = None + attn_39 = torch.nn.functional.dropout(attn_38, 0.0, False, False) + attn_38 = None + matmul_15 = attn_39 @ v_7 + attn_39 = v_7 = None + transpose_16 = matmul_15.transpose(1, 2) + matmul_15 = None + x_74 = transpose_16.reshape(1, 196, 192) + transpose_16 = None + x_75 = torch._C._nn.linear( + x_74, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_74 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_76 = torch.nn.functional.dropout(x_75, 0.0, False, False) + x_75 = None + mul_22 = l_self_modules_blocks_modules_7_parameters_gamma_1_ * x_76 + l_self_modules_blocks_modules_7_parameters_gamma_1_ = x_76 = None + x_77 = x_73 + mul_22 + x_73 = mul_22 = None + layer_norm_15 = torch.nn.functional.layer_norm( + x_77, + (192,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_78 = torch._C._nn.linear( + layer_norm_15, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_15 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_79 = torch._C._nn.gelu(x_78, approximate="none") + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = torch._C._nn.linear( + x_80, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_80 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_82 = torch.nn.functional.dropout(x_81, 0.0, False, False) + x_81 = None + mul_23 = l_self_modules_blocks_modules_7_parameters_gamma_2_ * x_82 + l_self_modules_blocks_modules_7_parameters_gamma_2_ = x_82 = None + x_83 = x_77 + mul_23 + x_77 = mul_23 = None + layer_norm_16 = torch.nn.functional.layer_norm( + x_83, + (192,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_48 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_16 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_48.reshape(1, 196, 3, 4, 48) + linear_48 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + getitem_28 = qkv_8[0] + q_8 = getitem_28 * 0.14433756729740643 + getitem_28 = None + k_8 = qkv_8[1] + v_8 = qkv_8[2] + qkv_8 = None + transpose_17 = k_8.transpose(-2, -1) + k_8 = None + attn_40 = q_8 @ transpose_17 + q_8 = transpose_17 = None + permute_41 = attn_40.permute(0, 2, 3, 1) + attn_40 = None + linear_49 = torch._C._nn.linear( + permute_41, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_41 = l_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_41 = linear_49.permute(0, 3, 1, 2) + linear_49 = None + attn_42 = attn_41.softmax(dim=-1) + attn_41 = None + permute_43 = attn_42.permute(0, 2, 3, 1) + attn_42 = None + linear_50 = torch._C._nn.linear( + permute_43, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_43 = l_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_43 = linear_50.permute(0, 3, 1, 2) + linear_50 = None + attn_44 = torch.nn.functional.dropout(attn_43, 0.0, False, False) + attn_43 = None + matmul_17 = attn_44 @ v_8 + attn_44 = v_8 = None + transpose_18 = matmul_17.transpose(1, 2) + matmul_17 = None + x_84 = transpose_18.reshape(1, 196, 192) + transpose_18 = None + x_85 = torch._C._nn.linear( + x_84, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_84 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_86 = torch.nn.functional.dropout(x_85, 0.0, False, False) + x_85 = None + mul_25 = l_self_modules_blocks_modules_8_parameters_gamma_1_ * x_86 + l_self_modules_blocks_modules_8_parameters_gamma_1_ = x_86 = None + x_87 = x_83 + mul_25 + x_83 = mul_25 = None + layer_norm_17 = torch.nn.functional.layer_norm( + x_87, + (192,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_88 = torch._C._nn.linear( + layer_norm_17, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_17 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_89 = torch._C._nn.gelu(x_88, approximate="none") + x_88 = None + x_90 = torch.nn.functional.dropout(x_89, 0.0, False, False) + x_89 = None + x_91 = torch._C._nn.linear( + x_90, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_90 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + mul_26 = l_self_modules_blocks_modules_8_parameters_gamma_2_ * x_92 + l_self_modules_blocks_modules_8_parameters_gamma_2_ = x_92 = None + x_93 = x_87 + mul_26 + x_87 = mul_26 = None + layer_norm_18 = torch.nn.functional.layer_norm( + x_93, + (192,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_54 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_18 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_54.reshape(1, 196, 3, 4, 48) + linear_54 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + getitem_31 = qkv_9[0] + q_9 = getitem_31 * 0.14433756729740643 + getitem_31 = None + k_9 = qkv_9[1] + v_9 = qkv_9[2] + qkv_9 = None + transpose_19 = k_9.transpose(-2, -1) + k_9 = None + attn_45 = q_9 @ transpose_19 + q_9 = transpose_19 = None + permute_46 = attn_45.permute(0, 2, 3, 1) + attn_45 = None + linear_55 = torch._C._nn.linear( + permute_46, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_46 = l_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_bias_ + ) = None + attn_46 = linear_55.permute(0, 3, 1, 2) + linear_55 = None + attn_47 = attn_46.softmax(dim=-1) + attn_46 = None + permute_48 = attn_47.permute(0, 2, 3, 1) + attn_47 = None + linear_56 = torch._C._nn.linear( + permute_48, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_48 = l_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_bias_ + ) = None + attn_48 = linear_56.permute(0, 3, 1, 2) + linear_56 = None + attn_49 = torch.nn.functional.dropout(attn_48, 0.0, False, False) + attn_48 = None + matmul_19 = attn_49 @ v_9 + attn_49 = v_9 = None + transpose_20 = matmul_19.transpose(1, 2) + matmul_19 = None + x_94 = transpose_20.reshape(1, 196, 192) + transpose_20 = None + x_95 = torch._C._nn.linear( + x_94, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_94 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_96 = torch.nn.functional.dropout(x_95, 0.0, False, False) + x_95 = None + mul_28 = l_self_modules_blocks_modules_9_parameters_gamma_1_ * x_96 + l_self_modules_blocks_modules_9_parameters_gamma_1_ = x_96 = None + x_97 = x_93 + mul_28 + x_93 = mul_28 = None + layer_norm_19 = torch.nn.functional.layer_norm( + x_97, + (192,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_98 = torch._C._nn.linear( + layer_norm_19, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_19 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_99 = torch._C._nn.gelu(x_98, approximate="none") + x_98 = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + x_101 = torch._C._nn.linear( + x_100, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_100 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + mul_29 = l_self_modules_blocks_modules_9_parameters_gamma_2_ * x_102 + l_self_modules_blocks_modules_9_parameters_gamma_2_ = x_102 = None + x_103 = x_97 + mul_29 + x_97 = mul_29 = None + layer_norm_20 = torch.nn.functional.layer_norm( + x_103, + (192,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_60 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_20 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_60.reshape(1, 196, 3, 4, 48) + linear_60 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + getitem_34 = qkv_10[0] + q_10 = getitem_34 * 0.14433756729740643 + getitem_34 = None + k_10 = qkv_10[1] + v_10 = qkv_10[2] + qkv_10 = None + transpose_21 = k_10.transpose(-2, -1) + k_10 = None + attn_50 = q_10 @ transpose_21 + q_10 = transpose_21 = None + permute_51 = attn_50.permute(0, 2, 3, 1) + attn_50 = None + linear_61 = torch._C._nn.linear( + permute_51, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_51 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_51 = linear_61.permute(0, 3, 1, 2) + linear_61 = None + attn_52 = attn_51.softmax(dim=-1) + attn_51 = None + permute_53 = attn_52.permute(0, 2, 3, 1) + attn_52 = None + linear_62 = torch._C._nn.linear( + permute_53, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_53 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_53 = linear_62.permute(0, 3, 1, 2) + linear_62 = None + attn_54 = torch.nn.functional.dropout(attn_53, 0.0, False, False) + attn_53 = None + matmul_21 = attn_54 @ v_10 + attn_54 = v_10 = None + transpose_22 = matmul_21.transpose(1, 2) + matmul_21 = None + x_104 = transpose_22.reshape(1, 196, 192) + transpose_22 = None + x_105 = torch._C._nn.linear( + x_104, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_104 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + mul_31 = l_self_modules_blocks_modules_10_parameters_gamma_1_ * x_106 + l_self_modules_blocks_modules_10_parameters_gamma_1_ = x_106 = None + x_107 = x_103 + mul_31 + x_103 = mul_31 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_107, + (192,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_108 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_21 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_109 = torch._C._nn.gelu(x_108, approximate="none") + x_108 = None + x_110 = torch.nn.functional.dropout(x_109, 0.0, False, False) + x_109 = None + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_110 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_112 = torch.nn.functional.dropout(x_111, 0.0, False, False) + x_111 = None + mul_32 = l_self_modules_blocks_modules_10_parameters_gamma_2_ * x_112 + l_self_modules_blocks_modules_10_parameters_gamma_2_ = x_112 = None + x_113 = x_107 + mul_32 + x_107 = mul_32 = None + layer_norm_22 = torch.nn.functional.layer_norm( + x_113, + (192,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_66 = torch._C._nn.linear( + layer_norm_22, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_22 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_66.reshape(1, 196, 3, 4, 48) + linear_66 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + getitem_37 = qkv_11[0] + q_11 = getitem_37 * 0.14433756729740643 + getitem_37 = None + k_11 = qkv_11[1] + v_11 = qkv_11[2] + qkv_11 = None + transpose_23 = k_11.transpose(-2, -1) + k_11 = None + attn_55 = q_11 @ transpose_23 + q_11 = transpose_23 = None + permute_56 = attn_55.permute(0, 2, 3, 1) + attn_55 = None + linear_67 = torch._C._nn.linear( + permute_56, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_56 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_56 = linear_67.permute(0, 3, 1, 2) + linear_67 = None + attn_57 = attn_56.softmax(dim=-1) + attn_56 = None + permute_58 = attn_57.permute(0, 2, 3, 1) + attn_57 = None + linear_68 = torch._C._nn.linear( + permute_58, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_58 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_58 = linear_68.permute(0, 3, 1, 2) + linear_68 = None + attn_59 = torch.nn.functional.dropout(attn_58, 0.0, False, False) + attn_58 = None + matmul_23 = attn_59 @ v_11 + attn_59 = v_11 = None + transpose_24 = matmul_23.transpose(1, 2) + matmul_23 = None + x_114 = transpose_24.reshape(1, 196, 192) + transpose_24 = None + x_115 = torch._C._nn.linear( + x_114, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_114 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + mul_34 = l_self_modules_blocks_modules_11_parameters_gamma_1_ * x_116 + l_self_modules_blocks_modules_11_parameters_gamma_1_ = x_116 = None + x_117 = x_113 + mul_34 + x_113 = mul_34 = None + layer_norm_23 = torch.nn.functional.layer_norm( + x_117, + (192,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_118 = torch._C._nn.linear( + layer_norm_23, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_23 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_119 = torch._C._nn.gelu(x_118, approximate="none") + x_118 = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + x_121 = torch._C._nn.linear( + x_120, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_120 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_122 = torch.nn.functional.dropout(x_121, 0.0, False, False) + x_121 = None + mul_35 = l_self_modules_blocks_modules_11_parameters_gamma_2_ * x_122 + l_self_modules_blocks_modules_11_parameters_gamma_2_ = x_122 = None + x_123 = x_117 + mul_35 + x_117 = mul_35 = None + layer_norm_24 = torch.nn.functional.layer_norm( + x_123, + (192,), + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ + ) = None + linear_72 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_24 = ( + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_24 = linear_72.reshape(1, 196, 3, 4, 48) + linear_72 = None + qkv_12 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + getitem_40 = qkv_12[0] + q_12 = getitem_40 * 0.14433756729740643 + getitem_40 = None + k_12 = qkv_12[1] + v_12 = qkv_12[2] + qkv_12 = None + transpose_25 = k_12.transpose(-2, -1) + k_12 = None + attn_60 = q_12 @ transpose_25 + q_12 = transpose_25 = None + permute_61 = attn_60.permute(0, 2, 3, 1) + attn_60 = None + linear_73 = torch._C._nn.linear( + permute_61, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_61 = l_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_61 = linear_73.permute(0, 3, 1, 2) + linear_73 = None + attn_62 = attn_61.softmax(dim=-1) + attn_61 = None + permute_63 = attn_62.permute(0, 2, 3, 1) + attn_62 = None + linear_74 = torch._C._nn.linear( + permute_63, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_63 = l_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_63 = linear_74.permute(0, 3, 1, 2) + linear_74 = None + attn_64 = torch.nn.functional.dropout(attn_63, 0.0, False, False) + attn_63 = None + matmul_25 = attn_64 @ v_12 + attn_64 = v_12 = None + transpose_26 = matmul_25.transpose(1, 2) + matmul_25 = None + x_124 = transpose_26.reshape(1, 196, 192) + transpose_26 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_, + ) + x_124 = l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ + ) = None + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + mul_37 = l_self_modules_blocks_modules_12_parameters_gamma_1_ * x_126 + l_self_modules_blocks_modules_12_parameters_gamma_1_ = x_126 = None + x_127 = x_123 + mul_37 + x_123 = mul_37 = None + layer_norm_25 = torch.nn.functional.layer_norm( + x_127, + (192,), + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ + ) = None + x_128 = torch._C._nn.linear( + layer_norm_25, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_25 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_129 = torch._C._nn.gelu(x_128, approximate="none") + x_128 = None + x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) + x_129 = None + x_131 = torch._C._nn.linear( + x_130, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_130 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_132 = torch.nn.functional.dropout(x_131, 0.0, False, False) + x_131 = None + mul_38 = l_self_modules_blocks_modules_12_parameters_gamma_2_ * x_132 + l_self_modules_blocks_modules_12_parameters_gamma_2_ = x_132 = None + x_133 = x_127 + mul_38 + x_127 = mul_38 = None + layer_norm_26 = torch.nn.functional.layer_norm( + x_133, + (192,), + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ + ) = None + linear_78 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_26 = ( + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_26 = linear_78.reshape(1, 196, 3, 4, 48) + linear_78 = None + qkv_13 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + getitem_43 = qkv_13[0] + q_13 = getitem_43 * 0.14433756729740643 + getitem_43 = None + k_13 = qkv_13[1] + v_13 = qkv_13[2] + qkv_13 = None + transpose_27 = k_13.transpose(-2, -1) + k_13 = None + attn_65 = q_13 @ transpose_27 + q_13 = transpose_27 = None + permute_66 = attn_65.permute(0, 2, 3, 1) + attn_65 = None + linear_79 = torch._C._nn.linear( + permute_66, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_66 = l_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_66 = linear_79.permute(0, 3, 1, 2) + linear_79 = None + attn_67 = attn_66.softmax(dim=-1) + attn_66 = None + permute_68 = attn_67.permute(0, 2, 3, 1) + attn_67 = None + linear_80 = torch._C._nn.linear( + permute_68, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_68 = l_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_68 = linear_80.permute(0, 3, 1, 2) + linear_80 = None + attn_69 = torch.nn.functional.dropout(attn_68, 0.0, False, False) + attn_68 = None + matmul_27 = attn_69 @ v_13 + attn_69 = v_13 = None + transpose_28 = matmul_27.transpose(1, 2) + matmul_27 = None + x_134 = transpose_28.reshape(1, 196, 192) + transpose_28 = None + x_135 = torch._C._nn.linear( + x_134, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_, + ) + x_134 = l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ + ) = None + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + mul_40 = l_self_modules_blocks_modules_13_parameters_gamma_1_ * x_136 + l_self_modules_blocks_modules_13_parameters_gamma_1_ = x_136 = None + x_137 = x_133 + mul_40 + x_133 = mul_40 = None + layer_norm_27 = torch.nn.functional.layer_norm( + x_137, + (192,), + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ + ) = None + x_138 = torch._C._nn.linear( + layer_norm_27, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_27 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_139 = torch._C._nn.gelu(x_138, approximate="none") + x_138 = None + x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) + x_139 = None + x_141 = torch._C._nn.linear( + x_140, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_140 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + mul_41 = l_self_modules_blocks_modules_13_parameters_gamma_2_ * x_142 + l_self_modules_blocks_modules_13_parameters_gamma_2_ = x_142 = None + x_143 = x_137 + mul_41 + x_137 = mul_41 = None + layer_norm_28 = torch.nn.functional.layer_norm( + x_143, + (192,), + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ + ) = None + linear_84 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_28 = ( + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_28 = linear_84.reshape(1, 196, 3, 4, 48) + linear_84 = None + qkv_14 = reshape_28.permute(2, 0, 3, 1, 4) + reshape_28 = None + getitem_46 = qkv_14[0] + q_14 = getitem_46 * 0.14433756729740643 + getitem_46 = None + k_14 = qkv_14[1] + v_14 = qkv_14[2] + qkv_14 = None + transpose_29 = k_14.transpose(-2, -1) + k_14 = None + attn_70 = q_14 @ transpose_29 + q_14 = transpose_29 = None + permute_71 = attn_70.permute(0, 2, 3, 1) + attn_70 = None + linear_85 = torch._C._nn.linear( + permute_71, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_71 = l_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_71 = linear_85.permute(0, 3, 1, 2) + linear_85 = None + attn_72 = attn_71.softmax(dim=-1) + attn_71 = None + permute_73 = attn_72.permute(0, 2, 3, 1) + attn_72 = None + linear_86 = torch._C._nn.linear( + permute_73, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_73 = l_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_73 = linear_86.permute(0, 3, 1, 2) + linear_86 = None + attn_74 = torch.nn.functional.dropout(attn_73, 0.0, False, False) + attn_73 = None + matmul_29 = attn_74 @ v_14 + attn_74 = v_14 = None + transpose_30 = matmul_29.transpose(1, 2) + matmul_29 = None + x_144 = transpose_30.reshape(1, 196, 192) + transpose_30 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_, + ) + x_144 = l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ + ) = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + mul_43 = l_self_modules_blocks_modules_14_parameters_gamma_1_ * x_146 + l_self_modules_blocks_modules_14_parameters_gamma_1_ = x_146 = None + x_147 = x_143 + mul_43 + x_143 = mul_43 = None + layer_norm_29 = torch.nn.functional.layer_norm( + x_147, + (192,), + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ + ) = None + x_148 = torch._C._nn.linear( + layer_norm_29, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_29 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_149 = torch._C._nn.gelu(x_148, approximate="none") + x_148 = None + x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) + x_149 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_150 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + mul_44 = l_self_modules_blocks_modules_14_parameters_gamma_2_ * x_152 + l_self_modules_blocks_modules_14_parameters_gamma_2_ = x_152 = None + x_153 = x_147 + mul_44 + x_147 = mul_44 = None + layer_norm_30 = torch.nn.functional.layer_norm( + x_153, + (192,), + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ + ) = None + linear_90 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_30 = ( + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_30 = linear_90.reshape(1, 196, 3, 4, 48) + linear_90 = None + qkv_15 = reshape_30.permute(2, 0, 3, 1, 4) + reshape_30 = None + getitem_49 = qkv_15[0] + q_15 = getitem_49 * 0.14433756729740643 + getitem_49 = None + k_15 = qkv_15[1] + v_15 = qkv_15[2] + qkv_15 = None + transpose_31 = k_15.transpose(-2, -1) + k_15 = None + attn_75 = q_15 @ transpose_31 + q_15 = transpose_31 = None + permute_76 = attn_75.permute(0, 2, 3, 1) + attn_75 = None + linear_91 = torch._C._nn.linear( + permute_76, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_76 = l_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_76 = linear_91.permute(0, 3, 1, 2) + linear_91 = None + attn_77 = attn_76.softmax(dim=-1) + attn_76 = None + permute_78 = attn_77.permute(0, 2, 3, 1) + attn_77 = None + linear_92 = torch._C._nn.linear( + permute_78, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_78 = l_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_78 = linear_92.permute(0, 3, 1, 2) + linear_92 = None + attn_79 = torch.nn.functional.dropout(attn_78, 0.0, False, False) + attn_78 = None + matmul_31 = attn_79 @ v_15 + attn_79 = v_15 = None + transpose_32 = matmul_31.transpose(1, 2) + matmul_31 = None + x_154 = transpose_32.reshape(1, 196, 192) + transpose_32 = None + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_, + ) + x_154 = l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ + ) = None + x_156 = torch.nn.functional.dropout(x_155, 0.0, False, False) + x_155 = None + mul_46 = l_self_modules_blocks_modules_15_parameters_gamma_1_ * x_156 + l_self_modules_blocks_modules_15_parameters_gamma_1_ = x_156 = None + x_157 = x_153 + mul_46 + x_153 = mul_46 = None + layer_norm_31 = torch.nn.functional.layer_norm( + x_157, + (192,), + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ + ) = None + x_158 = torch._C._nn.linear( + layer_norm_31, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_31 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_159 = torch._C._nn.gelu(x_158, approximate="none") + x_158 = None + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = torch._C._nn.linear( + x_160, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_160 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_162 = torch.nn.functional.dropout(x_161, 0.0, False, False) + x_161 = None + mul_47 = l_self_modules_blocks_modules_15_parameters_gamma_2_ * x_162 + l_self_modules_blocks_modules_15_parameters_gamma_2_ = x_162 = None + x_163 = x_157 + mul_47 + x_157 = mul_47 = None + layer_norm_32 = torch.nn.functional.layer_norm( + x_163, + (192,), + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ + ) = None + linear_96 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_32 = ( + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_32 = linear_96.reshape(1, 196, 3, 4, 48) + linear_96 = None + qkv_16 = reshape_32.permute(2, 0, 3, 1, 4) + reshape_32 = None + getitem_52 = qkv_16[0] + q_16 = getitem_52 * 0.14433756729740643 + getitem_52 = None + k_16 = qkv_16[1] + v_16 = qkv_16[2] + qkv_16 = None + transpose_33 = k_16.transpose(-2, -1) + k_16 = None + attn_80 = q_16 @ transpose_33 + q_16 = transpose_33 = None + permute_81 = attn_80.permute(0, 2, 3, 1) + attn_80 = None + linear_97 = torch._C._nn.linear( + permute_81, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_81 = l_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_81 = linear_97.permute(0, 3, 1, 2) + linear_97 = None + attn_82 = attn_81.softmax(dim=-1) + attn_81 = None + permute_83 = attn_82.permute(0, 2, 3, 1) + attn_82 = None + linear_98 = torch._C._nn.linear( + permute_83, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_83 = l_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_83 = linear_98.permute(0, 3, 1, 2) + linear_98 = None + attn_84 = torch.nn.functional.dropout(attn_83, 0.0, False, False) + attn_83 = None + matmul_33 = attn_84 @ v_16 + attn_84 = v_16 = None + transpose_34 = matmul_33.transpose(1, 2) + matmul_33 = None + x_164 = transpose_34.reshape(1, 196, 192) + transpose_34 = None + x_165 = torch._C._nn.linear( + x_164, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_, + ) + x_164 = l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ + ) = None + x_166 = torch.nn.functional.dropout(x_165, 0.0, False, False) + x_165 = None + mul_49 = l_self_modules_blocks_modules_16_parameters_gamma_1_ * x_166 + l_self_modules_blocks_modules_16_parameters_gamma_1_ = x_166 = None + x_167 = x_163 + mul_49 + x_163 = mul_49 = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_167, + (192,), + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ + ) = None + x_168 = torch._C._nn.linear( + layer_norm_33, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_33 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_169 = torch._C._nn.gelu(x_168, approximate="none") + x_168 = None + x_170 = torch.nn.functional.dropout(x_169, 0.0, False, False) + x_169 = None + x_171 = torch._C._nn.linear( + x_170, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_170 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_172 = torch.nn.functional.dropout(x_171, 0.0, False, False) + x_171 = None + mul_50 = l_self_modules_blocks_modules_16_parameters_gamma_2_ * x_172 + l_self_modules_blocks_modules_16_parameters_gamma_2_ = x_172 = None + x_173 = x_167 + mul_50 + x_167 = mul_50 = None + layer_norm_34 = torch.nn.functional.layer_norm( + x_173, + (192,), + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ + ) = None + linear_102 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_34 = ( + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_34 = linear_102.reshape(1, 196, 3, 4, 48) + linear_102 = None + qkv_17 = reshape_34.permute(2, 0, 3, 1, 4) + reshape_34 = None + getitem_55 = qkv_17[0] + q_17 = getitem_55 * 0.14433756729740643 + getitem_55 = None + k_17 = qkv_17[1] + v_17 = qkv_17[2] + qkv_17 = None + transpose_35 = k_17.transpose(-2, -1) + k_17 = None + attn_85 = q_17 @ transpose_35 + q_17 = transpose_35 = None + permute_86 = attn_85.permute(0, 2, 3, 1) + attn_85 = None + linear_103 = torch._C._nn.linear( + permute_86, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_86 = l_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_86 = linear_103.permute(0, 3, 1, 2) + linear_103 = None + attn_87 = attn_86.softmax(dim=-1) + attn_86 = None + permute_88 = attn_87.permute(0, 2, 3, 1) + attn_87 = None + linear_104 = torch._C._nn.linear( + permute_88, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_88 = l_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_88 = linear_104.permute(0, 3, 1, 2) + linear_104 = None + attn_89 = torch.nn.functional.dropout(attn_88, 0.0, False, False) + attn_88 = None + matmul_35 = attn_89 @ v_17 + attn_89 = v_17 = None + transpose_36 = matmul_35.transpose(1, 2) + matmul_35 = None + x_174 = transpose_36.reshape(1, 196, 192) + transpose_36 = None + x_175 = torch._C._nn.linear( + x_174, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_, + ) + x_174 = l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ + ) = None + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + mul_52 = l_self_modules_blocks_modules_17_parameters_gamma_1_ * x_176 + l_self_modules_blocks_modules_17_parameters_gamma_1_ = x_176 = None + x_177 = x_173 + mul_52 + x_173 = mul_52 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_177, + (192,), + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ + ) = None + x_178 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_35 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_179 = torch._C._nn.gelu(x_178, approximate="none") + x_178 = None + x_180 = torch.nn.functional.dropout(x_179, 0.0, False, False) + x_179 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + mul_53 = l_self_modules_blocks_modules_17_parameters_gamma_2_ * x_182 + l_self_modules_blocks_modules_17_parameters_gamma_2_ = x_182 = None + x_183 = x_177 + mul_53 + x_177 = mul_53 = None + layer_norm_36 = torch.nn.functional.layer_norm( + x_183, + (192,), + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ + ) = None + linear_108 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_36 = ( + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_36 = linear_108.reshape(1, 196, 3, 4, 48) + linear_108 = None + qkv_18 = reshape_36.permute(2, 0, 3, 1, 4) + reshape_36 = None + getitem_58 = qkv_18[0] + q_18 = getitem_58 * 0.14433756729740643 + getitem_58 = None + k_18 = qkv_18[1] + v_18 = qkv_18[2] + qkv_18 = None + transpose_37 = k_18.transpose(-2, -1) + k_18 = None + attn_90 = q_18 @ transpose_37 + q_18 = transpose_37 = None + permute_91 = attn_90.permute(0, 2, 3, 1) + attn_90 = None + linear_109 = torch._C._nn.linear( + permute_91, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_91 = l_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_91 = linear_109.permute(0, 3, 1, 2) + linear_109 = None + attn_92 = attn_91.softmax(dim=-1) + attn_91 = None + permute_93 = attn_92.permute(0, 2, 3, 1) + attn_92 = None + linear_110 = torch._C._nn.linear( + permute_93, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_93 = l_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_93 = linear_110.permute(0, 3, 1, 2) + linear_110 = None + attn_94 = torch.nn.functional.dropout(attn_93, 0.0, False, False) + attn_93 = None + matmul_37 = attn_94 @ v_18 + attn_94 = v_18 = None + transpose_38 = matmul_37.transpose(1, 2) + matmul_37 = None + x_184 = transpose_38.reshape(1, 196, 192) + transpose_38 = None + x_185 = torch._C._nn.linear( + x_184, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_, + ) + x_184 = l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ + ) = None + x_186 = torch.nn.functional.dropout(x_185, 0.0, False, False) + x_185 = None + mul_55 = l_self_modules_blocks_modules_18_parameters_gamma_1_ * x_186 + l_self_modules_blocks_modules_18_parameters_gamma_1_ = x_186 = None + x_187 = x_183 + mul_55 + x_183 = mul_55 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_187, + (192,), + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ + ) = None + x_188 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_37 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_189 = torch._C._nn.gelu(x_188, approximate="none") + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = torch._C._nn.linear( + x_190, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_190 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_192 = torch.nn.functional.dropout(x_191, 0.0, False, False) + x_191 = None + mul_56 = l_self_modules_blocks_modules_18_parameters_gamma_2_ * x_192 + l_self_modules_blocks_modules_18_parameters_gamma_2_ = x_192 = None + x_193 = x_187 + mul_56 + x_187 = mul_56 = None + layer_norm_38 = torch.nn.functional.layer_norm( + x_193, + (192,), + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ + ) = None + linear_114 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_38 = ( + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_38 = linear_114.reshape(1, 196, 3, 4, 48) + linear_114 = None + qkv_19 = reshape_38.permute(2, 0, 3, 1, 4) + reshape_38 = None + getitem_61 = qkv_19[0] + q_19 = getitem_61 * 0.14433756729740643 + getitem_61 = None + k_19 = qkv_19[1] + v_19 = qkv_19[2] + qkv_19 = None + transpose_39 = k_19.transpose(-2, -1) + k_19 = None + attn_95 = q_19 @ transpose_39 + q_19 = transpose_39 = None + permute_96 = attn_95.permute(0, 2, 3, 1) + attn_95 = None + linear_115 = torch._C._nn.linear( + permute_96, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_96 = l_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_96 = linear_115.permute(0, 3, 1, 2) + linear_115 = None + attn_97 = attn_96.softmax(dim=-1) + attn_96 = None + permute_98 = attn_97.permute(0, 2, 3, 1) + attn_97 = None + linear_116 = torch._C._nn.linear( + permute_98, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_98 = l_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_98 = linear_116.permute(0, 3, 1, 2) + linear_116 = None + attn_99 = torch.nn.functional.dropout(attn_98, 0.0, False, False) + attn_98 = None + matmul_39 = attn_99 @ v_19 + attn_99 = v_19 = None + transpose_40 = matmul_39.transpose(1, 2) + matmul_39 = None + x_194 = transpose_40.reshape(1, 196, 192) + transpose_40 = None + x_195 = torch._C._nn.linear( + x_194, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_, + ) + x_194 = l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ + ) = None + x_196 = torch.nn.functional.dropout(x_195, 0.0, False, False) + x_195 = None + mul_58 = l_self_modules_blocks_modules_19_parameters_gamma_1_ * x_196 + l_self_modules_blocks_modules_19_parameters_gamma_1_ = x_196 = None + x_197 = x_193 + mul_58 + x_193 = mul_58 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_197, + (192,), + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ + ) = None + x_198 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_39 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_199 = torch._C._nn.gelu(x_198, approximate="none") + x_198 = None + x_200 = torch.nn.functional.dropout(x_199, 0.0, False, False) + x_199 = None + x_201 = torch._C._nn.linear( + x_200, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_200 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_202 = torch.nn.functional.dropout(x_201, 0.0, False, False) + x_201 = None + mul_59 = l_self_modules_blocks_modules_19_parameters_gamma_2_ * x_202 + l_self_modules_blocks_modules_19_parameters_gamma_2_ = x_202 = None + x_203 = x_197 + mul_59 + x_197 = mul_59 = None + layer_norm_40 = torch.nn.functional.layer_norm( + x_203, + (192,), + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ + ) = None + linear_120 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_40 = ( + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_40 = linear_120.reshape(1, 196, 3, 4, 48) + linear_120 = None + qkv_20 = reshape_40.permute(2, 0, 3, 1, 4) + reshape_40 = None + getitem_64 = qkv_20[0] + q_20 = getitem_64 * 0.14433756729740643 + getitem_64 = None + k_20 = qkv_20[1] + v_20 = qkv_20[2] + qkv_20 = None + transpose_41 = k_20.transpose(-2, -1) + k_20 = None + attn_100 = q_20 @ transpose_41 + q_20 = transpose_41 = None + permute_101 = attn_100.permute(0, 2, 3, 1) + attn_100 = None + linear_121 = torch._C._nn.linear( + permute_101, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_101 = l_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_101 = linear_121.permute(0, 3, 1, 2) + linear_121 = None + attn_102 = attn_101.softmax(dim=-1) + attn_101 = None + permute_103 = attn_102.permute(0, 2, 3, 1) + attn_102 = None + linear_122 = torch._C._nn.linear( + permute_103, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_103 = l_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_103 = linear_122.permute(0, 3, 1, 2) + linear_122 = None + attn_104 = torch.nn.functional.dropout(attn_103, 0.0, False, False) + attn_103 = None + matmul_41 = attn_104 @ v_20 + attn_104 = v_20 = None + transpose_42 = matmul_41.transpose(1, 2) + matmul_41 = None + x_204 = transpose_42.reshape(1, 196, 192) + transpose_42 = None + x_205 = torch._C._nn.linear( + x_204, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_, + ) + x_204 = l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ + ) = None + x_206 = torch.nn.functional.dropout(x_205, 0.0, False, False) + x_205 = None + mul_61 = l_self_modules_blocks_modules_20_parameters_gamma_1_ * x_206 + l_self_modules_blocks_modules_20_parameters_gamma_1_ = x_206 = None + x_207 = x_203 + mul_61 + x_203 = mul_61 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_207, + (192,), + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ + ) = None + x_208 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_41 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_209 = torch._C._nn.gelu(x_208, approximate="none") + x_208 = None + x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) + x_209 = None + x_211 = torch._C._nn.linear( + x_210, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_210 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + mul_62 = l_self_modules_blocks_modules_20_parameters_gamma_2_ * x_212 + l_self_modules_blocks_modules_20_parameters_gamma_2_ = x_212 = None + x_213 = x_207 + mul_62 + x_207 = mul_62 = None + layer_norm_42 = torch.nn.functional.layer_norm( + x_213, + (192,), + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ + ) = None + linear_126 = torch._C._nn.linear( + layer_norm_42, + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_42 = ( + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_42 = linear_126.reshape(1, 196, 3, 4, 48) + linear_126 = None + qkv_21 = reshape_42.permute(2, 0, 3, 1, 4) + reshape_42 = None + getitem_67 = qkv_21[0] + q_21 = getitem_67 * 0.14433756729740643 + getitem_67 = None + k_21 = qkv_21[1] + v_21 = qkv_21[2] + qkv_21 = None + transpose_43 = k_21.transpose(-2, -1) + k_21 = None + attn_105 = q_21 @ transpose_43 + q_21 = transpose_43 = None + permute_106 = attn_105.permute(0, 2, 3, 1) + attn_105 = None + linear_127 = torch._C._nn.linear( + permute_106, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_106 = l_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_106 = linear_127.permute(0, 3, 1, 2) + linear_127 = None + attn_107 = attn_106.softmax(dim=-1) + attn_106 = None + permute_108 = attn_107.permute(0, 2, 3, 1) + attn_107 = None + linear_128 = torch._C._nn.linear( + permute_108, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_108 = l_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_108 = linear_128.permute(0, 3, 1, 2) + linear_128 = None + attn_109 = torch.nn.functional.dropout(attn_108, 0.0, False, False) + attn_108 = None + matmul_43 = attn_109 @ v_21 + attn_109 = v_21 = None + transpose_44 = matmul_43.transpose(1, 2) + matmul_43 = None + x_214 = transpose_44.reshape(1, 196, 192) + transpose_44 = None + x_215 = torch._C._nn.linear( + x_214, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_, + ) + x_214 = l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ + ) = None + x_216 = torch.nn.functional.dropout(x_215, 0.0, False, False) + x_215 = None + mul_64 = l_self_modules_blocks_modules_21_parameters_gamma_1_ * x_216 + l_self_modules_blocks_modules_21_parameters_gamma_1_ = x_216 = None + x_217 = x_213 + mul_64 + x_213 = mul_64 = None + layer_norm_43 = torch.nn.functional.layer_norm( + x_217, + (192,), + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ + ) = None + x_218 = torch._C._nn.linear( + layer_norm_43, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_43 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_219 = torch._C._nn.gelu(x_218, approximate="none") + x_218 = None + x_220 = torch.nn.functional.dropout(x_219, 0.0, False, False) + x_219 = None + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_220 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_222 = torch.nn.functional.dropout(x_221, 0.0, False, False) + x_221 = None + mul_65 = l_self_modules_blocks_modules_21_parameters_gamma_2_ * x_222 + l_self_modules_blocks_modules_21_parameters_gamma_2_ = x_222 = None + x_223 = x_217 + mul_65 + x_217 = mul_65 = None + layer_norm_44 = torch.nn.functional.layer_norm( + x_223, + (192,), + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ + ) = None + linear_132 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_44 = ( + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_44 = linear_132.reshape(1, 196, 3, 4, 48) + linear_132 = None + qkv_22 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + getitem_70 = qkv_22[0] + q_22 = getitem_70 * 0.14433756729740643 + getitem_70 = None + k_22 = qkv_22[1] + v_22 = qkv_22[2] + qkv_22 = None + transpose_45 = k_22.transpose(-2, -1) + k_22 = None + attn_110 = q_22 @ transpose_45 + q_22 = transpose_45 = None + permute_111 = attn_110.permute(0, 2, 3, 1) + attn_110 = None + linear_133 = torch._C._nn.linear( + permute_111, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_111 = l_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_111 = linear_133.permute(0, 3, 1, 2) + linear_133 = None + attn_112 = attn_111.softmax(dim=-1) + attn_111 = None + permute_113 = attn_112.permute(0, 2, 3, 1) + attn_112 = None + linear_134 = torch._C._nn.linear( + permute_113, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_113 = l_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_113 = linear_134.permute(0, 3, 1, 2) + linear_134 = None + attn_114 = torch.nn.functional.dropout(attn_113, 0.0, False, False) + attn_113 = None + matmul_45 = attn_114 @ v_22 + attn_114 = v_22 = None + transpose_46 = matmul_45.transpose(1, 2) + matmul_45 = None + x_224 = transpose_46.reshape(1, 196, 192) + transpose_46 = None + x_225 = torch._C._nn.linear( + x_224, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_, + ) + x_224 = l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ + ) = None + x_226 = torch.nn.functional.dropout(x_225, 0.0, False, False) + x_225 = None + mul_67 = l_self_modules_blocks_modules_22_parameters_gamma_1_ * x_226 + l_self_modules_blocks_modules_22_parameters_gamma_1_ = x_226 = None + x_227 = x_223 + mul_67 + x_223 = mul_67 = None + layer_norm_45 = torch.nn.functional.layer_norm( + x_227, + (192,), + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ + ) = None + x_228 = torch._C._nn.linear( + layer_norm_45, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_45 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_229 = torch._C._nn.gelu(x_228, approximate="none") + x_228 = None + x_230 = torch.nn.functional.dropout(x_229, 0.0, False, False) + x_229 = None + x_231 = torch._C._nn.linear( + x_230, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_230 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_232 = torch.nn.functional.dropout(x_231, 0.0, False, False) + x_231 = None + mul_68 = l_self_modules_blocks_modules_22_parameters_gamma_2_ * x_232 + l_self_modules_blocks_modules_22_parameters_gamma_2_ = x_232 = None + x_233 = x_227 + mul_68 + x_227 = mul_68 = None + layer_norm_46 = torch.nn.functional.layer_norm( + x_233, + (192,), + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ + ) = None + linear_138 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_46 = ( + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_46 = linear_138.reshape(1, 196, 3, 4, 48) + linear_138 = None + qkv_23 = reshape_46.permute(2, 0, 3, 1, 4) + reshape_46 = None + getitem_73 = qkv_23[0] + q_23 = getitem_73 * 0.14433756729740643 + getitem_73 = None + k_23 = qkv_23[1] + v_23 = qkv_23[2] + qkv_23 = None + transpose_47 = k_23.transpose(-2, -1) + k_23 = None + attn_115 = q_23 @ transpose_47 + q_23 = transpose_47 = None + permute_116 = attn_115.permute(0, 2, 3, 1) + attn_115 = None + linear_139 = torch._C._nn.linear( + permute_116, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_116 = l_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_116 = linear_139.permute(0, 3, 1, 2) + linear_139 = None + attn_117 = attn_116.softmax(dim=-1) + attn_116 = None + permute_118 = attn_117.permute(0, 2, 3, 1) + attn_117 = None + linear_140 = torch._C._nn.linear( + permute_118, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_118 = l_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_118 = linear_140.permute(0, 3, 1, 2) + linear_140 = None + attn_119 = torch.nn.functional.dropout(attn_118, 0.0, False, False) + attn_118 = None + matmul_47 = attn_119 @ v_23 + attn_119 = v_23 = None + transpose_48 = matmul_47.transpose(1, 2) + matmul_47 = None + x_234 = transpose_48.reshape(1, 196, 192) + transpose_48 = None + x_235 = torch._C._nn.linear( + x_234, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_, + ) + x_234 = l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ + ) = None + x_236 = torch.nn.functional.dropout(x_235, 0.0, False, False) + x_235 = None + mul_70 = l_self_modules_blocks_modules_23_parameters_gamma_1_ * x_236 + l_self_modules_blocks_modules_23_parameters_gamma_1_ = x_236 = None + x_237 = x_233 + mul_70 + x_233 = mul_70 = None + layer_norm_47 = torch.nn.functional.layer_norm( + x_237, + (192,), + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ + ) = None + x_238 = torch._C._nn.linear( + layer_norm_47, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_47 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_239 = torch._C._nn.gelu(x_238, approximate="none") + x_238 = None + x_240 = torch.nn.functional.dropout(x_239, 0.0, False, False) + x_239 = None + x_241 = torch._C._nn.linear( + x_240, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_240 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) + x_241 = None + mul_71 = l_self_modules_blocks_modules_23_parameters_gamma_2_ * x_242 + l_self_modules_blocks_modules_23_parameters_gamma_2_ = x_242 = None + x_243 = x_237 + mul_71 + x_237 = mul_71 = None + layer_norm_48 = torch.nn.functional.layer_norm( + x_243, + (192,), + l_self_modules_blocks_modules_24_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_24_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_24_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_24_modules_norm1_parameters_bias_ + ) = None + linear_144 = torch._C._nn.linear( + layer_norm_48, + l_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_48 = ( + l_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_48 = linear_144.reshape(1, 196, 3, 4, 48) + linear_144 = None + qkv_24 = reshape_48.permute(2, 0, 3, 1, 4) + reshape_48 = None + getitem_76 = qkv_24[0] + q_24 = getitem_76 * 0.14433756729740643 + getitem_76 = None + k_24 = qkv_24[1] + v_24 = qkv_24[2] + qkv_24 = None + transpose_49 = k_24.transpose(-2, -1) + k_24 = None + attn_120 = q_24 @ transpose_49 + q_24 = transpose_49 = None + permute_121 = attn_120.permute(0, 2, 3, 1) + attn_120 = None + linear_145 = torch._C._nn.linear( + permute_121, + l_self_modules_blocks_modules_24_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_24_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_121 = l_self_modules_blocks_modules_24_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_24_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_121 = linear_145.permute(0, 3, 1, 2) + linear_145 = None + attn_122 = attn_121.softmax(dim=-1) + attn_121 = None + permute_123 = attn_122.permute(0, 2, 3, 1) + attn_122 = None + linear_146 = torch._C._nn.linear( + permute_123, + l_self_modules_blocks_modules_24_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_24_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_123 = l_self_modules_blocks_modules_24_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_24_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_123 = linear_146.permute(0, 3, 1, 2) + linear_146 = None + attn_124 = torch.nn.functional.dropout(attn_123, 0.0, False, False) + attn_123 = None + matmul_49 = attn_124 @ v_24 + attn_124 = v_24 = None + transpose_50 = matmul_49.transpose(1, 2) + matmul_49 = None + x_244 = transpose_50.reshape(1, 196, 192) + transpose_50 = None + x_245 = torch._C._nn.linear( + x_244, + l_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_bias_, + ) + x_244 = l_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_bias_ + ) = None + x_246 = torch.nn.functional.dropout(x_245, 0.0, False, False) + x_245 = None + mul_73 = l_self_modules_blocks_modules_24_parameters_gamma_1_ * x_246 + l_self_modules_blocks_modules_24_parameters_gamma_1_ = x_246 = None + x_247 = x_243 + mul_73 + x_243 = mul_73 = None + layer_norm_49 = torch.nn.functional.layer_norm( + x_247, + (192,), + l_self_modules_blocks_modules_24_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_24_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_24_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_24_modules_norm2_parameters_bias_ + ) = None + x_248 = torch._C._nn.linear( + layer_norm_49, + l_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_49 = ( + l_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_249 = torch._C._nn.gelu(x_248, approximate="none") + x_248 = None + x_250 = torch.nn.functional.dropout(x_249, 0.0, False, False) + x_249 = None + x_251 = torch._C._nn.linear( + x_250, + l_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_250 = ( + l_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_252 = torch.nn.functional.dropout(x_251, 0.0, False, False) + x_251 = None + mul_74 = l_self_modules_blocks_modules_24_parameters_gamma_2_ * x_252 + l_self_modules_blocks_modules_24_parameters_gamma_2_ = x_252 = None + x_253 = x_247 + mul_74 + x_247 = mul_74 = None + layer_norm_50 = torch.nn.functional.layer_norm( + x_253, + (192,), + l_self_modules_blocks_modules_25_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_25_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_25_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_25_modules_norm1_parameters_bias_ + ) = None + linear_150 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_50 = ( + l_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_50 = linear_150.reshape(1, 196, 3, 4, 48) + linear_150 = None + qkv_25 = reshape_50.permute(2, 0, 3, 1, 4) + reshape_50 = None + getitem_79 = qkv_25[0] + q_25 = getitem_79 * 0.14433756729740643 + getitem_79 = None + k_25 = qkv_25[1] + v_25 = qkv_25[2] + qkv_25 = None + transpose_51 = k_25.transpose(-2, -1) + k_25 = None + attn_125 = q_25 @ transpose_51 + q_25 = transpose_51 = None + permute_126 = attn_125.permute(0, 2, 3, 1) + attn_125 = None + linear_151 = torch._C._nn.linear( + permute_126, + l_self_modules_blocks_modules_25_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_25_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_126 = l_self_modules_blocks_modules_25_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_25_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_126 = linear_151.permute(0, 3, 1, 2) + linear_151 = None + attn_127 = attn_126.softmax(dim=-1) + attn_126 = None + permute_128 = attn_127.permute(0, 2, 3, 1) + attn_127 = None + linear_152 = torch._C._nn.linear( + permute_128, + l_self_modules_blocks_modules_25_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_25_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_128 = l_self_modules_blocks_modules_25_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_25_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_128 = linear_152.permute(0, 3, 1, 2) + linear_152 = None + attn_129 = torch.nn.functional.dropout(attn_128, 0.0, False, False) + attn_128 = None + matmul_51 = attn_129 @ v_25 + attn_129 = v_25 = None + transpose_52 = matmul_51.transpose(1, 2) + matmul_51 = None + x_254 = transpose_52.reshape(1, 196, 192) + transpose_52 = None + x_255 = torch._C._nn.linear( + x_254, + l_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_bias_, + ) + x_254 = l_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_bias_ + ) = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + mul_76 = l_self_modules_blocks_modules_25_parameters_gamma_1_ * x_256 + l_self_modules_blocks_modules_25_parameters_gamma_1_ = x_256 = None + x_257 = x_253 + mul_76 + x_253 = mul_76 = None + layer_norm_51 = torch.nn.functional.layer_norm( + x_257, + (192,), + l_self_modules_blocks_modules_25_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_25_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_25_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_25_modules_norm2_parameters_bias_ + ) = None + x_258 = torch._C._nn.linear( + layer_norm_51, + l_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_51 = ( + l_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_259 = torch._C._nn.gelu(x_258, approximate="none") + x_258 = None + x_260 = torch.nn.functional.dropout(x_259, 0.0, False, False) + x_259 = None + x_261 = torch._C._nn.linear( + x_260, + l_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_260 = ( + l_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_262 = torch.nn.functional.dropout(x_261, 0.0, False, False) + x_261 = None + mul_77 = l_self_modules_blocks_modules_25_parameters_gamma_2_ * x_262 + l_self_modules_blocks_modules_25_parameters_gamma_2_ = x_262 = None + x_263 = x_257 + mul_77 + x_257 = mul_77 = None + layer_norm_52 = torch.nn.functional.layer_norm( + x_263, + (192,), + l_self_modules_blocks_modules_26_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_26_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_26_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_26_modules_norm1_parameters_bias_ + ) = None + linear_156 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_52 = ( + l_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_52 = linear_156.reshape(1, 196, 3, 4, 48) + linear_156 = None + qkv_26 = reshape_52.permute(2, 0, 3, 1, 4) + reshape_52 = None + getitem_82 = qkv_26[0] + q_26 = getitem_82 * 0.14433756729740643 + getitem_82 = None + k_26 = qkv_26[1] + v_26 = qkv_26[2] + qkv_26 = None + transpose_53 = k_26.transpose(-2, -1) + k_26 = None + attn_130 = q_26 @ transpose_53 + q_26 = transpose_53 = None + permute_131 = attn_130.permute(0, 2, 3, 1) + attn_130 = None + linear_157 = torch._C._nn.linear( + permute_131, + l_self_modules_blocks_modules_26_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_26_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_131 = l_self_modules_blocks_modules_26_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_26_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_131 = linear_157.permute(0, 3, 1, 2) + linear_157 = None + attn_132 = attn_131.softmax(dim=-1) + attn_131 = None + permute_133 = attn_132.permute(0, 2, 3, 1) + attn_132 = None + linear_158 = torch._C._nn.linear( + permute_133, + l_self_modules_blocks_modules_26_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_26_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_133 = l_self_modules_blocks_modules_26_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_26_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_133 = linear_158.permute(0, 3, 1, 2) + linear_158 = None + attn_134 = torch.nn.functional.dropout(attn_133, 0.0, False, False) + attn_133 = None + matmul_53 = attn_134 @ v_26 + attn_134 = v_26 = None + transpose_54 = matmul_53.transpose(1, 2) + matmul_53 = None + x_264 = transpose_54.reshape(1, 196, 192) + transpose_54 = None + x_265 = torch._C._nn.linear( + x_264, + l_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_bias_, + ) + x_264 = l_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_bias_ + ) = None + x_266 = torch.nn.functional.dropout(x_265, 0.0, False, False) + x_265 = None + mul_79 = l_self_modules_blocks_modules_26_parameters_gamma_1_ * x_266 + l_self_modules_blocks_modules_26_parameters_gamma_1_ = x_266 = None + x_267 = x_263 + mul_79 + x_263 = mul_79 = None + layer_norm_53 = torch.nn.functional.layer_norm( + x_267, + (192,), + l_self_modules_blocks_modules_26_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_26_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_26_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_26_modules_norm2_parameters_bias_ + ) = None + x_268 = torch._C._nn.linear( + layer_norm_53, + l_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_53 = ( + l_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_269 = torch._C._nn.gelu(x_268, approximate="none") + x_268 = None + x_270 = torch.nn.functional.dropout(x_269, 0.0, False, False) + x_269 = None + x_271 = torch._C._nn.linear( + x_270, + l_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_270 = ( + l_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_272 = torch.nn.functional.dropout(x_271, 0.0, False, False) + x_271 = None + mul_80 = l_self_modules_blocks_modules_26_parameters_gamma_2_ * x_272 + l_self_modules_blocks_modules_26_parameters_gamma_2_ = x_272 = None + x_273 = x_267 + mul_80 + x_267 = mul_80 = None + layer_norm_54 = torch.nn.functional.layer_norm( + x_273, + (192,), + l_self_modules_blocks_modules_27_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_27_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_27_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_27_modules_norm1_parameters_bias_ + ) = None + linear_162 = torch._C._nn.linear( + layer_norm_54, + l_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_54 = ( + l_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_54 = linear_162.reshape(1, 196, 3, 4, 48) + linear_162 = None + qkv_27 = reshape_54.permute(2, 0, 3, 1, 4) + reshape_54 = None + getitem_85 = qkv_27[0] + q_27 = getitem_85 * 0.14433756729740643 + getitem_85 = None + k_27 = qkv_27[1] + v_27 = qkv_27[2] + qkv_27 = None + transpose_55 = k_27.transpose(-2, -1) + k_27 = None + attn_135 = q_27 @ transpose_55 + q_27 = transpose_55 = None + permute_136 = attn_135.permute(0, 2, 3, 1) + attn_135 = None + linear_163 = torch._C._nn.linear( + permute_136, + l_self_modules_blocks_modules_27_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_27_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_136 = l_self_modules_blocks_modules_27_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_27_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_136 = linear_163.permute(0, 3, 1, 2) + linear_163 = None + attn_137 = attn_136.softmax(dim=-1) + attn_136 = None + permute_138 = attn_137.permute(0, 2, 3, 1) + attn_137 = None + linear_164 = torch._C._nn.linear( + permute_138, + l_self_modules_blocks_modules_27_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_27_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_138 = l_self_modules_blocks_modules_27_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_27_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_138 = linear_164.permute(0, 3, 1, 2) + linear_164 = None + attn_139 = torch.nn.functional.dropout(attn_138, 0.0, False, False) + attn_138 = None + matmul_55 = attn_139 @ v_27 + attn_139 = v_27 = None + transpose_56 = matmul_55.transpose(1, 2) + matmul_55 = None + x_274 = transpose_56.reshape(1, 196, 192) + transpose_56 = None + x_275 = torch._C._nn.linear( + x_274, + l_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_bias_, + ) + x_274 = l_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_bias_ + ) = None + x_276 = torch.nn.functional.dropout(x_275, 0.0, False, False) + x_275 = None + mul_82 = l_self_modules_blocks_modules_27_parameters_gamma_1_ * x_276 + l_self_modules_blocks_modules_27_parameters_gamma_1_ = x_276 = None + x_277 = x_273 + mul_82 + x_273 = mul_82 = None + layer_norm_55 = torch.nn.functional.layer_norm( + x_277, + (192,), + l_self_modules_blocks_modules_27_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_27_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_27_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_27_modules_norm2_parameters_bias_ + ) = None + x_278 = torch._C._nn.linear( + layer_norm_55, + l_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_55 = ( + l_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_279 = torch._C._nn.gelu(x_278, approximate="none") + x_278 = None + x_280 = torch.nn.functional.dropout(x_279, 0.0, False, False) + x_279 = None + x_281 = torch._C._nn.linear( + x_280, + l_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_, + ) + x_280 = ( + l_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) + x_281 = None + mul_83 = l_self_modules_blocks_modules_27_parameters_gamma_2_ * x_282 + l_self_modules_blocks_modules_27_parameters_gamma_2_ = x_282 = None + x_283 = x_277 + mul_83 + x_277 = mul_83 = None + layer_norm_56 = torch.nn.functional.layer_norm( + x_283, + (192,), + l_self_modules_blocks_modules_28_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_28_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_28_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_28_modules_norm1_parameters_bias_ + ) = None + linear_168 = torch._C._nn.linear( + layer_norm_56, + l_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_56 = ( + l_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_56 = linear_168.reshape(1, 196, 3, 4, 48) + linear_168 = None + qkv_28 = reshape_56.permute(2, 0, 3, 1, 4) + reshape_56 = None + getitem_88 = qkv_28[0] + q_28 = getitem_88 * 0.14433756729740643 + getitem_88 = None + k_28 = qkv_28[1] + v_28 = qkv_28[2] + qkv_28 = None + transpose_57 = k_28.transpose(-2, -1) + k_28 = None + attn_140 = q_28 @ transpose_57 + q_28 = transpose_57 = None + permute_141 = attn_140.permute(0, 2, 3, 1) + attn_140 = None + linear_169 = torch._C._nn.linear( + permute_141, + l_self_modules_blocks_modules_28_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_28_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_141 = l_self_modules_blocks_modules_28_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_28_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_141 = linear_169.permute(0, 3, 1, 2) + linear_169 = None + attn_142 = attn_141.softmax(dim=-1) + attn_141 = None + permute_143 = attn_142.permute(0, 2, 3, 1) + attn_142 = None + linear_170 = torch._C._nn.linear( + permute_143, + l_self_modules_blocks_modules_28_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_28_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_143 = l_self_modules_blocks_modules_28_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_28_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_143 = linear_170.permute(0, 3, 1, 2) + linear_170 = None + attn_144 = torch.nn.functional.dropout(attn_143, 0.0, False, False) + attn_143 = None + matmul_57 = attn_144 @ v_28 + attn_144 = v_28 = None + transpose_58 = matmul_57.transpose(1, 2) + matmul_57 = None + x_284 = transpose_58.reshape(1, 196, 192) + transpose_58 = None + x_285 = torch._C._nn.linear( + x_284, + l_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_bias_, + ) + x_284 = l_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_bias_ + ) = None + x_286 = torch.nn.functional.dropout(x_285, 0.0, False, False) + x_285 = None + mul_85 = l_self_modules_blocks_modules_28_parameters_gamma_1_ * x_286 + l_self_modules_blocks_modules_28_parameters_gamma_1_ = x_286 = None + x_287 = x_283 + mul_85 + x_283 = mul_85 = None + layer_norm_57 = torch.nn.functional.layer_norm( + x_287, + (192,), + l_self_modules_blocks_modules_28_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_28_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_28_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_28_modules_norm2_parameters_bias_ + ) = None + x_288 = torch._C._nn.linear( + layer_norm_57, + l_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_57 = ( + l_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_289 = torch._C._nn.gelu(x_288, approximate="none") + x_288 = None + x_290 = torch.nn.functional.dropout(x_289, 0.0, False, False) + x_289 = None + x_291 = torch._C._nn.linear( + x_290, + l_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_, + ) + x_290 = ( + l_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_292 = torch.nn.functional.dropout(x_291, 0.0, False, False) + x_291 = None + mul_86 = l_self_modules_blocks_modules_28_parameters_gamma_2_ * x_292 + l_self_modules_blocks_modules_28_parameters_gamma_2_ = x_292 = None + x_293 = x_287 + mul_86 + x_287 = mul_86 = None + layer_norm_58 = torch.nn.functional.layer_norm( + x_293, + (192,), + l_self_modules_blocks_modules_29_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_29_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_29_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_29_modules_norm1_parameters_bias_ + ) = None + linear_174 = torch._C._nn.linear( + layer_norm_58, + l_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_58 = ( + l_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_58 = linear_174.reshape(1, 196, 3, 4, 48) + linear_174 = None + qkv_29 = reshape_58.permute(2, 0, 3, 1, 4) + reshape_58 = None + getitem_91 = qkv_29[0] + q_29 = getitem_91 * 0.14433756729740643 + getitem_91 = None + k_29 = qkv_29[1] + v_29 = qkv_29[2] + qkv_29 = None + transpose_59 = k_29.transpose(-2, -1) + k_29 = None + attn_145 = q_29 @ transpose_59 + q_29 = transpose_59 = None + permute_146 = attn_145.permute(0, 2, 3, 1) + attn_145 = None + linear_175 = torch._C._nn.linear( + permute_146, + l_self_modules_blocks_modules_29_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_29_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_146 = l_self_modules_blocks_modules_29_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_29_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_146 = linear_175.permute(0, 3, 1, 2) + linear_175 = None + attn_147 = attn_146.softmax(dim=-1) + attn_146 = None + permute_148 = attn_147.permute(0, 2, 3, 1) + attn_147 = None + linear_176 = torch._C._nn.linear( + permute_148, + l_self_modules_blocks_modules_29_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_29_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_148 = l_self_modules_blocks_modules_29_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_29_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_148 = linear_176.permute(0, 3, 1, 2) + linear_176 = None + attn_149 = torch.nn.functional.dropout(attn_148, 0.0, False, False) + attn_148 = None + matmul_59 = attn_149 @ v_29 + attn_149 = v_29 = None + transpose_60 = matmul_59.transpose(1, 2) + matmul_59 = None + x_294 = transpose_60.reshape(1, 196, 192) + transpose_60 = None + x_295 = torch._C._nn.linear( + x_294, + l_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_bias_, + ) + x_294 = l_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_bias_ + ) = None + x_296 = torch.nn.functional.dropout(x_295, 0.0, False, False) + x_295 = None + mul_88 = l_self_modules_blocks_modules_29_parameters_gamma_1_ * x_296 + l_self_modules_blocks_modules_29_parameters_gamma_1_ = x_296 = None + x_297 = x_293 + mul_88 + x_293 = mul_88 = None + layer_norm_59 = torch.nn.functional.layer_norm( + x_297, + (192,), + l_self_modules_blocks_modules_29_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_29_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_29_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_29_modules_norm2_parameters_bias_ + ) = None + x_298 = torch._C._nn.linear( + layer_norm_59, + l_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_59 = ( + l_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_299 = torch._C._nn.gelu(x_298, approximate="none") + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_301 = torch._C._nn.linear( + x_300, + l_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_, + ) + x_300 = ( + l_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_302 = torch.nn.functional.dropout(x_301, 0.0, False, False) + x_301 = None + mul_89 = l_self_modules_blocks_modules_29_parameters_gamma_2_ * x_302 + l_self_modules_blocks_modules_29_parameters_gamma_2_ = x_302 = None + x_303 = x_297 + mul_89 + x_297 = mul_89 = None + layer_norm_60 = torch.nn.functional.layer_norm( + x_303, + (192,), + l_self_modules_blocks_modules_30_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_30_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_30_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_30_modules_norm1_parameters_bias_ + ) = None + linear_180 = torch._C._nn.linear( + layer_norm_60, + l_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_60 = ( + l_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_60 = linear_180.reshape(1, 196, 3, 4, 48) + linear_180 = None + qkv_30 = reshape_60.permute(2, 0, 3, 1, 4) + reshape_60 = None + getitem_94 = qkv_30[0] + q_30 = getitem_94 * 0.14433756729740643 + getitem_94 = None + k_30 = qkv_30[1] + v_30 = qkv_30[2] + qkv_30 = None + transpose_61 = k_30.transpose(-2, -1) + k_30 = None + attn_150 = q_30 @ transpose_61 + q_30 = transpose_61 = None + permute_151 = attn_150.permute(0, 2, 3, 1) + attn_150 = None + linear_181 = torch._C._nn.linear( + permute_151, + l_self_modules_blocks_modules_30_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_30_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_151 = l_self_modules_blocks_modules_30_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_30_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_151 = linear_181.permute(0, 3, 1, 2) + linear_181 = None + attn_152 = attn_151.softmax(dim=-1) + attn_151 = None + permute_153 = attn_152.permute(0, 2, 3, 1) + attn_152 = None + linear_182 = torch._C._nn.linear( + permute_153, + l_self_modules_blocks_modules_30_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_30_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_153 = l_self_modules_blocks_modules_30_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_30_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_153 = linear_182.permute(0, 3, 1, 2) + linear_182 = None + attn_154 = torch.nn.functional.dropout(attn_153, 0.0, False, False) + attn_153 = None + matmul_61 = attn_154 @ v_30 + attn_154 = v_30 = None + transpose_62 = matmul_61.transpose(1, 2) + matmul_61 = None + x_304 = transpose_62.reshape(1, 196, 192) + transpose_62 = None + x_305 = torch._C._nn.linear( + x_304, + l_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_bias_, + ) + x_304 = l_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_bias_ + ) = None + x_306 = torch.nn.functional.dropout(x_305, 0.0, False, False) + x_305 = None + mul_91 = l_self_modules_blocks_modules_30_parameters_gamma_1_ * x_306 + l_self_modules_blocks_modules_30_parameters_gamma_1_ = x_306 = None + x_307 = x_303 + mul_91 + x_303 = mul_91 = None + layer_norm_61 = torch.nn.functional.layer_norm( + x_307, + (192,), + l_self_modules_blocks_modules_30_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_30_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_30_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_30_modules_norm2_parameters_bias_ + ) = None + x_308 = torch._C._nn.linear( + layer_norm_61, + l_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_61 = ( + l_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_309 = torch._C._nn.gelu(x_308, approximate="none") + x_308 = None + x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) + x_309 = None + x_311 = torch._C._nn.linear( + x_310, + l_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_bias_, + ) + x_310 = ( + l_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_312 = torch.nn.functional.dropout(x_311, 0.0, False, False) + x_311 = None + mul_92 = l_self_modules_blocks_modules_30_parameters_gamma_2_ * x_312 + l_self_modules_blocks_modules_30_parameters_gamma_2_ = x_312 = None + x_313 = x_307 + mul_92 + x_307 = mul_92 = None + layer_norm_62 = torch.nn.functional.layer_norm( + x_313, + (192,), + l_self_modules_blocks_modules_31_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_31_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_31_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_31_modules_norm1_parameters_bias_ + ) = None + linear_186 = torch._C._nn.linear( + layer_norm_62, + l_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_62 = ( + l_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_62 = linear_186.reshape(1, 196, 3, 4, 48) + linear_186 = None + qkv_31 = reshape_62.permute(2, 0, 3, 1, 4) + reshape_62 = None + getitem_97 = qkv_31[0] + q_31 = getitem_97 * 0.14433756729740643 + getitem_97 = None + k_31 = qkv_31[1] + v_31 = qkv_31[2] + qkv_31 = None + transpose_63 = k_31.transpose(-2, -1) + k_31 = None + attn_155 = q_31 @ transpose_63 + q_31 = transpose_63 = None + permute_156 = attn_155.permute(0, 2, 3, 1) + attn_155 = None + linear_187 = torch._C._nn.linear( + permute_156, + l_self_modules_blocks_modules_31_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_31_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_156 = l_self_modules_blocks_modules_31_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_31_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_156 = linear_187.permute(0, 3, 1, 2) + linear_187 = None + attn_157 = attn_156.softmax(dim=-1) + attn_156 = None + permute_158 = attn_157.permute(0, 2, 3, 1) + attn_157 = None + linear_188 = torch._C._nn.linear( + permute_158, + l_self_modules_blocks_modules_31_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_31_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_158 = l_self_modules_blocks_modules_31_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_31_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_158 = linear_188.permute(0, 3, 1, 2) + linear_188 = None + attn_159 = torch.nn.functional.dropout(attn_158, 0.0, False, False) + attn_158 = None + matmul_63 = attn_159 @ v_31 + attn_159 = v_31 = None + transpose_64 = matmul_63.transpose(1, 2) + matmul_63 = None + x_314 = transpose_64.reshape(1, 196, 192) + transpose_64 = None + x_315 = torch._C._nn.linear( + x_314, + l_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_bias_, + ) + x_314 = l_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_bias_ + ) = None + x_316 = torch.nn.functional.dropout(x_315, 0.0, False, False) + x_315 = None + mul_94 = l_self_modules_blocks_modules_31_parameters_gamma_1_ * x_316 + l_self_modules_blocks_modules_31_parameters_gamma_1_ = x_316 = None + x_317 = x_313 + mul_94 + x_313 = mul_94 = None + layer_norm_63 = torch.nn.functional.layer_norm( + x_317, + (192,), + l_self_modules_blocks_modules_31_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_31_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_31_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_31_modules_norm2_parameters_bias_ + ) = None + x_318 = torch._C._nn.linear( + layer_norm_63, + l_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_63 = ( + l_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_319 = torch._C._nn.gelu(x_318, approximate="none") + x_318 = None + x_320 = torch.nn.functional.dropout(x_319, 0.0, False, False) + x_319 = None + x_321 = torch._C._nn.linear( + x_320, + l_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_bias_, + ) + x_320 = ( + l_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_322 = torch.nn.functional.dropout(x_321, 0.0, False, False) + x_321 = None + mul_95 = l_self_modules_blocks_modules_31_parameters_gamma_2_ * x_322 + l_self_modules_blocks_modules_31_parameters_gamma_2_ = x_322 = None + x_323 = x_317 + mul_95 + x_317 = mul_95 = None + layer_norm_64 = torch.nn.functional.layer_norm( + x_323, + (192,), + l_self_modules_blocks_modules_32_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_32_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_32_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_32_modules_norm1_parameters_bias_ + ) = None + linear_192 = torch._C._nn.linear( + layer_norm_64, + l_self_modules_blocks_modules_32_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_32_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_64 = ( + l_self_modules_blocks_modules_32_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_32_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_64 = linear_192.reshape(1, 196, 3, 4, 48) + linear_192 = None + qkv_32 = reshape_64.permute(2, 0, 3, 1, 4) + reshape_64 = None + getitem_100 = qkv_32[0] + q_32 = getitem_100 * 0.14433756729740643 + getitem_100 = None + k_32 = qkv_32[1] + v_32 = qkv_32[2] + qkv_32 = None + transpose_65 = k_32.transpose(-2, -1) + k_32 = None + attn_160 = q_32 @ transpose_65 + q_32 = transpose_65 = None + permute_161 = attn_160.permute(0, 2, 3, 1) + attn_160 = None + linear_193 = torch._C._nn.linear( + permute_161, + l_self_modules_blocks_modules_32_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_32_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_161 = l_self_modules_blocks_modules_32_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_32_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_161 = linear_193.permute(0, 3, 1, 2) + linear_193 = None + attn_162 = attn_161.softmax(dim=-1) + attn_161 = None + permute_163 = attn_162.permute(0, 2, 3, 1) + attn_162 = None + linear_194 = torch._C._nn.linear( + permute_163, + l_self_modules_blocks_modules_32_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_32_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_163 = l_self_modules_blocks_modules_32_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_32_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_163 = linear_194.permute(0, 3, 1, 2) + linear_194 = None + attn_164 = torch.nn.functional.dropout(attn_163, 0.0, False, False) + attn_163 = None + matmul_65 = attn_164 @ v_32 + attn_164 = v_32 = None + transpose_66 = matmul_65.transpose(1, 2) + matmul_65 = None + x_324 = transpose_66.reshape(1, 196, 192) + transpose_66 = None + x_325 = torch._C._nn.linear( + x_324, + l_self_modules_blocks_modules_32_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_32_modules_attn_modules_proj_parameters_bias_, + ) + x_324 = l_self_modules_blocks_modules_32_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_32_modules_attn_modules_proj_parameters_bias_ + ) = None + x_326 = torch.nn.functional.dropout(x_325, 0.0, False, False) + x_325 = None + mul_97 = l_self_modules_blocks_modules_32_parameters_gamma_1_ * x_326 + l_self_modules_blocks_modules_32_parameters_gamma_1_ = x_326 = None + x_327 = x_323 + mul_97 + x_323 = mul_97 = None + layer_norm_65 = torch.nn.functional.layer_norm( + x_327, + (192,), + l_self_modules_blocks_modules_32_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_32_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_32_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_32_modules_norm2_parameters_bias_ + ) = None + x_328 = torch._C._nn.linear( + layer_norm_65, + l_self_modules_blocks_modules_32_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_32_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_65 = ( + l_self_modules_blocks_modules_32_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_32_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_329 = torch._C._nn.gelu(x_328, approximate="none") + x_328 = None + x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) + x_329 = None + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_blocks_modules_32_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_32_modules_mlp_modules_fc2_parameters_bias_, + ) + x_330 = ( + l_self_modules_blocks_modules_32_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_32_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_332 = torch.nn.functional.dropout(x_331, 0.0, False, False) + x_331 = None + mul_98 = l_self_modules_blocks_modules_32_parameters_gamma_2_ * x_332 + l_self_modules_blocks_modules_32_parameters_gamma_2_ = x_332 = None + x_333 = x_327 + mul_98 + x_327 = mul_98 = None + layer_norm_66 = torch.nn.functional.layer_norm( + x_333, + (192,), + l_self_modules_blocks_modules_33_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_33_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_33_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_33_modules_norm1_parameters_bias_ + ) = None + linear_198 = torch._C._nn.linear( + layer_norm_66, + l_self_modules_blocks_modules_33_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_33_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_66 = ( + l_self_modules_blocks_modules_33_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_33_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_66 = linear_198.reshape(1, 196, 3, 4, 48) + linear_198 = None + qkv_33 = reshape_66.permute(2, 0, 3, 1, 4) + reshape_66 = None + getitem_103 = qkv_33[0] + q_33 = getitem_103 * 0.14433756729740643 + getitem_103 = None + k_33 = qkv_33[1] + v_33 = qkv_33[2] + qkv_33 = None + transpose_67 = k_33.transpose(-2, -1) + k_33 = None + attn_165 = q_33 @ transpose_67 + q_33 = transpose_67 = None + permute_166 = attn_165.permute(0, 2, 3, 1) + attn_165 = None + linear_199 = torch._C._nn.linear( + permute_166, + l_self_modules_blocks_modules_33_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_33_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_166 = l_self_modules_blocks_modules_33_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_33_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_166 = linear_199.permute(0, 3, 1, 2) + linear_199 = None + attn_167 = attn_166.softmax(dim=-1) + attn_166 = None + permute_168 = attn_167.permute(0, 2, 3, 1) + attn_167 = None + linear_200 = torch._C._nn.linear( + permute_168, + l_self_modules_blocks_modules_33_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_33_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_168 = l_self_modules_blocks_modules_33_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_33_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_168 = linear_200.permute(0, 3, 1, 2) + linear_200 = None + attn_169 = torch.nn.functional.dropout(attn_168, 0.0, False, False) + attn_168 = None + matmul_67 = attn_169 @ v_33 + attn_169 = v_33 = None + transpose_68 = matmul_67.transpose(1, 2) + matmul_67 = None + x_334 = transpose_68.reshape(1, 196, 192) + transpose_68 = None + x_335 = torch._C._nn.linear( + x_334, + l_self_modules_blocks_modules_33_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_33_modules_attn_modules_proj_parameters_bias_, + ) + x_334 = l_self_modules_blocks_modules_33_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_33_modules_attn_modules_proj_parameters_bias_ + ) = None + x_336 = torch.nn.functional.dropout(x_335, 0.0, False, False) + x_335 = None + mul_100 = l_self_modules_blocks_modules_33_parameters_gamma_1_ * x_336 + l_self_modules_blocks_modules_33_parameters_gamma_1_ = x_336 = None + x_337 = x_333 + mul_100 + x_333 = mul_100 = None + layer_norm_67 = torch.nn.functional.layer_norm( + x_337, + (192,), + l_self_modules_blocks_modules_33_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_33_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_33_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_33_modules_norm2_parameters_bias_ + ) = None + x_338 = torch._C._nn.linear( + layer_norm_67, + l_self_modules_blocks_modules_33_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_33_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_67 = ( + l_self_modules_blocks_modules_33_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_33_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_339 = torch._C._nn.gelu(x_338, approximate="none") + x_338 = None + x_340 = torch.nn.functional.dropout(x_339, 0.0, False, False) + x_339 = None + x_341 = torch._C._nn.linear( + x_340, + l_self_modules_blocks_modules_33_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_33_modules_mlp_modules_fc2_parameters_bias_, + ) + x_340 = ( + l_self_modules_blocks_modules_33_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_33_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_342 = torch.nn.functional.dropout(x_341, 0.0, False, False) + x_341 = None + mul_101 = l_self_modules_blocks_modules_33_parameters_gamma_2_ * x_342 + l_self_modules_blocks_modules_33_parameters_gamma_2_ = x_342 = None + x_343 = x_337 + mul_101 + x_337 = mul_101 = None + layer_norm_68 = torch.nn.functional.layer_norm( + x_343, + (192,), + l_self_modules_blocks_modules_34_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_34_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_34_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_34_modules_norm1_parameters_bias_ + ) = None + linear_204 = torch._C._nn.linear( + layer_norm_68, + l_self_modules_blocks_modules_34_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_34_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_68 = ( + l_self_modules_blocks_modules_34_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_34_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_68 = linear_204.reshape(1, 196, 3, 4, 48) + linear_204 = None + qkv_34 = reshape_68.permute(2, 0, 3, 1, 4) + reshape_68 = None + getitem_106 = qkv_34[0] + q_34 = getitem_106 * 0.14433756729740643 + getitem_106 = None + k_34 = qkv_34[1] + v_34 = qkv_34[2] + qkv_34 = None + transpose_69 = k_34.transpose(-2, -1) + k_34 = None + attn_170 = q_34 @ transpose_69 + q_34 = transpose_69 = None + permute_171 = attn_170.permute(0, 2, 3, 1) + attn_170 = None + linear_205 = torch._C._nn.linear( + permute_171, + l_self_modules_blocks_modules_34_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_34_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_171 = l_self_modules_blocks_modules_34_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_34_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_171 = linear_205.permute(0, 3, 1, 2) + linear_205 = None + attn_172 = attn_171.softmax(dim=-1) + attn_171 = None + permute_173 = attn_172.permute(0, 2, 3, 1) + attn_172 = None + linear_206 = torch._C._nn.linear( + permute_173, + l_self_modules_blocks_modules_34_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_34_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_173 = l_self_modules_blocks_modules_34_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_34_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_173 = linear_206.permute(0, 3, 1, 2) + linear_206 = None + attn_174 = torch.nn.functional.dropout(attn_173, 0.0, False, False) + attn_173 = None + matmul_69 = attn_174 @ v_34 + attn_174 = v_34 = None + transpose_70 = matmul_69.transpose(1, 2) + matmul_69 = None + x_344 = transpose_70.reshape(1, 196, 192) + transpose_70 = None + x_345 = torch._C._nn.linear( + x_344, + l_self_modules_blocks_modules_34_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_34_modules_attn_modules_proj_parameters_bias_, + ) + x_344 = l_self_modules_blocks_modules_34_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_34_modules_attn_modules_proj_parameters_bias_ + ) = None + x_346 = torch.nn.functional.dropout(x_345, 0.0, False, False) + x_345 = None + mul_103 = l_self_modules_blocks_modules_34_parameters_gamma_1_ * x_346 + l_self_modules_blocks_modules_34_parameters_gamma_1_ = x_346 = None + x_347 = x_343 + mul_103 + x_343 = mul_103 = None + layer_norm_69 = torch.nn.functional.layer_norm( + x_347, + (192,), + l_self_modules_blocks_modules_34_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_34_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_34_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_34_modules_norm2_parameters_bias_ + ) = None + x_348 = torch._C._nn.linear( + layer_norm_69, + l_self_modules_blocks_modules_34_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_34_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_69 = ( + l_self_modules_blocks_modules_34_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_34_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_349 = torch._C._nn.gelu(x_348, approximate="none") + x_348 = None + x_350 = torch.nn.functional.dropout(x_349, 0.0, False, False) + x_349 = None + x_351 = torch._C._nn.linear( + x_350, + l_self_modules_blocks_modules_34_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_34_modules_mlp_modules_fc2_parameters_bias_, + ) + x_350 = ( + l_self_modules_blocks_modules_34_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_34_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_352 = torch.nn.functional.dropout(x_351, 0.0, False, False) + x_351 = None + mul_104 = l_self_modules_blocks_modules_34_parameters_gamma_2_ * x_352 + l_self_modules_blocks_modules_34_parameters_gamma_2_ = x_352 = None + x_353 = x_347 + mul_104 + x_347 = mul_104 = None + layer_norm_70 = torch.nn.functional.layer_norm( + x_353, + (192,), + l_self_modules_blocks_modules_35_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_35_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_35_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_35_modules_norm1_parameters_bias_ + ) = None + linear_210 = torch._C._nn.linear( + layer_norm_70, + l_self_modules_blocks_modules_35_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_35_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_70 = ( + l_self_modules_blocks_modules_35_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_35_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_70 = linear_210.reshape(1, 196, 3, 4, 48) + linear_210 = None + qkv_35 = reshape_70.permute(2, 0, 3, 1, 4) + reshape_70 = None + getitem_109 = qkv_35[0] + q_35 = getitem_109 * 0.14433756729740643 + getitem_109 = None + k_35 = qkv_35[1] + v_35 = qkv_35[2] + qkv_35 = None + transpose_71 = k_35.transpose(-2, -1) + k_35 = None + attn_175 = q_35 @ transpose_71 + q_35 = transpose_71 = None + permute_176 = attn_175.permute(0, 2, 3, 1) + attn_175 = None + linear_211 = torch._C._nn.linear( + permute_176, + l_self_modules_blocks_modules_35_modules_attn_modules_proj_l_parameters_weight_, + l_self_modules_blocks_modules_35_modules_attn_modules_proj_l_parameters_bias_, + ) + permute_176 = l_self_modules_blocks_modules_35_modules_attn_modules_proj_l_parameters_weight_ = l_self_modules_blocks_modules_35_modules_attn_modules_proj_l_parameters_bias_ = (None) + attn_176 = linear_211.permute(0, 3, 1, 2) + linear_211 = None + attn_177 = attn_176.softmax(dim=-1) + attn_176 = None + permute_178 = attn_177.permute(0, 2, 3, 1) + attn_177 = None + linear_212 = torch._C._nn.linear( + permute_178, + l_self_modules_blocks_modules_35_modules_attn_modules_proj_w_parameters_weight_, + l_self_modules_blocks_modules_35_modules_attn_modules_proj_w_parameters_bias_, + ) + permute_178 = l_self_modules_blocks_modules_35_modules_attn_modules_proj_w_parameters_weight_ = l_self_modules_blocks_modules_35_modules_attn_modules_proj_w_parameters_bias_ = (None) + attn_178 = linear_212.permute(0, 3, 1, 2) + linear_212 = None + attn_179 = torch.nn.functional.dropout(attn_178, 0.0, False, False) + attn_178 = None + matmul_71 = attn_179 @ v_35 + attn_179 = v_35 = None + transpose_72 = matmul_71.transpose(1, 2) + matmul_71 = None + x_354 = transpose_72.reshape(1, 196, 192) + transpose_72 = None + x_355 = torch._C._nn.linear( + x_354, + l_self_modules_blocks_modules_35_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_35_modules_attn_modules_proj_parameters_bias_, + ) + x_354 = l_self_modules_blocks_modules_35_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_35_modules_attn_modules_proj_parameters_bias_ + ) = None + x_356 = torch.nn.functional.dropout(x_355, 0.0, False, False) + x_355 = None + mul_106 = l_self_modules_blocks_modules_35_parameters_gamma_1_ * x_356 + l_self_modules_blocks_modules_35_parameters_gamma_1_ = x_356 = None + x_357 = x_353 + mul_106 + x_353 = mul_106 = None + layer_norm_71 = torch.nn.functional.layer_norm( + x_357, + (192,), + l_self_modules_blocks_modules_35_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_35_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_35_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_35_modules_norm2_parameters_bias_ + ) = None + x_358 = torch._C._nn.linear( + layer_norm_71, + l_self_modules_blocks_modules_35_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_35_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_71 = ( + l_self_modules_blocks_modules_35_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_35_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_359 = torch._C._nn.gelu(x_358, approximate="none") + x_358 = None + x_360 = torch.nn.functional.dropout(x_359, 0.0, False, False) + x_359 = None + x_361 = torch._C._nn.linear( + x_360, + l_self_modules_blocks_modules_35_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_35_modules_mlp_modules_fc2_parameters_bias_, + ) + x_360 = ( + l_self_modules_blocks_modules_35_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_35_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_362 = torch.nn.functional.dropout(x_361, 0.0, False, False) + x_361 = None + mul_107 = l_self_modules_blocks_modules_35_parameters_gamma_2_ * x_362 + l_self_modules_blocks_modules_35_parameters_gamma_2_ = x_362 = None + x_363 = x_357 + mul_107 + x_357 = mul_107 = None + cls_tokens = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + u = torch.cat((cls_tokens, x_363), dim=1) + layer_norm_72 = torch.nn.functional.layer_norm( + u, + (192,), + l_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + u = ( + l_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_weight_ + ) = ( + l_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_bias_ + ) = None + getitem_112 = layer_norm_72[(slice(None, None, None), 0)] + linear_216 = torch._C._nn.linear( + getitem_112, + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_weight_, + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_bias_, + ) + getitem_112 = l_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_weight_ = l_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_bias_ = (None) + unsqueeze = linear_216.unsqueeze(1) + linear_216 = None + reshape_72 = unsqueeze.reshape(1, 1, 4, 48) + unsqueeze = None + q_36 = reshape_72.permute(0, 2, 1, 3) + reshape_72 = None + linear_217 = torch._C._nn.linear( + layer_norm_72, + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_weight_, + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_bias_, + ) + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_weight_ = l_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_bias_ = (None) + reshape_73 = linear_217.reshape(1, 197, 4, 48) + linear_217 = None + k_36 = reshape_73.permute(0, 2, 1, 3) + reshape_73 = None + linear_218 = torch._C._nn.linear( + layer_norm_72, + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_weight_, + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_bias_, + ) + layer_norm_72 = l_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_weight_ = l_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_bias_ = (None) + reshape_74 = linear_218.reshape(1, 197, 4, 48) + linear_218 = None + v_36 = reshape_74.permute(0, 2, 1, 3) + reshape_74 = None + x_cls = torch._C._nn.scaled_dot_product_attention( + q_36, k_36, v_36, dropout_p=0.0 + ) + q_36 = k_36 = v_36 = None + transpose_73 = x_cls.transpose(1, 2) + x_cls = None + x_cls_1 = transpose_73.reshape(1, 1, 192) + transpose_73 = None + x_cls_2 = torch._C._nn.linear( + x_cls_1, + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_cls_1 = l_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_cls_3 = torch.nn.functional.dropout(x_cls_2, 0.0, False, False) + x_cls_2 = None + mul_108 = ( + l_self_modules_blocks_token_only_modules_0_parameters_gamma_1_ * x_cls_3 + ) + l_self_modules_blocks_token_only_modules_0_parameters_gamma_1_ = x_cls_3 = None + x_cls_4 = cls_tokens + mul_108 + cls_tokens = mul_108 = None + layer_norm_73 = torch.nn.functional.layer_norm( + x_cls_4, + (192,), + l_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_bias_ + ) = None + x_364 = torch._C._nn.linear( + layer_norm_73, + l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_73 = l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_365 = torch._C._nn.gelu(x_364, approximate="none") + x_364 = None + x_366 = torch.nn.functional.dropout(x_365, 0.0, False, False) + x_365 = None + x_367 = torch._C._nn.linear( + x_366, + l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_366 = l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_368 = torch.nn.functional.dropout(x_367, 0.0, False, False) + x_367 = None + mul_109 = l_self_modules_blocks_token_only_modules_0_parameters_gamma_2_ * x_368 + l_self_modules_blocks_token_only_modules_0_parameters_gamma_2_ = x_368 = None + x_cls_5 = x_cls_4 + mul_109 + x_cls_4 = mul_109 = None + u_1 = torch.cat((x_cls_5, x_363), dim=1) + layer_norm_74 = torch.nn.functional.layer_norm( + u_1, + (192,), + l_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + u_1 = ( + l_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_weight_ + ) = ( + l_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_bias_ + ) = None + getitem_113 = layer_norm_74[(slice(None, None, None), 0)] + linear_222 = torch._C._nn.linear( + getitem_113, + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_weight_, + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_bias_, + ) + getitem_113 = l_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_weight_ = l_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_bias_ = (None) + unsqueeze_1 = linear_222.unsqueeze(1) + linear_222 = None + reshape_76 = unsqueeze_1.reshape(1, 1, 4, 48) + unsqueeze_1 = None + q_37 = reshape_76.permute(0, 2, 1, 3) + reshape_76 = None + linear_223 = torch._C._nn.linear( + layer_norm_74, + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_weight_, + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_bias_, + ) + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_weight_ = l_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_bias_ = (None) + reshape_77 = linear_223.reshape(1, 197, 4, 48) + linear_223 = None + k_37 = reshape_77.permute(0, 2, 1, 3) + reshape_77 = None + linear_224 = torch._C._nn.linear( + layer_norm_74, + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_weight_, + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_bias_, + ) + layer_norm_74 = l_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_weight_ = l_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_bias_ = (None) + reshape_78 = linear_224.reshape(1, 197, 4, 48) + linear_224 = None + v_37 = reshape_78.permute(0, 2, 1, 3) + reshape_78 = None + x_cls_6 = torch._C._nn.scaled_dot_product_attention( + q_37, k_37, v_37, dropout_p=0.0 + ) + q_37 = k_37 = v_37 = None + transpose_74 = x_cls_6.transpose(1, 2) + x_cls_6 = None + x_cls_7 = transpose_74.reshape(1, 1, 192) + transpose_74 = None + x_cls_8 = torch._C._nn.linear( + x_cls_7, + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_cls_7 = l_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_cls_9 = torch.nn.functional.dropout(x_cls_8, 0.0, False, False) + x_cls_8 = None + mul_110 = ( + l_self_modules_blocks_token_only_modules_1_parameters_gamma_1_ * x_cls_9 + ) + l_self_modules_blocks_token_only_modules_1_parameters_gamma_1_ = x_cls_9 = None + x_cls_10 = x_cls_5 + mul_110 + x_cls_5 = mul_110 = None + layer_norm_75 = torch.nn.functional.layer_norm( + x_cls_10, + (192,), + l_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_bias_ + ) = None + x_369 = torch._C._nn.linear( + layer_norm_75, + l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_75 = l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_370 = torch._C._nn.gelu(x_369, approximate="none") + x_369 = None + x_371 = torch.nn.functional.dropout(x_370, 0.0, False, False) + x_370 = None + x_372 = torch._C._nn.linear( + x_371, + l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_371 = l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_373 = torch.nn.functional.dropout(x_372, 0.0, False, False) + x_372 = None + mul_111 = l_self_modules_blocks_token_only_modules_1_parameters_gamma_2_ * x_373 + l_self_modules_blocks_token_only_modules_1_parameters_gamma_2_ = x_373 = None + x_cls_11 = x_cls_10 + mul_111 + x_cls_10 = mul_111 = None + x_374 = torch.cat((x_cls_11, x_363), dim=1) + x_cls_11 = x_363 = None + x_375 = torch.nn.functional.layer_norm( + x_374, + (192,), + l_self_modules_norm_parameters_weight_, + l_self_modules_norm_parameters_bias_, + 1e-06, + ) + x_374 = ( + l_self_modules_norm_parameters_weight_ + ) = l_self_modules_norm_parameters_bias_ = None + x_376 = x_375[(slice(None, None, None), 0)] + x_375 = None + x_377 = torch.nn.functional.dropout(x_376, 0.0, False, False) + x_376 = None + x_378 = torch._C._nn.linear( + x_377, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_377 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_378,) diff --git a/samples/timm/cait_xxs36_224.fb_dist_in1k/weight_meta.py b/samples/timm/cait_xxs36_224.fb_dist_in1k/weight_meta.py new file mode 100644 index 0000000000..f4107ae439 --- /dev/null +++ b/samples/timm/cait_xxs36_224.fb_dist_in1k/weight_meta.py @@ -0,0 +1,7236 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.232 + std = 1.282 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [192, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 196, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_0_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.005 + std = 0.027 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.006 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_0_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_1_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.023 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.009 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_1_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_2_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_2_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_3_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.016 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.005 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_3_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_4_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.005 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.025 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_4_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_5_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.008 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.025 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_5_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_6_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_6_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_7_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_7_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_8_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.012 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_8_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_9_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_weight_" + ) + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.017 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_9_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_10_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.008 + std = 0.023 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.006 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_10_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_11_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.006 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_11_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_12_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.012 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.006 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_12_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_13_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.005 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_13_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_14_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.008 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.006 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_14_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_15_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.023 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.006 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_15_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_16_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.005 + std = 0.017 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_16_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_17_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_17_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_18_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.006 + std = 0.023 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.004 + std = 0.025 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_18_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_19_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.009 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.009 + std = 0.017 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_19_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_20_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_20_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_21_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.016 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.011 + std = 0.023 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_21_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_22_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.017 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.007 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_22_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_23_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.005 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.010 + std = 0.027 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_23_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_24_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_24_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_24_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_24_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_24_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_24_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.004 + std = 0.017 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_24_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_24_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_24_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_24_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_25_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_25_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_25_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_25_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_25_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_25_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.009 + std = 0.016 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_25_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_25_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_25_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_25_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_26_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_26_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_26_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_26_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_26_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_26_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.014 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_26_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_26_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_26_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_26_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_27_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_27_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_27_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_27_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.008 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_27_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_27_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.007 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_27_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_27_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_27_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_27_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_28_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_28_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_28_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_28_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_28_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_28_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.016 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_28_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_28_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_28_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_28_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_29_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_29_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_29_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_29_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_29_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_29_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_29_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_29_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_29_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_29_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_30_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_30_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_30_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_30_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.009 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_30_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_30_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_30_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_30_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_30_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_30_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_31_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_31_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_31_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_31_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_31_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_31_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.023 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_31_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_31_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_31_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_31_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_32_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_32_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_32_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_32_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_32_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_32_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_32_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_32_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_32_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_32_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_32_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_32_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.008 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_32_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_32_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_32_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_32_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.010 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_32_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_32_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_32_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_32_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_32_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_32_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_32_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_32_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_32_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_32_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_32_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_32_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_32_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_32_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_32_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_32_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_32_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_32_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_32_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_32_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_33_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_33_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_33_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_33_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_33_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_33_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_33_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_33_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_33_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_33_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_33_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_33_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_33_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_33_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_33_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_33_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_33_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_33_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_33_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_33_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_33_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_33_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_33_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_33_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_33_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_33_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_33_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_33_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_33_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_33_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_33_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_33_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_33_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_33_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_33_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_33_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_34_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_34_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_34_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_34_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_34_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_34_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_34_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_34_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_34_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_34_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_34_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_34_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.027 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_34_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_34_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_34_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_34_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_34_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_34_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_34_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_34_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_34_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_34_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_34_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_34_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_34_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_34_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_34_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_34_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_34_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_34_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_34_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_34_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_34_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_34_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_34_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_34_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_35_parameters_gamma_1_: + name = "L_self_modules_blocks_modules_35_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_35_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_35_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_35_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_35_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_35_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_35_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_35_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_35_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_35_modules_attn_modules_proj_l_parameters_weight_: + name = "L_self_modules_blocks_modules_35_modules_attn_modules_proj_l_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_35_modules_attn_modules_proj_l_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_35_modules_attn_modules_proj_l_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_35_modules_attn_modules_proj_w_parameters_weight_: + name = "L_self_modules_blocks_modules_35_modules_attn_modules_proj_w_parameters_weight_" + shape = [4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_35_modules_attn_modules_proj_w_parameters_bias_: + name = ( + "L_self_modules_blocks_modules_35_modules_attn_modules_proj_w_parameters_bias_" + ) + shape = [4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_35_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_35_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_35_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_35_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_35_parameters_gamma_2_: + name = "L_self_modules_blocks_modules_35_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_35_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_35_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_35_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_35_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_35_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_35_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_35_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_35_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_35_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_35_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_35_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_35_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_parameters_gamma_1_: + name = "L_self_modules_blocks_token_only_modules_0_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_0_modules_attn_modules_q_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_0_modules_attn_modules_k_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_0_modules_attn_modules_v_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_parameters_gamma_2_: + name = "L_self_modules_blocks_token_only_modules_0_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_parameters_gamma_1_: + name = "L_self_modules_blocks_token_only_modules_1_parameters_gamma_1_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_1_modules_attn_modules_q_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_1_modules_attn_modules_k_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_1_modules_attn_modules_v_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_parameters_gamma_2_: + name = "L_self_modules_blocks_token_only_modules_1_parameters_gamma_2_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_1_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_token_only_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_weight_: + name = "L_self_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_bias_: + name = "L_self_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/coat_lite_medium.in1k/graph_hash.txt b/samples/timm/coat_lite_medium.in1k/graph_hash.txt new file mode 100644 index 0000000000..00c332427b --- /dev/null +++ b/samples/timm/coat_lite_medium.in1k/graph_hash.txt @@ -0,0 +1 @@ +89c00326da6b2b9930919a98f8db1c2b3d4a880a204269070dd51b4ddf3793a1 \ No newline at end of file diff --git a/samples/timm/coat_lite_medium.in1k/graph_net.json b/samples/timm/coat_lite_medium.in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/coat_lite_medium.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/coat_lite_medium.in1k/input_meta.py b/samples/timm/coat_lite_medium.in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/coat_lite_medium.in1k/input_tensor_constraints.py b/samples/timm/coat_lite_medium.in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/coat_lite_medium.in1k/model.py b/samples/timm/coat_lite_medium.in1k/model.py new file mode 100644 index 0000000000..7115d6a0ac --- /dev/null +++ b/samples/timm/coat_lite_medium.in1k/model.py @@ -0,0 +1,5903 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed1_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed1_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token1_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed2_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed2_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token2_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed3_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed3_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token3_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed4_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed4_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token4_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm4_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm4_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed1_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed1_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed1_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed1_modules_proj_parameters_bias_ + ) + l_self_modules_patch_embed1_modules_norm_parameters_weight_ = ( + L_self_modules_patch_embed1_modules_norm_parameters_weight_ + ) + l_self_modules_patch_embed1_modules_norm_parameters_bias_ = ( + L_self_modules_patch_embed1_modules_norm_parameters_bias_ + ) + l_self_parameters_cls_token1_ = L_self_parameters_cls_token1_ + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_patch_embed2_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed2_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed2_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed2_modules_proj_parameters_bias_ + ) + l_self_modules_patch_embed2_modules_norm_parameters_weight_ = ( + L_self_modules_patch_embed2_modules_norm_parameters_weight_ + ) + l_self_modules_patch_embed2_modules_norm_parameters_bias_ = ( + L_self_modules_patch_embed2_modules_norm_parameters_bias_ + ) + l_self_parameters_cls_token2_ = L_self_parameters_cls_token2_ + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks2_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks2_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks2_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks2_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_patch_embed3_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed3_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed3_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed3_modules_proj_parameters_bias_ + ) + l_self_modules_patch_embed3_modules_norm_parameters_weight_ = ( + L_self_modules_patch_embed3_modules_norm_parameters_weight_ + ) + l_self_modules_patch_embed3_modules_norm_parameters_bias_ = ( + L_self_modules_patch_embed3_modules_norm_parameters_bias_ + ) + l_self_parameters_cls_token3_ = L_self_parameters_cls_token3_ + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks3_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks3_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks3_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks3_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks3_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks3_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks3_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks3_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_patch_embed4_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed4_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed4_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed4_modules_proj_parameters_bias_ + ) + l_self_modules_patch_embed4_modules_norm_parameters_weight_ = ( + L_self_modules_patch_embed4_modules_norm_parameters_weight_ + ) + l_self_modules_patch_embed4_modules_norm_parameters_bias_ = ( + L_self_modules_patch_embed4_modules_norm_parameters_bias_ + ) + l_self_parameters_cls_token4_ = L_self_parameters_cls_token4_ + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks4_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks4_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks4_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks4_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks4_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks4_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks4_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks4_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks4_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks4_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_norm4_parameters_weight_ = ( + L_self_modules_norm4_parameters_weight_ + ) + l_self_modules_norm4_parameters_bias_ = L_self_modules_norm4_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed1_modules_proj_parameters_weight_, + l_self_modules_patch_embed1_modules_proj_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed1_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed1_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + x_2 = torch.nn.functional.layer_norm( + x_1, + (128,), + l_self_modules_patch_embed1_modules_norm_parameters_weight_, + l_self_modules_patch_embed1_modules_norm_parameters_bias_, + 1e-05, + ) + x_1 = ( + l_self_modules_patch_embed1_modules_norm_parameters_weight_ + ) = l_self_modules_patch_embed1_modules_norm_parameters_bias_ = None + cls_tokens = l_self_parameters_cls_token1_.expand(1, -1, -1) + l_self_parameters_cls_token1_ = None + x_3 = torch.cat((cls_tokens, x_2), dim=1) + cls_tokens = x_2 = None + cls_token = x_3[(slice(None, None, None), slice(None, 1, None))] + img_tokens = x_3[(slice(None, None, None), slice(1, None, None))] + x_3 = None + transpose_1 = img_tokens.transpose(1, 2) + img_tokens = None + feat = transpose_1.view(1, 128, 56, 56) + transpose_1 = None + conv2d_1 = torch.conv2d( + feat, + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 128, + ) + x_4 = conv2d_1 + feat + conv2d_1 = feat = None + flatten_1 = x_4.flatten(2) + x_4 = None + x_5 = flatten_1.transpose(1, 2) + flatten_1 = None + x_6 = torch.cat((cls_token, x_5), dim=1) + cls_token = x_5 = None + x_7 = torch.nn.functional.layer_norm( + x_6, + (128,), + l_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_7, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_7 = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 3137, 3, 8, 16) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + k_softmax = k.softmax(dim=2) + k = None + transpose_3 = k_softmax.transpose(-1, -2) + k_softmax = None + factor_att = transpose_3 @ v + transpose_3 = None + factor_att_1 = q @ factor_att + factor_att = None + q_img = q[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q = None + v_img = v[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v = None + transpose_4 = v_img.transpose(-1, -2) + v_img = None + v_img_1 = transpose_4.reshape(1, 128, 56, 56) + transpose_4 = None + split = torch.functional.split(v_img_1, [32, 48, 48], dim=1) + v_img_1 = None + getitem_15 = split[0] + getitem_16 = split[1] + getitem_17 = split[2] + split = None + conv2d_2 = torch.conv2d( + getitem_15, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + getitem_15 = None + conv2d_3 = torch.conv2d( + getitem_16, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 48, + ) + getitem_16 = None + conv2d_4 = torch.conv2d( + getitem_17, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 48, + ) + getitem_17 = None + conv_v_img = torch.cat([conv2d_2, conv2d_3, conv2d_4], dim=1) + conv2d_2 = conv2d_3 = conv2d_4 = None + reshape_2 = conv_v_img.reshape(1, 8, 16, 3136) + conv_v_img = None + conv_v_img_1 = reshape_2.transpose(-1, -2) + reshape_2 = None + EV_hat = q_img * conv_v_img_1 + q_img = conv_v_img_1 = None + EV_hat_1 = torch._C._nn.pad(EV_hat, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat = None + mul_1 = 0.25 * factor_att_1 + factor_att_1 = None + x_8 = mul_1 + EV_hat_1 + mul_1 = EV_hat_1 = None + transpose_6 = x_8.transpose(1, 2) + x_8 = None + x_9 = transpose_6.reshape(1, 3137, 128) + transpose_6 = None + x_10 = torch._C._nn.linear( + x_9, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_9 = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = x_6 + x_11 + x_6 = x_11 = None + x_13 = torch.nn.functional.layer_norm( + x_12, + (128,), + l_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_ + ) = None + x_14 = torch._C._nn.linear( + x_13, + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_13 = l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_15 = torch._C._nn.gelu(x_14, approximate="none") + x_14 = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_16 = l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_18 = torch.nn.functional.dropout(x_17, 0.0, False, False) + x_17 = None + x_19 = x_12 + x_18 + x_12 = x_18 = None + cls_token_1 = x_19[(slice(None, None, None), slice(None, 1, None))] + img_tokens_1 = x_19[(slice(None, None, None), slice(1, None, None))] + x_19 = None + transpose_7 = img_tokens_1.transpose(1, 2) + img_tokens_1 = None + feat_1 = transpose_7.view(1, 128, 56, 56) + transpose_7 = None + conv2d_5 = torch.conv2d( + feat_1, + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 128, + ) + x_20 = conv2d_5 + feat_1 + conv2d_5 = feat_1 = None + flatten_2 = x_20.flatten(2) + x_20 = None + x_21 = flatten_2.transpose(1, 2) + flatten_2 = None + x_22 = torch.cat((cls_token_1, x_21), dim=1) + cls_token_1 = x_21 = None + x_23 = torch.nn.functional.layer_norm( + x_22, + (128,), + l_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_23, + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_23 = l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_4.reshape(1, 3137, 3, 8, 16) + linear_4 = None + qkv_1 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + k_softmax_1 = k_1.softmax(dim=2) + k_1 = None + transpose_9 = k_softmax_1.transpose(-1, -2) + k_softmax_1 = None + factor_att_2 = transpose_9 @ v_1 + transpose_9 = None + factor_att_3 = q_1 @ factor_att_2 + factor_att_2 = None + q_img_1 = q_1[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_1 = None + v_img_2 = v_1[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_1 = None + transpose_10 = v_img_2.transpose(-1, -2) + v_img_2 = None + v_img_3 = transpose_10.reshape(1, 128, 56, 56) + transpose_10 = None + split_1 = torch.functional.split(v_img_3, [32, 48, 48], dim=1) + v_img_3 = None + getitem_25 = split_1[0] + getitem_26 = split_1[1] + getitem_27 = split_1[2] + split_1 = None + conv2d_6 = torch.conv2d( + getitem_25, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + getitem_25 = None + conv2d_7 = torch.conv2d( + getitem_26, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 48, + ) + getitem_26 = None + conv2d_8 = torch.conv2d( + getitem_27, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 48, + ) + getitem_27 = None + conv_v_img_2 = torch.cat([conv2d_6, conv2d_7, conv2d_8], dim=1) + conv2d_6 = conv2d_7 = conv2d_8 = None + reshape_6 = conv_v_img_2.reshape(1, 8, 16, 3136) + conv_v_img_2 = None + conv_v_img_3 = reshape_6.transpose(-1, -2) + reshape_6 = None + EV_hat_2 = q_img_1 * conv_v_img_3 + q_img_1 = conv_v_img_3 = None + EV_hat_3 = torch._C._nn.pad(EV_hat_2, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_2 = None + mul_3 = 0.25 * factor_att_3 + factor_att_3 = None + x_24 = mul_3 + EV_hat_3 + mul_3 = EV_hat_3 = None + transpose_12 = x_24.transpose(1, 2) + x_24 = None + x_25 = transpose_12.reshape(1, 3137, 128) + transpose_12 = None + x_26 = torch._C._nn.linear( + x_25, + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_25 = l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = x_22 + x_27 + x_22 = x_27 = None + x_29 = torch.nn.functional.layer_norm( + x_28, + (128,), + l_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_ + ) = None + x_30 = torch._C._nn.linear( + x_29, + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_29 = l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_31 = torch._C._nn.gelu(x_30, approximate="none") + x_30 = None + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = torch._C._nn.linear( + x_32, + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_32 = l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = x_28 + x_34 + x_28 = x_34 = None + cls_token_2 = x_35[(slice(None, None, None), slice(None, 1, None))] + img_tokens_2 = x_35[(slice(None, None, None), slice(1, None, None))] + x_35 = None + transpose_13 = img_tokens_2.transpose(1, 2) + img_tokens_2 = None + feat_2 = transpose_13.view(1, 128, 56, 56) + transpose_13 = None + conv2d_9 = torch.conv2d( + feat_2, + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 128, + ) + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_ = (None) + x_36 = conv2d_9 + feat_2 + conv2d_9 = feat_2 = None + flatten_3 = x_36.flatten(2) + x_36 = None + x_37 = flatten_3.transpose(1, 2) + flatten_3 = None + x_38 = torch.cat((cls_token_2, x_37), dim=1) + cls_token_2 = x_37 = None + x_39 = torch.nn.functional.layer_norm( + x_38, + (128,), + l_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_39, + l_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_39 = l_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_8 = linear_8.reshape(1, 3137, 3, 8, 16) + linear_8 = None + qkv_2 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + k_softmax_2 = k_2.softmax(dim=2) + k_2 = None + transpose_15 = k_softmax_2.transpose(-1, -2) + k_softmax_2 = None + factor_att_4 = transpose_15 @ v_2 + transpose_15 = None + factor_att_5 = q_2 @ factor_att_4 + factor_att_4 = None + q_img_2 = q_2[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_2 = None + v_img_4 = v_2[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_2 = None + transpose_16 = v_img_4.transpose(-1, -2) + v_img_4 = None + v_img_5 = transpose_16.reshape(1, 128, 56, 56) + transpose_16 = None + split_2 = torch.functional.split(v_img_5, [32, 48, 48], dim=1) + v_img_5 = None + getitem_35 = split_2[0] + getitem_36 = split_2[1] + getitem_37 = split_2[2] + split_2 = None + conv2d_10 = torch.conv2d( + getitem_35, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + getitem_35 = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = (None) + conv2d_11 = torch.conv2d( + getitem_36, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 48, + ) + getitem_36 = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = (None) + conv2d_12 = torch.conv2d( + getitem_37, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 48, + ) + getitem_37 = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = (None) + conv_v_img_4 = torch.cat([conv2d_10, conv2d_11, conv2d_12], dim=1) + conv2d_10 = conv2d_11 = conv2d_12 = None + reshape_10 = conv_v_img_4.reshape(1, 8, 16, 3136) + conv_v_img_4 = None + conv_v_img_5 = reshape_10.transpose(-1, -2) + reshape_10 = None + EV_hat_4 = q_img_2 * conv_v_img_5 + q_img_2 = conv_v_img_5 = None + EV_hat_5 = torch._C._nn.pad(EV_hat_4, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_4 = None + mul_5 = 0.25 * factor_att_5 + factor_att_5 = None + x_40 = mul_5 + EV_hat_5 + mul_5 = EV_hat_5 = None + transpose_18 = x_40.transpose(1, 2) + x_40 = None + x_41 = transpose_18.reshape(1, 3137, 128) + transpose_18 = None + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_41 = l_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_43 = torch.nn.functional.dropout(x_42, 0.0, False, False) + x_42 = None + x_44 = x_38 + x_43 + x_38 = x_43 = None + x_45 = torch.nn.functional.layer_norm( + x_44, + (128,), + l_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_bias_ + ) = None + x_46 = torch._C._nn.linear( + x_45, + l_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_45 = l_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_47 = torch._C._nn.gelu(x_46, approximate="none") + x_46 = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = torch._C._nn.linear( + x_48, + l_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_48 = l_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_50 = torch.nn.functional.dropout(x_49, 0.0, False, False) + x_49 = None + x_51 = x_44 + x_50 + x_44 = x_50 = None + getitem_38 = x_51[ + (slice(None, None, None), slice(1, None, None), slice(None, None, None)) + ] + x_51 = None + reshape_12 = getitem_38.reshape(1, 56, 56, -1) + getitem_38 = None + permute_3 = reshape_12.permute(0, 3, 1, 2) + reshape_12 = None + x1_nocls = permute_3.contiguous() + permute_3 = None + x_52 = torch.conv2d( + x1_nocls, + l_self_modules_patch_embed2_modules_proj_parameters_weight_, + l_self_modules_patch_embed2_modules_proj_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x1_nocls = ( + l_self_modules_patch_embed2_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed2_modules_proj_parameters_bias_ = None + flatten_4 = x_52.flatten(2) + x_52 = None + x_53 = flatten_4.transpose(1, 2) + flatten_4 = None + x_54 = torch.nn.functional.layer_norm( + x_53, + (256,), + l_self_modules_patch_embed2_modules_norm_parameters_weight_, + l_self_modules_patch_embed2_modules_norm_parameters_bias_, + 1e-05, + ) + x_53 = ( + l_self_modules_patch_embed2_modules_norm_parameters_weight_ + ) = l_self_modules_patch_embed2_modules_norm_parameters_bias_ = None + cls_tokens_1 = l_self_parameters_cls_token2_.expand(1, -1, -1) + l_self_parameters_cls_token2_ = None + x_55 = torch.cat((cls_tokens_1, x_54), dim=1) + cls_tokens_1 = x_54 = None + cls_token_3 = x_55[(slice(None, None, None), slice(None, 1, None))] + img_tokens_3 = x_55[(slice(None, None, None), slice(1, None, None))] + x_55 = None + transpose_20 = img_tokens_3.transpose(1, 2) + img_tokens_3 = None + feat_3 = transpose_20.view(1, 256, 28, 28) + transpose_20 = None + conv2d_14 = torch.conv2d( + feat_3, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 256, + ) + x_56 = conv2d_14 + feat_3 + conv2d_14 = feat_3 = None + flatten_5 = x_56.flatten(2) + x_56 = None + x_57 = flatten_5.transpose(1, 2) + flatten_5 = None + x_58 = torch.cat((cls_token_3, x_57), dim=1) + cls_token_3 = x_57 = None + x_59 = torch.nn.functional.layer_norm( + x_58, + (256,), + l_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_59, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_59 = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_13 = linear_12.reshape(1, 785, 3, 8, 32) + linear_12 = None + qkv_3 = reshape_13.permute(2, 0, 3, 1, 4) + reshape_13 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + k_softmax_3 = k_3.softmax(dim=2) + k_3 = None + transpose_22 = k_softmax_3.transpose(-1, -2) + k_softmax_3 = None + factor_att_6 = transpose_22 @ v_3 + transpose_22 = None + factor_att_7 = q_3 @ factor_att_6 + factor_att_6 = None + q_img_3 = q_3[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_3 = None + v_img_6 = v_3[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_3 = None + transpose_23 = v_img_6.transpose(-1, -2) + v_img_6 = None + v_img_7 = transpose_23.reshape(1, 256, 28, 28) + transpose_23 = None + split_3 = torch.functional.split(v_img_7, [64, 96, 96], dim=1) + v_img_7 = None + getitem_46 = split_3[0] + getitem_47 = split_3[1] + getitem_48 = split_3[2] + split_3 = None + conv2d_15 = torch.conv2d( + getitem_46, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 64, + ) + getitem_46 = None + conv2d_16 = torch.conv2d( + getitem_47, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 96, + ) + getitem_47 = None + conv2d_17 = torch.conv2d( + getitem_48, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + getitem_48 = None + conv_v_img_6 = torch.cat([conv2d_15, conv2d_16, conv2d_17], dim=1) + conv2d_15 = conv2d_16 = conv2d_17 = None + reshape_15 = conv_v_img_6.reshape(1, 8, 32, 784) + conv_v_img_6 = None + conv_v_img_7 = reshape_15.transpose(-1, -2) + reshape_15 = None + EV_hat_6 = q_img_3 * conv_v_img_7 + q_img_3 = conv_v_img_7 = None + EV_hat_7 = torch._C._nn.pad(EV_hat_6, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_6 = None + mul_7 = 0.1767766952966369 * factor_att_7 + factor_att_7 = None + x_60 = mul_7 + EV_hat_7 + mul_7 = EV_hat_7 = None + transpose_25 = x_60.transpose(1, 2) + x_60 = None + x_61 = transpose_25.reshape(1, 785, 256) + transpose_25 = None + x_62 = torch._C._nn.linear( + x_61, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_61 = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_63 = torch.nn.functional.dropout(x_62, 0.0, False, False) + x_62 = None + x_64 = x_58 + x_63 + x_58 = x_63 = None + x_65 = torch.nn.functional.layer_norm( + x_64, + (256,), + l_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_ + ) = None + x_66 = torch._C._nn.linear( + x_65, + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_65 = l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_67 = torch._C._nn.gelu(x_66, approximate="none") + x_66 = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = torch._C._nn.linear( + x_68, + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_68 = l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_71 = x_64 + x_70 + x_64 = x_70 = None + cls_token_4 = x_71[(slice(None, None, None), slice(None, 1, None))] + img_tokens_4 = x_71[(slice(None, None, None), slice(1, None, None))] + x_71 = None + transpose_26 = img_tokens_4.transpose(1, 2) + img_tokens_4 = None + feat_4 = transpose_26.view(1, 256, 28, 28) + transpose_26 = None + conv2d_18 = torch.conv2d( + feat_4, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 256, + ) + x_72 = conv2d_18 + feat_4 + conv2d_18 = feat_4 = None + flatten_6 = x_72.flatten(2) + x_72 = None + x_73 = flatten_6.transpose(1, 2) + flatten_6 = None + x_74 = torch.cat((cls_token_4, x_73), dim=1) + cls_token_4 = x_73 = None + x_75 = torch.nn.functional.layer_norm( + x_74, + (256,), + l_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_75, + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_75 = l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_17 = linear_16.reshape(1, 785, 3, 8, 32) + linear_16 = None + qkv_4 = reshape_17.permute(2, 0, 3, 1, 4) + reshape_17 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + k_softmax_4 = k_4.softmax(dim=2) + k_4 = None + transpose_28 = k_softmax_4.transpose(-1, -2) + k_softmax_4 = None + factor_att_8 = transpose_28 @ v_4 + transpose_28 = None + factor_att_9 = q_4 @ factor_att_8 + factor_att_8 = None + q_img_4 = q_4[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_4 = None + v_img_8 = v_4[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_4 = None + transpose_29 = v_img_8.transpose(-1, -2) + v_img_8 = None + v_img_9 = transpose_29.reshape(1, 256, 28, 28) + transpose_29 = None + split_4 = torch.functional.split(v_img_9, [64, 96, 96], dim=1) + v_img_9 = None + getitem_56 = split_4[0] + getitem_57 = split_4[1] + getitem_58 = split_4[2] + split_4 = None + conv2d_19 = torch.conv2d( + getitem_56, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 64, + ) + getitem_56 = None + conv2d_20 = torch.conv2d( + getitem_57, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 96, + ) + getitem_57 = None + conv2d_21 = torch.conv2d( + getitem_58, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + getitem_58 = None + conv_v_img_8 = torch.cat([conv2d_19, conv2d_20, conv2d_21], dim=1) + conv2d_19 = conv2d_20 = conv2d_21 = None + reshape_19 = conv_v_img_8.reshape(1, 8, 32, 784) + conv_v_img_8 = None + conv_v_img_9 = reshape_19.transpose(-1, -2) + reshape_19 = None + EV_hat_8 = q_img_4 * conv_v_img_9 + q_img_4 = conv_v_img_9 = None + EV_hat_9 = torch._C._nn.pad(EV_hat_8, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_8 = None + mul_9 = 0.1767766952966369 * factor_att_9 + factor_att_9 = None + x_76 = mul_9 + EV_hat_9 + mul_9 = EV_hat_9 = None + transpose_31 = x_76.transpose(1, 2) + x_76 = None + x_77 = transpose_31.reshape(1, 785, 256) + transpose_31 = None + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_77 = l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = x_74 + x_79 + x_74 = x_79 = None + x_81 = torch.nn.functional.layer_norm( + x_80, + (256,), + l_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_ + ) = None + x_82 = torch._C._nn.linear( + x_81, + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_81 = l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_83 = torch._C._nn.gelu(x_82, approximate="none") + x_82 = None + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = torch._C._nn.linear( + x_84, + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_84 = l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_86 = torch.nn.functional.dropout(x_85, 0.0, False, False) + x_85 = None + x_87 = x_80 + x_86 + x_80 = x_86 = None + cls_token_5 = x_87[(slice(None, None, None), slice(None, 1, None))] + img_tokens_5 = x_87[(slice(None, None, None), slice(1, None, None))] + x_87 = None + transpose_32 = img_tokens_5.transpose(1, 2) + img_tokens_5 = None + feat_5 = transpose_32.view(1, 256, 28, 28) + transpose_32 = None + conv2d_22 = torch.conv2d( + feat_5, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 256, + ) + x_88 = conv2d_22 + feat_5 + conv2d_22 = feat_5 = None + flatten_7 = x_88.flatten(2) + x_88 = None + x_89 = flatten_7.transpose(1, 2) + flatten_7 = None + x_90 = torch.cat((cls_token_5, x_89), dim=1) + cls_token_5 = x_89 = None + x_91 = torch.nn.functional.layer_norm( + x_90, + (256,), + l_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_91, + l_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_91 = l_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_21 = linear_20.reshape(1, 785, 3, 8, 32) + linear_20 = None + qkv_5 = reshape_21.permute(2, 0, 3, 1, 4) + reshape_21 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + k_softmax_5 = k_5.softmax(dim=2) + k_5 = None + transpose_34 = k_softmax_5.transpose(-1, -2) + k_softmax_5 = None + factor_att_10 = transpose_34 @ v_5 + transpose_34 = None + factor_att_11 = q_5 @ factor_att_10 + factor_att_10 = None + q_img_5 = q_5[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_5 = None + v_img_10 = v_5[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_5 = None + transpose_35 = v_img_10.transpose(-1, -2) + v_img_10 = None + v_img_11 = transpose_35.reshape(1, 256, 28, 28) + transpose_35 = None + split_5 = torch.functional.split(v_img_11, [64, 96, 96], dim=1) + v_img_11 = None + getitem_66 = split_5[0] + getitem_67 = split_5[1] + getitem_68 = split_5[2] + split_5 = None + conv2d_23 = torch.conv2d( + getitem_66, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 64, + ) + getitem_66 = None + conv2d_24 = torch.conv2d( + getitem_67, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 96, + ) + getitem_67 = None + conv2d_25 = torch.conv2d( + getitem_68, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + getitem_68 = None + conv_v_img_10 = torch.cat([conv2d_23, conv2d_24, conv2d_25], dim=1) + conv2d_23 = conv2d_24 = conv2d_25 = None + reshape_23 = conv_v_img_10.reshape(1, 8, 32, 784) + conv_v_img_10 = None + conv_v_img_11 = reshape_23.transpose(-1, -2) + reshape_23 = None + EV_hat_10 = q_img_5 * conv_v_img_11 + q_img_5 = conv_v_img_11 = None + EV_hat_11 = torch._C._nn.pad(EV_hat_10, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_10 = None + mul_11 = 0.1767766952966369 * factor_att_11 + factor_att_11 = None + x_92 = mul_11 + EV_hat_11 + mul_11 = EV_hat_11 = None + transpose_37 = x_92.transpose(1, 2) + x_92 = None + x_93 = transpose_37.reshape(1, 785, 256) + transpose_37 = None + x_94 = torch._C._nn.linear( + x_93, + l_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_93 = l_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = x_90 + x_95 + x_90 = x_95 = None + x_97 = torch.nn.functional.layer_norm( + x_96, + (256,), + l_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_bias_ + ) = None + x_98 = torch._C._nn.linear( + x_97, + l_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_97 = l_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_99 = torch._C._nn.gelu(x_98, approximate="none") + x_98 = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + x_101 = torch._C._nn.linear( + x_100, + l_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_100 = l_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = x_96 + x_102 + x_96 = x_102 = None + cls_token_6 = x_103[(slice(None, None, None), slice(None, 1, None))] + img_tokens_6 = x_103[(slice(None, None, None), slice(1, None, None))] + x_103 = None + transpose_38 = img_tokens_6.transpose(1, 2) + img_tokens_6 = None + feat_6 = transpose_38.view(1, 256, 28, 28) + transpose_38 = None + conv2d_26 = torch.conv2d( + feat_6, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 256, + ) + x_104 = conv2d_26 + feat_6 + conv2d_26 = feat_6 = None + flatten_8 = x_104.flatten(2) + x_104 = None + x_105 = flatten_8.transpose(1, 2) + flatten_8 = None + x_106 = torch.cat((cls_token_6, x_105), dim=1) + cls_token_6 = x_105 = None + x_107 = torch.nn.functional.layer_norm( + x_106, + (256,), + l_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_107, + l_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_107 = l_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_25 = linear_24.reshape(1, 785, 3, 8, 32) + linear_24 = None + qkv_6 = reshape_25.permute(2, 0, 3, 1, 4) + reshape_25 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + k_softmax_6 = k_6.softmax(dim=2) + k_6 = None + transpose_40 = k_softmax_6.transpose(-1, -2) + k_softmax_6 = None + factor_att_12 = transpose_40 @ v_6 + transpose_40 = None + factor_att_13 = q_6 @ factor_att_12 + factor_att_12 = None + q_img_6 = q_6[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_6 = None + v_img_12 = v_6[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_6 = None + transpose_41 = v_img_12.transpose(-1, -2) + v_img_12 = None + v_img_13 = transpose_41.reshape(1, 256, 28, 28) + transpose_41 = None + split_6 = torch.functional.split(v_img_13, [64, 96, 96], dim=1) + v_img_13 = None + getitem_76 = split_6[0] + getitem_77 = split_6[1] + getitem_78 = split_6[2] + split_6 = None + conv2d_27 = torch.conv2d( + getitem_76, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 64, + ) + getitem_76 = None + conv2d_28 = torch.conv2d( + getitem_77, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 96, + ) + getitem_77 = None + conv2d_29 = torch.conv2d( + getitem_78, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + getitem_78 = None + conv_v_img_12 = torch.cat([conv2d_27, conv2d_28, conv2d_29], dim=1) + conv2d_27 = conv2d_28 = conv2d_29 = None + reshape_27 = conv_v_img_12.reshape(1, 8, 32, 784) + conv_v_img_12 = None + conv_v_img_13 = reshape_27.transpose(-1, -2) + reshape_27 = None + EV_hat_12 = q_img_6 * conv_v_img_13 + q_img_6 = conv_v_img_13 = None + EV_hat_13 = torch._C._nn.pad(EV_hat_12, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_12 = None + mul_13 = 0.1767766952966369 * factor_att_13 + factor_att_13 = None + x_108 = mul_13 + EV_hat_13 + mul_13 = EV_hat_13 = None + transpose_43 = x_108.transpose(1, 2) + x_108 = None + x_109 = transpose_43.reshape(1, 785, 256) + transpose_43 = None + x_110 = torch._C._nn.linear( + x_109, + l_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_109 = l_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_111 = torch.nn.functional.dropout(x_110, 0.0, False, False) + x_110 = None + x_112 = x_106 + x_111 + x_106 = x_111 = None + x_113 = torch.nn.functional.layer_norm( + x_112, + (256,), + l_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_bias_ + ) = None + x_114 = torch._C._nn.linear( + x_113, + l_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_113 = l_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_115 = torch._C._nn.gelu(x_114, approximate="none") + x_114 = None + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + x_117 = torch._C._nn.linear( + x_116, + l_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_116 = l_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = x_112 + x_118 + x_112 = x_118 = None + cls_token_7 = x_119[(slice(None, None, None), slice(None, 1, None))] + img_tokens_7 = x_119[(slice(None, None, None), slice(1, None, None))] + x_119 = None + transpose_44 = img_tokens_7.transpose(1, 2) + img_tokens_7 = None + feat_7 = transpose_44.view(1, 256, 28, 28) + transpose_44 = None + conv2d_30 = torch.conv2d( + feat_7, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 256, + ) + x_120 = conv2d_30 + feat_7 + conv2d_30 = feat_7 = None + flatten_9 = x_120.flatten(2) + x_120 = None + x_121 = flatten_9.transpose(1, 2) + flatten_9 = None + x_122 = torch.cat((cls_token_7, x_121), dim=1) + cls_token_7 = x_121 = None + x_123 = torch.nn.functional.layer_norm( + x_122, + (256,), + l_self_modules_serial_blocks2_modules_4_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks2_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_123, + l_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_123 = l_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_29 = linear_28.reshape(1, 785, 3, 8, 32) + linear_28 = None + qkv_7 = reshape_29.permute(2, 0, 3, 1, 4) + reshape_29 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + k_softmax_7 = k_7.softmax(dim=2) + k_7 = None + transpose_46 = k_softmax_7.transpose(-1, -2) + k_softmax_7 = None + factor_att_14 = transpose_46 @ v_7 + transpose_46 = None + factor_att_15 = q_7 @ factor_att_14 + factor_att_14 = None + q_img_7 = q_7[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_7 = None + v_img_14 = v_7[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_7 = None + transpose_47 = v_img_14.transpose(-1, -2) + v_img_14 = None + v_img_15 = transpose_47.reshape(1, 256, 28, 28) + transpose_47 = None + split_7 = torch.functional.split(v_img_15, [64, 96, 96], dim=1) + v_img_15 = None + getitem_86 = split_7[0] + getitem_87 = split_7[1] + getitem_88 = split_7[2] + split_7 = None + conv2d_31 = torch.conv2d( + getitem_86, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 64, + ) + getitem_86 = None + conv2d_32 = torch.conv2d( + getitem_87, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 96, + ) + getitem_87 = None + conv2d_33 = torch.conv2d( + getitem_88, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + getitem_88 = None + conv_v_img_14 = torch.cat([conv2d_31, conv2d_32, conv2d_33], dim=1) + conv2d_31 = conv2d_32 = conv2d_33 = None + reshape_31 = conv_v_img_14.reshape(1, 8, 32, 784) + conv_v_img_14 = None + conv_v_img_15 = reshape_31.transpose(-1, -2) + reshape_31 = None + EV_hat_14 = q_img_7 * conv_v_img_15 + q_img_7 = conv_v_img_15 = None + EV_hat_15 = torch._C._nn.pad(EV_hat_14, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_14 = None + mul_15 = 0.1767766952966369 * factor_att_15 + factor_att_15 = None + x_124 = mul_15 + EV_hat_15 + mul_15 = EV_hat_15 = None + transpose_49 = x_124.transpose(1, 2) + x_124 = None + x_125 = transpose_49.reshape(1, 785, 256) + transpose_49 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_125 = l_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = x_122 + x_127 + x_122 = x_127 = None + x_129 = torch.nn.functional.layer_norm( + x_128, + (256,), + l_self_modules_serial_blocks2_modules_4_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks2_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_4_modules_norm2_parameters_bias_ + ) = None + x_130 = torch._C._nn.linear( + x_129, + l_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_129 = l_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_131 = torch._C._nn.gelu(x_130, approximate="none") + x_130 = None + x_132 = torch.nn.functional.dropout(x_131, 0.0, False, False) + x_131 = None + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_132 = l_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_134 = torch.nn.functional.dropout(x_133, 0.0, False, False) + x_133 = None + x_135 = x_128 + x_134 + x_128 = x_134 = None + cls_token_8 = x_135[(slice(None, None, None), slice(None, 1, None))] + img_tokens_8 = x_135[(slice(None, None, None), slice(1, None, None))] + x_135 = None + transpose_50 = img_tokens_8.transpose(1, 2) + img_tokens_8 = None + feat_8 = transpose_50.view(1, 256, 28, 28) + transpose_50 = None + conv2d_34 = torch.conv2d( + feat_8, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 256, + ) + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_ = (None) + x_136 = conv2d_34 + feat_8 + conv2d_34 = feat_8 = None + flatten_10 = x_136.flatten(2) + x_136 = None + x_137 = flatten_10.transpose(1, 2) + flatten_10 = None + x_138 = torch.cat((cls_token_8, x_137), dim=1) + cls_token_8 = x_137 = None + x_139 = torch.nn.functional.layer_norm( + x_138, + (256,), + l_self_modules_serial_blocks2_modules_5_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks2_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_139, + l_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_139 = l_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_33 = linear_32.reshape(1, 785, 3, 8, 32) + linear_32 = None + qkv_8 = reshape_33.permute(2, 0, 3, 1, 4) + reshape_33 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + k_softmax_8 = k_8.softmax(dim=2) + k_8 = None + transpose_52 = k_softmax_8.transpose(-1, -2) + k_softmax_8 = None + factor_att_16 = transpose_52 @ v_8 + transpose_52 = None + factor_att_17 = q_8 @ factor_att_16 + factor_att_16 = None + q_img_8 = q_8[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_8 = None + v_img_16 = v_8[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_8 = None + transpose_53 = v_img_16.transpose(-1, -2) + v_img_16 = None + v_img_17 = transpose_53.reshape(1, 256, 28, 28) + transpose_53 = None + split_8 = torch.functional.split(v_img_17, [64, 96, 96], dim=1) + v_img_17 = None + getitem_96 = split_8[0] + getitem_97 = split_8[1] + getitem_98 = split_8[2] + split_8 = None + conv2d_35 = torch.conv2d( + getitem_96, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 64, + ) + getitem_96 = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = (None) + conv2d_36 = torch.conv2d( + getitem_97, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 96, + ) + getitem_97 = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = (None) + conv2d_37 = torch.conv2d( + getitem_98, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + getitem_98 = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = (None) + conv_v_img_16 = torch.cat([conv2d_35, conv2d_36, conv2d_37], dim=1) + conv2d_35 = conv2d_36 = conv2d_37 = None + reshape_35 = conv_v_img_16.reshape(1, 8, 32, 784) + conv_v_img_16 = None + conv_v_img_17 = reshape_35.transpose(-1, -2) + reshape_35 = None + EV_hat_16 = q_img_8 * conv_v_img_17 + q_img_8 = conv_v_img_17 = None + EV_hat_17 = torch._C._nn.pad(EV_hat_16, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_16 = None + mul_17 = 0.1767766952966369 * factor_att_17 + factor_att_17 = None + x_140 = mul_17 + EV_hat_17 + mul_17 = EV_hat_17 = None + transpose_55 = x_140.transpose(1, 2) + x_140 = None + x_141 = transpose_55.reshape(1, 785, 256) + transpose_55 = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_141 = l_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_143 = torch.nn.functional.dropout(x_142, 0.0, False, False) + x_142 = None + x_144 = x_138 + x_143 + x_138 = x_143 = None + x_145 = torch.nn.functional.layer_norm( + x_144, + (256,), + l_self_modules_serial_blocks2_modules_5_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks2_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_5_modules_norm2_parameters_bias_ + ) = None + x_146 = torch._C._nn.linear( + x_145, + l_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_145 = l_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_147 = torch._C._nn.gelu(x_146, approximate="none") + x_146 = None + x_148 = torch.nn.functional.dropout(x_147, 0.0, False, False) + x_147 = None + x_149 = torch._C._nn.linear( + x_148, + l_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_148 = l_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) + x_149 = None + x_151 = x_144 + x_150 + x_144 = x_150 = None + getitem_99 = x_151[ + (slice(None, None, None), slice(1, None, None), slice(None, None, None)) + ] + x_151 = None + reshape_37 = getitem_99.reshape(1, 28, 28, -1) + getitem_99 = None + permute_10 = reshape_37.permute(0, 3, 1, 2) + reshape_37 = None + x2_nocls = permute_10.contiguous() + permute_10 = None + x_152 = torch.conv2d( + x2_nocls, + l_self_modules_patch_embed3_modules_proj_parameters_weight_, + l_self_modules_patch_embed3_modules_proj_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x2_nocls = ( + l_self_modules_patch_embed3_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed3_modules_proj_parameters_bias_ = None + flatten_11 = x_152.flatten(2) + x_152 = None + x_153 = flatten_11.transpose(1, 2) + flatten_11 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (320,), + l_self_modules_patch_embed3_modules_norm_parameters_weight_, + l_self_modules_patch_embed3_modules_norm_parameters_bias_, + 1e-05, + ) + x_153 = ( + l_self_modules_patch_embed3_modules_norm_parameters_weight_ + ) = l_self_modules_patch_embed3_modules_norm_parameters_bias_ = None + cls_tokens_2 = l_self_parameters_cls_token3_.expand(1, -1, -1) + l_self_parameters_cls_token3_ = None + x_155 = torch.cat((cls_tokens_2, x_154), dim=1) + cls_tokens_2 = x_154 = None + cls_token_9 = x_155[(slice(None, None, None), slice(None, 1, None))] + img_tokens_9 = x_155[(slice(None, None, None), slice(1, None, None))] + x_155 = None + transpose_57 = img_tokens_9.transpose(1, 2) + img_tokens_9 = None + feat_9 = transpose_57.view(1, 320, 14, 14) + transpose_57 = None + conv2d_39 = torch.conv2d( + feat_9, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 320, + ) + x_156 = conv2d_39 + feat_9 + conv2d_39 = feat_9 = None + flatten_12 = x_156.flatten(2) + x_156 = None + x_157 = flatten_12.transpose(1, 2) + flatten_12 = None + x_158 = torch.cat((cls_token_9, x_157), dim=1) + cls_token_9 = x_157 = None + x_159 = torch.nn.functional.layer_norm( + x_158, + (320,), + l_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_159, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_159 = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_38 = linear_36.reshape(1, 197, 3, 8, 40) + linear_36 = None + qkv_9 = reshape_38.permute(2, 0, 3, 1, 4) + reshape_38 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + k_softmax_9 = k_9.softmax(dim=2) + k_9 = None + transpose_59 = k_softmax_9.transpose(-1, -2) + k_softmax_9 = None + factor_att_18 = transpose_59 @ v_9 + transpose_59 = None + factor_att_19 = q_9 @ factor_att_18 + factor_att_18 = None + q_img_9 = q_9[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_9 = None + v_img_18 = v_9[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_9 = None + transpose_60 = v_img_18.transpose(-1, -2) + v_img_18 = None + v_img_19 = transpose_60.reshape(1, 320, 14, 14) + transpose_60 = None + split_9 = torch.functional.split(v_img_19, [80, 120, 120], dim=1) + v_img_19 = None + getitem_107 = split_9[0] + getitem_108 = split_9[1] + getitem_109 = split_9[2] + split_9 = None + conv2d_40 = torch.conv2d( + getitem_107, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 80, + ) + getitem_107 = None + conv2d_41 = torch.conv2d( + getitem_108, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 120, + ) + getitem_108 = None + conv2d_42 = torch.conv2d( + getitem_109, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 120, + ) + getitem_109 = None + conv_v_img_18 = torch.cat([conv2d_40, conv2d_41, conv2d_42], dim=1) + conv2d_40 = conv2d_41 = conv2d_42 = None + reshape_40 = conv_v_img_18.reshape(1, 8, 40, 196) + conv_v_img_18 = None + conv_v_img_19 = reshape_40.transpose(-1, -2) + reshape_40 = None + EV_hat_18 = q_img_9 * conv_v_img_19 + q_img_9 = conv_v_img_19 = None + EV_hat_19 = torch._C._nn.pad(EV_hat_18, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_18 = None + mul_19 = 0.15811388300841897 * factor_att_19 + factor_att_19 = None + x_160 = mul_19 + EV_hat_19 + mul_19 = EV_hat_19 = None + transpose_62 = x_160.transpose(1, 2) + x_160 = None + x_161 = transpose_62.reshape(1, 197, 320) + transpose_62 = None + x_162 = torch._C._nn.linear( + x_161, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_161 = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_163 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + x_164 = x_158 + x_163 + x_158 = x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (320,), + l_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_ + ) = None + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166, approximate="none") + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_169 = torch._C._nn.linear( + x_168, + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_168 = l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_170 = torch.nn.functional.dropout(x_169, 0.0, False, False) + x_169 = None + x_171 = x_164 + x_170 + x_164 = x_170 = None + cls_token_10 = x_171[(slice(None, None, None), slice(None, 1, None))] + img_tokens_10 = x_171[(slice(None, None, None), slice(1, None, None))] + x_171 = None + transpose_63 = img_tokens_10.transpose(1, 2) + img_tokens_10 = None + feat_10 = transpose_63.view(1, 320, 14, 14) + transpose_63 = None + conv2d_43 = torch.conv2d( + feat_10, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 320, + ) + x_172 = conv2d_43 + feat_10 + conv2d_43 = feat_10 = None + flatten_13 = x_172.flatten(2) + x_172 = None + x_173 = flatten_13.transpose(1, 2) + flatten_13 = None + x_174 = torch.cat((cls_token_10, x_173), dim=1) + cls_token_10 = x_173 = None + x_175 = torch.nn.functional.layer_norm( + x_174, + (320,), + l_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_175, + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_175 = l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_42 = linear_40.reshape(1, 197, 3, 8, 40) + linear_40 = None + qkv_10 = reshape_42.permute(2, 0, 3, 1, 4) + reshape_42 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + k_softmax_10 = k_10.softmax(dim=2) + k_10 = None + transpose_65 = k_softmax_10.transpose(-1, -2) + k_softmax_10 = None + factor_att_20 = transpose_65 @ v_10 + transpose_65 = None + factor_att_21 = q_10 @ factor_att_20 + factor_att_20 = None + q_img_10 = q_10[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_10 = None + v_img_20 = v_10[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_10 = None + transpose_66 = v_img_20.transpose(-1, -2) + v_img_20 = None + v_img_21 = transpose_66.reshape(1, 320, 14, 14) + transpose_66 = None + split_10 = torch.functional.split(v_img_21, [80, 120, 120], dim=1) + v_img_21 = None + getitem_117 = split_10[0] + getitem_118 = split_10[1] + getitem_119 = split_10[2] + split_10 = None + conv2d_44 = torch.conv2d( + getitem_117, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 80, + ) + getitem_117 = None + conv2d_45 = torch.conv2d( + getitem_118, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 120, + ) + getitem_118 = None + conv2d_46 = torch.conv2d( + getitem_119, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 120, + ) + getitem_119 = None + conv_v_img_20 = torch.cat([conv2d_44, conv2d_45, conv2d_46], dim=1) + conv2d_44 = conv2d_45 = conv2d_46 = None + reshape_44 = conv_v_img_20.reshape(1, 8, 40, 196) + conv_v_img_20 = None + conv_v_img_21 = reshape_44.transpose(-1, -2) + reshape_44 = None + EV_hat_20 = q_img_10 * conv_v_img_21 + q_img_10 = conv_v_img_21 = None + EV_hat_21 = torch._C._nn.pad(EV_hat_20, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_20 = None + mul_21 = 0.15811388300841897 * factor_att_21 + factor_att_21 = None + x_176 = mul_21 + EV_hat_21 + mul_21 = EV_hat_21 = None + transpose_68 = x_176.transpose(1, 2) + x_176 = None + x_177 = transpose_68.reshape(1, 197, 320) + transpose_68 = None + x_178 = torch._C._nn.linear( + x_177, + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_177 = l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_180 = x_174 + x_179 + x_174 = x_179 = None + x_181 = torch.nn.functional.layer_norm( + x_180, + (320,), + l_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_ + ) = None + x_182 = torch._C._nn.linear( + x_181, + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_181 = l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_183 = torch._C._nn.gelu(x_182, approximate="none") + x_182 = None + x_184 = torch.nn.functional.dropout(x_183, 0.0, False, False) + x_183 = None + x_185 = torch._C._nn.linear( + x_184, + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_184 = l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_186 = torch.nn.functional.dropout(x_185, 0.0, False, False) + x_185 = None + x_187 = x_180 + x_186 + x_180 = x_186 = None + cls_token_11 = x_187[(slice(None, None, None), slice(None, 1, None))] + img_tokens_11 = x_187[(slice(None, None, None), slice(1, None, None))] + x_187 = None + transpose_69 = img_tokens_11.transpose(1, 2) + img_tokens_11 = None + feat_11 = transpose_69.view(1, 320, 14, 14) + transpose_69 = None + conv2d_47 = torch.conv2d( + feat_11, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 320, + ) + x_188 = conv2d_47 + feat_11 + conv2d_47 = feat_11 = None + flatten_14 = x_188.flatten(2) + x_188 = None + x_189 = flatten_14.transpose(1, 2) + flatten_14 = None + x_190 = torch.cat((cls_token_11, x_189), dim=1) + cls_token_11 = x_189 = None + x_191 = torch.nn.functional.layer_norm( + x_190, + (320,), + l_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_191, + l_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_191 = l_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_46 = linear_44.reshape(1, 197, 3, 8, 40) + linear_44 = None + qkv_11 = reshape_46.permute(2, 0, 3, 1, 4) + reshape_46 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + k_softmax_11 = k_11.softmax(dim=2) + k_11 = None + transpose_71 = k_softmax_11.transpose(-1, -2) + k_softmax_11 = None + factor_att_22 = transpose_71 @ v_11 + transpose_71 = None + factor_att_23 = q_11 @ factor_att_22 + factor_att_22 = None + q_img_11 = q_11[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_11 = None + v_img_22 = v_11[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_11 = None + transpose_72 = v_img_22.transpose(-1, -2) + v_img_22 = None + v_img_23 = transpose_72.reshape(1, 320, 14, 14) + transpose_72 = None + split_11 = torch.functional.split(v_img_23, [80, 120, 120], dim=1) + v_img_23 = None + getitem_127 = split_11[0] + getitem_128 = split_11[1] + getitem_129 = split_11[2] + split_11 = None + conv2d_48 = torch.conv2d( + getitem_127, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 80, + ) + getitem_127 = None + conv2d_49 = torch.conv2d( + getitem_128, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 120, + ) + getitem_128 = None + conv2d_50 = torch.conv2d( + getitem_129, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 120, + ) + getitem_129 = None + conv_v_img_22 = torch.cat([conv2d_48, conv2d_49, conv2d_50], dim=1) + conv2d_48 = conv2d_49 = conv2d_50 = None + reshape_48 = conv_v_img_22.reshape(1, 8, 40, 196) + conv_v_img_22 = None + conv_v_img_23 = reshape_48.transpose(-1, -2) + reshape_48 = None + EV_hat_22 = q_img_11 * conv_v_img_23 + q_img_11 = conv_v_img_23 = None + EV_hat_23 = torch._C._nn.pad(EV_hat_22, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_22 = None + mul_23 = 0.15811388300841897 * factor_att_23 + factor_att_23 = None + x_192 = mul_23 + EV_hat_23 + mul_23 = EV_hat_23 = None + transpose_74 = x_192.transpose(1, 2) + x_192 = None + x_193 = transpose_74.reshape(1, 197, 320) + transpose_74 = None + x_194 = torch._C._nn.linear( + x_193, + l_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_193 = l_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_195 = torch.nn.functional.dropout(x_194, 0.0, False, False) + x_194 = None + x_196 = x_190 + x_195 + x_190 = x_195 = None + x_197 = torch.nn.functional.layer_norm( + x_196, + (320,), + l_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_bias_ + ) = None + x_198 = torch._C._nn.linear( + x_197, + l_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_197 = l_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_199 = torch._C._nn.gelu(x_198, approximate="none") + x_198 = None + x_200 = torch.nn.functional.dropout(x_199, 0.0, False, False) + x_199 = None + x_201 = torch._C._nn.linear( + x_200, + l_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_200 = l_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_202 = torch.nn.functional.dropout(x_201, 0.0, False, False) + x_201 = None + x_203 = x_196 + x_202 + x_196 = x_202 = None + cls_token_12 = x_203[(slice(None, None, None), slice(None, 1, None))] + img_tokens_12 = x_203[(slice(None, None, None), slice(1, None, None))] + x_203 = None + transpose_75 = img_tokens_12.transpose(1, 2) + img_tokens_12 = None + feat_12 = transpose_75.view(1, 320, 14, 14) + transpose_75 = None + conv2d_51 = torch.conv2d( + feat_12, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 320, + ) + x_204 = conv2d_51 + feat_12 + conv2d_51 = feat_12 = None + flatten_15 = x_204.flatten(2) + x_204 = None + x_205 = flatten_15.transpose(1, 2) + flatten_15 = None + x_206 = torch.cat((cls_token_12, x_205), dim=1) + cls_token_12 = x_205 = None + x_207 = torch.nn.functional.layer_norm( + x_206, + (320,), + l_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_48 = torch._C._nn.linear( + x_207, + l_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_207 = l_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_50 = linear_48.reshape(1, 197, 3, 8, 40) + linear_48 = None + qkv_12 = reshape_50.permute(2, 0, 3, 1, 4) + reshape_50 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_12 = unbind_12[0] + k_12 = unbind_12[1] + v_12 = unbind_12[2] + unbind_12 = None + k_softmax_12 = k_12.softmax(dim=2) + k_12 = None + transpose_77 = k_softmax_12.transpose(-1, -2) + k_softmax_12 = None + factor_att_24 = transpose_77 @ v_12 + transpose_77 = None + factor_att_25 = q_12 @ factor_att_24 + factor_att_24 = None + q_img_12 = q_12[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_12 = None + v_img_24 = v_12[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_12 = None + transpose_78 = v_img_24.transpose(-1, -2) + v_img_24 = None + v_img_25 = transpose_78.reshape(1, 320, 14, 14) + transpose_78 = None + split_12 = torch.functional.split(v_img_25, [80, 120, 120], dim=1) + v_img_25 = None + getitem_137 = split_12[0] + getitem_138 = split_12[1] + getitem_139 = split_12[2] + split_12 = None + conv2d_52 = torch.conv2d( + getitem_137, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 80, + ) + getitem_137 = None + conv2d_53 = torch.conv2d( + getitem_138, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 120, + ) + getitem_138 = None + conv2d_54 = torch.conv2d( + getitem_139, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 120, + ) + getitem_139 = None + conv_v_img_24 = torch.cat([conv2d_52, conv2d_53, conv2d_54], dim=1) + conv2d_52 = conv2d_53 = conv2d_54 = None + reshape_52 = conv_v_img_24.reshape(1, 8, 40, 196) + conv_v_img_24 = None + conv_v_img_25 = reshape_52.transpose(-1, -2) + reshape_52 = None + EV_hat_24 = q_img_12 * conv_v_img_25 + q_img_12 = conv_v_img_25 = None + EV_hat_25 = torch._C._nn.pad(EV_hat_24, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_24 = None + mul_25 = 0.15811388300841897 * factor_att_25 + factor_att_25 = None + x_208 = mul_25 + EV_hat_25 + mul_25 = EV_hat_25 = None + transpose_80 = x_208.transpose(1, 2) + x_208 = None + x_209 = transpose_80.reshape(1, 197, 320) + transpose_80 = None + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_209 = l_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_211 = torch.nn.functional.dropout(x_210, 0.0, False, False) + x_210 = None + x_212 = x_206 + x_211 + x_206 = x_211 = None + x_213 = torch.nn.functional.layer_norm( + x_212, + (320,), + l_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_bias_ + ) = None + x_214 = torch._C._nn.linear( + x_213, + l_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_213 = l_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_215 = torch._C._nn.gelu(x_214, approximate="none") + x_214 = None + x_216 = torch.nn.functional.dropout(x_215, 0.0, False, False) + x_215 = None + x_217 = torch._C._nn.linear( + x_216, + l_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_216 = l_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_218 = torch.nn.functional.dropout(x_217, 0.0, False, False) + x_217 = None + x_219 = x_212 + x_218 + x_212 = x_218 = None + cls_token_13 = x_219[(slice(None, None, None), slice(None, 1, None))] + img_tokens_13 = x_219[(slice(None, None, None), slice(1, None, None))] + x_219 = None + transpose_81 = img_tokens_13.transpose(1, 2) + img_tokens_13 = None + feat_13 = transpose_81.view(1, 320, 14, 14) + transpose_81 = None + conv2d_55 = torch.conv2d( + feat_13, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 320, + ) + x_220 = conv2d_55 + feat_13 + conv2d_55 = feat_13 = None + flatten_16 = x_220.flatten(2) + x_220 = None + x_221 = flatten_16.transpose(1, 2) + flatten_16 = None + x_222 = torch.cat((cls_token_13, x_221), dim=1) + cls_token_13 = x_221 = None + x_223 = torch.nn.functional.layer_norm( + x_222, + (320,), + l_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_52 = torch._C._nn.linear( + x_223, + l_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_223 = l_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_54 = linear_52.reshape(1, 197, 3, 8, 40) + linear_52 = None + qkv_13 = reshape_54.permute(2, 0, 3, 1, 4) + reshape_54 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_13 = unbind_13[0] + k_13 = unbind_13[1] + v_13 = unbind_13[2] + unbind_13 = None + k_softmax_13 = k_13.softmax(dim=2) + k_13 = None + transpose_83 = k_softmax_13.transpose(-1, -2) + k_softmax_13 = None + factor_att_26 = transpose_83 @ v_13 + transpose_83 = None + factor_att_27 = q_13 @ factor_att_26 + factor_att_26 = None + q_img_13 = q_13[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_13 = None + v_img_26 = v_13[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_13 = None + transpose_84 = v_img_26.transpose(-1, -2) + v_img_26 = None + v_img_27 = transpose_84.reshape(1, 320, 14, 14) + transpose_84 = None + split_13 = torch.functional.split(v_img_27, [80, 120, 120], dim=1) + v_img_27 = None + getitem_147 = split_13[0] + getitem_148 = split_13[1] + getitem_149 = split_13[2] + split_13 = None + conv2d_56 = torch.conv2d( + getitem_147, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 80, + ) + getitem_147 = None + conv2d_57 = torch.conv2d( + getitem_148, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 120, + ) + getitem_148 = None + conv2d_58 = torch.conv2d( + getitem_149, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 120, + ) + getitem_149 = None + conv_v_img_26 = torch.cat([conv2d_56, conv2d_57, conv2d_58], dim=1) + conv2d_56 = conv2d_57 = conv2d_58 = None + reshape_56 = conv_v_img_26.reshape(1, 8, 40, 196) + conv_v_img_26 = None + conv_v_img_27 = reshape_56.transpose(-1, -2) + reshape_56 = None + EV_hat_26 = q_img_13 * conv_v_img_27 + q_img_13 = conv_v_img_27 = None + EV_hat_27 = torch._C._nn.pad(EV_hat_26, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_26 = None + mul_27 = 0.15811388300841897 * factor_att_27 + factor_att_27 = None + x_224 = mul_27 + EV_hat_27 + mul_27 = EV_hat_27 = None + transpose_86 = x_224.transpose(1, 2) + x_224 = None + x_225 = transpose_86.reshape(1, 197, 320) + transpose_86 = None + x_226 = torch._C._nn.linear( + x_225, + l_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_225 = l_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) + x_226 = None + x_228 = x_222 + x_227 + x_222 = x_227 = None + x_229 = torch.nn.functional.layer_norm( + x_228, + (320,), + l_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_bias_ + ) = None + x_230 = torch._C._nn.linear( + x_229, + l_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_229 = l_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_231 = torch._C._nn.gelu(x_230, approximate="none") + x_230 = None + x_232 = torch.nn.functional.dropout(x_231, 0.0, False, False) + x_231 = None + x_233 = torch._C._nn.linear( + x_232, + l_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_232 = l_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_235 = x_228 + x_234 + x_228 = x_234 = None + cls_token_14 = x_235[(slice(None, None, None), slice(None, 1, None))] + img_tokens_14 = x_235[(slice(None, None, None), slice(1, None, None))] + x_235 = None + transpose_87 = img_tokens_14.transpose(1, 2) + img_tokens_14 = None + feat_14 = transpose_87.view(1, 320, 14, 14) + transpose_87 = None + conv2d_59 = torch.conv2d( + feat_14, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 320, + ) + x_236 = conv2d_59 + feat_14 + conv2d_59 = feat_14 = None + flatten_17 = x_236.flatten(2) + x_236 = None + x_237 = flatten_17.transpose(1, 2) + flatten_17 = None + x_238 = torch.cat((cls_token_14, x_237), dim=1) + cls_token_14 = x_237 = None + x_239 = torch.nn.functional.layer_norm( + x_238, + (320,), + l_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_56 = torch._C._nn.linear( + x_239, + l_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_239 = l_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_58 = linear_56.reshape(1, 197, 3, 8, 40) + linear_56 = None + qkv_14 = reshape_58.permute(2, 0, 3, 1, 4) + reshape_58 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_14 = unbind_14[0] + k_14 = unbind_14[1] + v_14 = unbind_14[2] + unbind_14 = None + k_softmax_14 = k_14.softmax(dim=2) + k_14 = None + transpose_89 = k_softmax_14.transpose(-1, -2) + k_softmax_14 = None + factor_att_28 = transpose_89 @ v_14 + transpose_89 = None + factor_att_29 = q_14 @ factor_att_28 + factor_att_28 = None + q_img_14 = q_14[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_14 = None + v_img_28 = v_14[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_14 = None + transpose_90 = v_img_28.transpose(-1, -2) + v_img_28 = None + v_img_29 = transpose_90.reshape(1, 320, 14, 14) + transpose_90 = None + split_14 = torch.functional.split(v_img_29, [80, 120, 120], dim=1) + v_img_29 = None + getitem_157 = split_14[0] + getitem_158 = split_14[1] + getitem_159 = split_14[2] + split_14 = None + conv2d_60 = torch.conv2d( + getitem_157, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 80, + ) + getitem_157 = None + conv2d_61 = torch.conv2d( + getitem_158, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 120, + ) + getitem_158 = None + conv2d_62 = torch.conv2d( + getitem_159, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 120, + ) + getitem_159 = None + conv_v_img_28 = torch.cat([conv2d_60, conv2d_61, conv2d_62], dim=1) + conv2d_60 = conv2d_61 = conv2d_62 = None + reshape_60 = conv_v_img_28.reshape(1, 8, 40, 196) + conv_v_img_28 = None + conv_v_img_29 = reshape_60.transpose(-1, -2) + reshape_60 = None + EV_hat_28 = q_img_14 * conv_v_img_29 + q_img_14 = conv_v_img_29 = None + EV_hat_29 = torch._C._nn.pad(EV_hat_28, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_28 = None + mul_29 = 0.15811388300841897 * factor_att_29 + factor_att_29 = None + x_240 = mul_29 + EV_hat_29 + mul_29 = EV_hat_29 = None + transpose_92 = x_240.transpose(1, 2) + x_240 = None + x_241 = transpose_92.reshape(1, 197, 320) + transpose_92 = None + x_242 = torch._C._nn.linear( + x_241, + l_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_241 = l_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_243 = torch.nn.functional.dropout(x_242, 0.0, False, False) + x_242 = None + x_244 = x_238 + x_243 + x_238 = x_243 = None + x_245 = torch.nn.functional.layer_norm( + x_244, + (320,), + l_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_bias_ + ) = None + x_246 = torch._C._nn.linear( + x_245, + l_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_245 = l_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_247 = torch._C._nn.gelu(x_246, approximate="none") + x_246 = None + x_248 = torch.nn.functional.dropout(x_247, 0.0, False, False) + x_247 = None + x_249 = torch._C._nn.linear( + x_248, + l_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_248 = l_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_250 = torch.nn.functional.dropout(x_249, 0.0, False, False) + x_249 = None + x_251 = x_244 + x_250 + x_244 = x_250 = None + cls_token_15 = x_251[(slice(None, None, None), slice(None, 1, None))] + img_tokens_15 = x_251[(slice(None, None, None), slice(1, None, None))] + x_251 = None + transpose_93 = img_tokens_15.transpose(1, 2) + img_tokens_15 = None + feat_15 = transpose_93.view(1, 320, 14, 14) + transpose_93 = None + conv2d_63 = torch.conv2d( + feat_15, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 320, + ) + x_252 = conv2d_63 + feat_15 + conv2d_63 = feat_15 = None + flatten_18 = x_252.flatten(2) + x_252 = None + x_253 = flatten_18.transpose(1, 2) + flatten_18 = None + x_254 = torch.cat((cls_token_15, x_253), dim=1) + cls_token_15 = x_253 = None + x_255 = torch.nn.functional.layer_norm( + x_254, + (320,), + l_self_modules_serial_blocks3_modules_6_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks3_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_60 = torch._C._nn.linear( + x_255, + l_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_255 = l_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_62 = linear_60.reshape(1, 197, 3, 8, 40) + linear_60 = None + qkv_15 = reshape_62.permute(2, 0, 3, 1, 4) + reshape_62 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_15 = unbind_15[0] + k_15 = unbind_15[1] + v_15 = unbind_15[2] + unbind_15 = None + k_softmax_15 = k_15.softmax(dim=2) + k_15 = None + transpose_95 = k_softmax_15.transpose(-1, -2) + k_softmax_15 = None + factor_att_30 = transpose_95 @ v_15 + transpose_95 = None + factor_att_31 = q_15 @ factor_att_30 + factor_att_30 = None + q_img_15 = q_15[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_15 = None + v_img_30 = v_15[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_15 = None + transpose_96 = v_img_30.transpose(-1, -2) + v_img_30 = None + v_img_31 = transpose_96.reshape(1, 320, 14, 14) + transpose_96 = None + split_15 = torch.functional.split(v_img_31, [80, 120, 120], dim=1) + v_img_31 = None + getitem_167 = split_15[0] + getitem_168 = split_15[1] + getitem_169 = split_15[2] + split_15 = None + conv2d_64 = torch.conv2d( + getitem_167, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 80, + ) + getitem_167 = None + conv2d_65 = torch.conv2d( + getitem_168, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 120, + ) + getitem_168 = None + conv2d_66 = torch.conv2d( + getitem_169, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 120, + ) + getitem_169 = None + conv_v_img_30 = torch.cat([conv2d_64, conv2d_65, conv2d_66], dim=1) + conv2d_64 = conv2d_65 = conv2d_66 = None + reshape_64 = conv_v_img_30.reshape(1, 8, 40, 196) + conv_v_img_30 = None + conv_v_img_31 = reshape_64.transpose(-1, -2) + reshape_64 = None + EV_hat_30 = q_img_15 * conv_v_img_31 + q_img_15 = conv_v_img_31 = None + EV_hat_31 = torch._C._nn.pad(EV_hat_30, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_30 = None + mul_31 = 0.15811388300841897 * factor_att_31 + factor_att_31 = None + x_256 = mul_31 + EV_hat_31 + mul_31 = EV_hat_31 = None + transpose_98 = x_256.transpose(1, 2) + x_256 = None + x_257 = transpose_98.reshape(1, 197, 320) + transpose_98 = None + x_258 = torch._C._nn.linear( + x_257, + l_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_257 = l_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = x_254 + x_259 + x_254 = x_259 = None + x_261 = torch.nn.functional.layer_norm( + x_260, + (320,), + l_self_modules_serial_blocks3_modules_6_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks3_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_6_modules_norm2_parameters_bias_ + ) = None + x_262 = torch._C._nn.linear( + x_261, + l_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_261 = l_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_263 = torch._C._nn.gelu(x_262, approximate="none") + x_262 = None + x_264 = torch.nn.functional.dropout(x_263, 0.0, False, False) + x_263 = None + x_265 = torch._C._nn.linear( + x_264, + l_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_264 = l_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_266 = torch.nn.functional.dropout(x_265, 0.0, False, False) + x_265 = None + x_267 = x_260 + x_266 + x_260 = x_266 = None + cls_token_16 = x_267[(slice(None, None, None), slice(None, 1, None))] + img_tokens_16 = x_267[(slice(None, None, None), slice(1, None, None))] + x_267 = None + transpose_99 = img_tokens_16.transpose(1, 2) + img_tokens_16 = None + feat_16 = transpose_99.view(1, 320, 14, 14) + transpose_99 = None + conv2d_67 = torch.conv2d( + feat_16, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 320, + ) + x_268 = conv2d_67 + feat_16 + conv2d_67 = feat_16 = None + flatten_19 = x_268.flatten(2) + x_268 = None + x_269 = flatten_19.transpose(1, 2) + flatten_19 = None + x_270 = torch.cat((cls_token_16, x_269), dim=1) + cls_token_16 = x_269 = None + x_271 = torch.nn.functional.layer_norm( + x_270, + (320,), + l_self_modules_serial_blocks3_modules_7_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks3_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_64 = torch._C._nn.linear( + x_271, + l_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_271 = l_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_66 = linear_64.reshape(1, 197, 3, 8, 40) + linear_64 = None + qkv_16 = reshape_66.permute(2, 0, 3, 1, 4) + reshape_66 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_16 = unbind_16[0] + k_16 = unbind_16[1] + v_16 = unbind_16[2] + unbind_16 = None + k_softmax_16 = k_16.softmax(dim=2) + k_16 = None + transpose_101 = k_softmax_16.transpose(-1, -2) + k_softmax_16 = None + factor_att_32 = transpose_101 @ v_16 + transpose_101 = None + factor_att_33 = q_16 @ factor_att_32 + factor_att_32 = None + q_img_16 = q_16[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_16 = None + v_img_32 = v_16[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_16 = None + transpose_102 = v_img_32.transpose(-1, -2) + v_img_32 = None + v_img_33 = transpose_102.reshape(1, 320, 14, 14) + transpose_102 = None + split_16 = torch.functional.split(v_img_33, [80, 120, 120], dim=1) + v_img_33 = None + getitem_177 = split_16[0] + getitem_178 = split_16[1] + getitem_179 = split_16[2] + split_16 = None + conv2d_68 = torch.conv2d( + getitem_177, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 80, + ) + getitem_177 = None + conv2d_69 = torch.conv2d( + getitem_178, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 120, + ) + getitem_178 = None + conv2d_70 = torch.conv2d( + getitem_179, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 120, + ) + getitem_179 = None + conv_v_img_32 = torch.cat([conv2d_68, conv2d_69, conv2d_70], dim=1) + conv2d_68 = conv2d_69 = conv2d_70 = None + reshape_68 = conv_v_img_32.reshape(1, 8, 40, 196) + conv_v_img_32 = None + conv_v_img_33 = reshape_68.transpose(-1, -2) + reshape_68 = None + EV_hat_32 = q_img_16 * conv_v_img_33 + q_img_16 = conv_v_img_33 = None + EV_hat_33 = torch._C._nn.pad(EV_hat_32, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_32 = None + mul_33 = 0.15811388300841897 * factor_att_33 + factor_att_33 = None + x_272 = mul_33 + EV_hat_33 + mul_33 = EV_hat_33 = None + transpose_104 = x_272.transpose(1, 2) + x_272 = None + x_273 = transpose_104.reshape(1, 197, 320) + transpose_104 = None + x_274 = torch._C._nn.linear( + x_273, + l_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_273 = l_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_275 = torch.nn.functional.dropout(x_274, 0.0, False, False) + x_274 = None + x_276 = x_270 + x_275 + x_270 = x_275 = None + x_277 = torch.nn.functional.layer_norm( + x_276, + (320,), + l_self_modules_serial_blocks3_modules_7_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks3_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_7_modules_norm2_parameters_bias_ + ) = None + x_278 = torch._C._nn.linear( + x_277, + l_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_277 = l_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_279 = torch._C._nn.gelu(x_278, approximate="none") + x_278 = None + x_280 = torch.nn.functional.dropout(x_279, 0.0, False, False) + x_279 = None + x_281 = torch._C._nn.linear( + x_280, + l_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_280 = l_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) + x_281 = None + x_283 = x_276 + x_282 + x_276 = x_282 = None + cls_token_17 = x_283[(slice(None, None, None), slice(None, 1, None))] + img_tokens_17 = x_283[(slice(None, None, None), slice(1, None, None))] + x_283 = None + transpose_105 = img_tokens_17.transpose(1, 2) + img_tokens_17 = None + feat_17 = transpose_105.view(1, 320, 14, 14) + transpose_105 = None + conv2d_71 = torch.conv2d( + feat_17, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 320, + ) + x_284 = conv2d_71 + feat_17 + conv2d_71 = feat_17 = None + flatten_20 = x_284.flatten(2) + x_284 = None + x_285 = flatten_20.transpose(1, 2) + flatten_20 = None + x_286 = torch.cat((cls_token_17, x_285), dim=1) + cls_token_17 = x_285 = None + x_287 = torch.nn.functional.layer_norm( + x_286, + (320,), + l_self_modules_serial_blocks3_modules_8_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks3_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_68 = torch._C._nn.linear( + x_287, + l_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_287 = l_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_70 = linear_68.reshape(1, 197, 3, 8, 40) + linear_68 = None + qkv_17 = reshape_70.permute(2, 0, 3, 1, 4) + reshape_70 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_17 = unbind_17[0] + k_17 = unbind_17[1] + v_17 = unbind_17[2] + unbind_17 = None + k_softmax_17 = k_17.softmax(dim=2) + k_17 = None + transpose_107 = k_softmax_17.transpose(-1, -2) + k_softmax_17 = None + factor_att_34 = transpose_107 @ v_17 + transpose_107 = None + factor_att_35 = q_17 @ factor_att_34 + factor_att_34 = None + q_img_17 = q_17[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_17 = None + v_img_34 = v_17[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_17 = None + transpose_108 = v_img_34.transpose(-1, -2) + v_img_34 = None + v_img_35 = transpose_108.reshape(1, 320, 14, 14) + transpose_108 = None + split_17 = torch.functional.split(v_img_35, [80, 120, 120], dim=1) + v_img_35 = None + getitem_187 = split_17[0] + getitem_188 = split_17[1] + getitem_189 = split_17[2] + split_17 = None + conv2d_72 = torch.conv2d( + getitem_187, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 80, + ) + getitem_187 = None + conv2d_73 = torch.conv2d( + getitem_188, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 120, + ) + getitem_188 = None + conv2d_74 = torch.conv2d( + getitem_189, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 120, + ) + getitem_189 = None + conv_v_img_34 = torch.cat([conv2d_72, conv2d_73, conv2d_74], dim=1) + conv2d_72 = conv2d_73 = conv2d_74 = None + reshape_72 = conv_v_img_34.reshape(1, 8, 40, 196) + conv_v_img_34 = None + conv_v_img_35 = reshape_72.transpose(-1, -2) + reshape_72 = None + EV_hat_34 = q_img_17 * conv_v_img_35 + q_img_17 = conv_v_img_35 = None + EV_hat_35 = torch._C._nn.pad(EV_hat_34, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_34 = None + mul_35 = 0.15811388300841897 * factor_att_35 + factor_att_35 = None + x_288 = mul_35 + EV_hat_35 + mul_35 = EV_hat_35 = None + transpose_110 = x_288.transpose(1, 2) + x_288 = None + x_289 = transpose_110.reshape(1, 197, 320) + transpose_110 = None + x_290 = torch._C._nn.linear( + x_289, + l_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_289 = l_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_291 = torch.nn.functional.dropout(x_290, 0.0, False, False) + x_290 = None + x_292 = x_286 + x_291 + x_286 = x_291 = None + x_293 = torch.nn.functional.layer_norm( + x_292, + (320,), + l_self_modules_serial_blocks3_modules_8_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks3_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_8_modules_norm2_parameters_bias_ + ) = None + x_294 = torch._C._nn.linear( + x_293, + l_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_293 = l_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_295 = torch._C._nn.gelu(x_294, approximate="none") + x_294 = None + x_296 = torch.nn.functional.dropout(x_295, 0.0, False, False) + x_295 = None + x_297 = torch._C._nn.linear( + x_296, + l_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_296 = l_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_298 = torch.nn.functional.dropout(x_297, 0.0, False, False) + x_297 = None + x_299 = x_292 + x_298 + x_292 = x_298 = None + cls_token_18 = x_299[(slice(None, None, None), slice(None, 1, None))] + img_tokens_18 = x_299[(slice(None, None, None), slice(1, None, None))] + x_299 = None + transpose_111 = img_tokens_18.transpose(1, 2) + img_tokens_18 = None + feat_18 = transpose_111.view(1, 320, 14, 14) + transpose_111 = None + conv2d_75 = torch.conv2d( + feat_18, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 320, + ) + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_ = (None) + x_300 = conv2d_75 + feat_18 + conv2d_75 = feat_18 = None + flatten_21 = x_300.flatten(2) + x_300 = None + x_301 = flatten_21.transpose(1, 2) + flatten_21 = None + x_302 = torch.cat((cls_token_18, x_301), dim=1) + cls_token_18 = x_301 = None + x_303 = torch.nn.functional.layer_norm( + x_302, + (320,), + l_self_modules_serial_blocks3_modules_9_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks3_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_72 = torch._C._nn.linear( + x_303, + l_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_303 = l_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_74 = linear_72.reshape(1, 197, 3, 8, 40) + linear_72 = None + qkv_18 = reshape_74.permute(2, 0, 3, 1, 4) + reshape_74 = None + unbind_18 = qkv_18.unbind(0) + qkv_18 = None + q_18 = unbind_18[0] + k_18 = unbind_18[1] + v_18 = unbind_18[2] + unbind_18 = None + k_softmax_18 = k_18.softmax(dim=2) + k_18 = None + transpose_113 = k_softmax_18.transpose(-1, -2) + k_softmax_18 = None + factor_att_36 = transpose_113 @ v_18 + transpose_113 = None + factor_att_37 = q_18 @ factor_att_36 + factor_att_36 = None + q_img_18 = q_18[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_18 = None + v_img_36 = v_18[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_18 = None + transpose_114 = v_img_36.transpose(-1, -2) + v_img_36 = None + v_img_37 = transpose_114.reshape(1, 320, 14, 14) + transpose_114 = None + split_18 = torch.functional.split(v_img_37, [80, 120, 120], dim=1) + v_img_37 = None + getitem_197 = split_18[0] + getitem_198 = split_18[1] + getitem_199 = split_18[2] + split_18 = None + conv2d_76 = torch.conv2d( + getitem_197, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 80, + ) + getitem_197 = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = (None) + conv2d_77 = torch.conv2d( + getitem_198, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 120, + ) + getitem_198 = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = (None) + conv2d_78 = torch.conv2d( + getitem_199, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 120, + ) + getitem_199 = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = (None) + conv_v_img_36 = torch.cat([conv2d_76, conv2d_77, conv2d_78], dim=1) + conv2d_76 = conv2d_77 = conv2d_78 = None + reshape_76 = conv_v_img_36.reshape(1, 8, 40, 196) + conv_v_img_36 = None + conv_v_img_37 = reshape_76.transpose(-1, -2) + reshape_76 = None + EV_hat_36 = q_img_18 * conv_v_img_37 + q_img_18 = conv_v_img_37 = None + EV_hat_37 = torch._C._nn.pad(EV_hat_36, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_36 = None + mul_37 = 0.15811388300841897 * factor_att_37 + factor_att_37 = None + x_304 = mul_37 + EV_hat_37 + mul_37 = EV_hat_37 = None + transpose_116 = x_304.transpose(1, 2) + x_304 = None + x_305 = transpose_116.reshape(1, 197, 320) + transpose_116 = None + x_306 = torch._C._nn.linear( + x_305, + l_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_305 = l_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_307 = torch.nn.functional.dropout(x_306, 0.0, False, False) + x_306 = None + x_308 = x_302 + x_307 + x_302 = x_307 = None + x_309 = torch.nn.functional.layer_norm( + x_308, + (320,), + l_self_modules_serial_blocks3_modules_9_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks3_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_9_modules_norm2_parameters_bias_ + ) = None + x_310 = torch._C._nn.linear( + x_309, + l_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_309 = l_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_311 = torch._C._nn.gelu(x_310, approximate="none") + x_310 = None + x_312 = torch.nn.functional.dropout(x_311, 0.0, False, False) + x_311 = None + x_313 = torch._C._nn.linear( + x_312, + l_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_312 = l_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_314 = torch.nn.functional.dropout(x_313, 0.0, False, False) + x_313 = None + x_315 = x_308 + x_314 + x_308 = x_314 = None + getitem_200 = x_315[ + (slice(None, None, None), slice(1, None, None), slice(None, None, None)) + ] + x_315 = None + reshape_78 = getitem_200.reshape(1, 14, 14, -1) + getitem_200 = None + permute_21 = reshape_78.permute(0, 3, 1, 2) + reshape_78 = None + x3_nocls = permute_21.contiguous() + permute_21 = None + x_316 = torch.conv2d( + x3_nocls, + l_self_modules_patch_embed4_modules_proj_parameters_weight_, + l_self_modules_patch_embed4_modules_proj_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x3_nocls = ( + l_self_modules_patch_embed4_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed4_modules_proj_parameters_bias_ = None + flatten_22 = x_316.flatten(2) + x_316 = None + x_317 = flatten_22.transpose(1, 2) + flatten_22 = None + x_318 = torch.nn.functional.layer_norm( + x_317, + (512,), + l_self_modules_patch_embed4_modules_norm_parameters_weight_, + l_self_modules_patch_embed4_modules_norm_parameters_bias_, + 1e-05, + ) + x_317 = ( + l_self_modules_patch_embed4_modules_norm_parameters_weight_ + ) = l_self_modules_patch_embed4_modules_norm_parameters_bias_ = None + cls_tokens_3 = l_self_parameters_cls_token4_.expand(1, -1, -1) + l_self_parameters_cls_token4_ = None + x_319 = torch.cat((cls_tokens_3, x_318), dim=1) + cls_tokens_3 = x_318 = None + cls_token_19 = x_319[(slice(None, None, None), slice(None, 1, None))] + img_tokens_19 = x_319[(slice(None, None, None), slice(1, None, None))] + x_319 = None + transpose_118 = img_tokens_19.transpose(1, 2) + img_tokens_19 = None + feat_19 = transpose_118.view(1, 512, 7, 7) + transpose_118 = None + conv2d_80 = torch.conv2d( + feat_19, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 512, + ) + x_320 = conv2d_80 + feat_19 + conv2d_80 = feat_19 = None + flatten_23 = x_320.flatten(2) + x_320 = None + x_321 = flatten_23.transpose(1, 2) + flatten_23 = None + x_322 = torch.cat((cls_token_19, x_321), dim=1) + cls_token_19 = x_321 = None + x_323 = torch.nn.functional.layer_norm( + x_322, + (512,), + l_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_ + ) = None + linear_76 = torch._C._nn.linear( + x_323, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_323 = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_79 = linear_76.reshape(1, 50, 3, 8, 64) + linear_76 = None + qkv_19 = reshape_79.permute(2, 0, 3, 1, 4) + reshape_79 = None + unbind_19 = qkv_19.unbind(0) + qkv_19 = None + q_19 = unbind_19[0] + k_19 = unbind_19[1] + v_19 = unbind_19[2] + unbind_19 = None + k_softmax_19 = k_19.softmax(dim=2) + k_19 = None + transpose_120 = k_softmax_19.transpose(-1, -2) + k_softmax_19 = None + factor_att_38 = transpose_120 @ v_19 + transpose_120 = None + factor_att_39 = q_19 @ factor_att_38 + factor_att_38 = None + q_img_19 = q_19[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_19 = None + v_img_38 = v_19[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_19 = None + transpose_121 = v_img_38.transpose(-1, -2) + v_img_38 = None + v_img_39 = transpose_121.reshape(1, 512, 7, 7) + transpose_121 = None + split_19 = torch.functional.split(v_img_39, [128, 192, 192], dim=1) + v_img_39 = None + getitem_208 = split_19[0] + getitem_209 = split_19[1] + getitem_210 = split_19[2] + split_19 = None + conv2d_81 = torch.conv2d( + getitem_208, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 128, + ) + getitem_208 = None + conv2d_82 = torch.conv2d( + getitem_209, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 192, + ) + getitem_209 = None + conv2d_83 = torch.conv2d( + getitem_210, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + getitem_210 = None + conv_v_img_38 = torch.cat([conv2d_81, conv2d_82, conv2d_83], dim=1) + conv2d_81 = conv2d_82 = conv2d_83 = None + reshape_81 = conv_v_img_38.reshape(1, 8, 64, 49) + conv_v_img_38 = None + conv_v_img_39 = reshape_81.transpose(-1, -2) + reshape_81 = None + EV_hat_38 = q_img_19 * conv_v_img_39 + q_img_19 = conv_v_img_39 = None + EV_hat_39 = torch._C._nn.pad(EV_hat_38, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_38 = None + mul_39 = 0.125 * factor_att_39 + factor_att_39 = None + x_324 = mul_39 + EV_hat_39 + mul_39 = EV_hat_39 = None + transpose_123 = x_324.transpose(1, 2) + x_324 = None + x_325 = transpose_123.reshape(1, 50, 512) + transpose_123 = None + x_326 = torch._C._nn.linear( + x_325, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_325 = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_327 = torch.nn.functional.dropout(x_326, 0.0, False, False) + x_326 = None + x_328 = x_322 + x_327 + x_322 = x_327 = None + x_329 = torch.nn.functional.layer_norm( + x_328, + (512,), + l_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_ + ) = None + x_330 = torch._C._nn.linear( + x_329, + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_329 = l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_331 = torch._C._nn.gelu(x_330, approximate="none") + x_330 = None + x_332 = torch.nn.functional.dropout(x_331, 0.0, False, False) + x_331 = None + x_333 = torch._C._nn.linear( + x_332, + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_332 = l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_334 = torch.nn.functional.dropout(x_333, 0.0, False, False) + x_333 = None + x_335 = x_328 + x_334 + x_328 = x_334 = None + cls_token_20 = x_335[(slice(None, None, None), slice(None, 1, None))] + img_tokens_20 = x_335[(slice(None, None, None), slice(1, None, None))] + x_335 = None + transpose_124 = img_tokens_20.transpose(1, 2) + img_tokens_20 = None + feat_20 = transpose_124.view(1, 512, 7, 7) + transpose_124 = None + conv2d_84 = torch.conv2d( + feat_20, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 512, + ) + x_336 = conv2d_84 + feat_20 + conv2d_84 = feat_20 = None + flatten_24 = x_336.flatten(2) + x_336 = None + x_337 = flatten_24.transpose(1, 2) + flatten_24 = None + x_338 = torch.cat((cls_token_20, x_337), dim=1) + cls_token_20 = x_337 = None + x_339 = torch.nn.functional.layer_norm( + x_338, + (512,), + l_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_80 = torch._C._nn.linear( + x_339, + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_339 = l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_83 = linear_80.reshape(1, 50, 3, 8, 64) + linear_80 = None + qkv_20 = reshape_83.permute(2, 0, 3, 1, 4) + reshape_83 = None + unbind_20 = qkv_20.unbind(0) + qkv_20 = None + q_20 = unbind_20[0] + k_20 = unbind_20[1] + v_20 = unbind_20[2] + unbind_20 = None + k_softmax_20 = k_20.softmax(dim=2) + k_20 = None + transpose_126 = k_softmax_20.transpose(-1, -2) + k_softmax_20 = None + factor_att_40 = transpose_126 @ v_20 + transpose_126 = None + factor_att_41 = q_20 @ factor_att_40 + factor_att_40 = None + q_img_20 = q_20[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_20 = None + v_img_40 = v_20[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_20 = None + transpose_127 = v_img_40.transpose(-1, -2) + v_img_40 = None + v_img_41 = transpose_127.reshape(1, 512, 7, 7) + transpose_127 = None + split_20 = torch.functional.split(v_img_41, [128, 192, 192], dim=1) + v_img_41 = None + getitem_218 = split_20[0] + getitem_219 = split_20[1] + getitem_220 = split_20[2] + split_20 = None + conv2d_85 = torch.conv2d( + getitem_218, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 128, + ) + getitem_218 = None + conv2d_86 = torch.conv2d( + getitem_219, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 192, + ) + getitem_219 = None + conv2d_87 = torch.conv2d( + getitem_220, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + getitem_220 = None + conv_v_img_40 = torch.cat([conv2d_85, conv2d_86, conv2d_87], dim=1) + conv2d_85 = conv2d_86 = conv2d_87 = None + reshape_85 = conv_v_img_40.reshape(1, 8, 64, 49) + conv_v_img_40 = None + conv_v_img_41 = reshape_85.transpose(-1, -2) + reshape_85 = None + EV_hat_40 = q_img_20 * conv_v_img_41 + q_img_20 = conv_v_img_41 = None + EV_hat_41 = torch._C._nn.pad(EV_hat_40, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_40 = None + mul_41 = 0.125 * factor_att_41 + factor_att_41 = None + x_340 = mul_41 + EV_hat_41 + mul_41 = EV_hat_41 = None + transpose_129 = x_340.transpose(1, 2) + x_340 = None + x_341 = transpose_129.reshape(1, 50, 512) + transpose_129 = None + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_341 = l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_343 = torch.nn.functional.dropout(x_342, 0.0, False, False) + x_342 = None + x_344 = x_338 + x_343 + x_338 = x_343 = None + x_345 = torch.nn.functional.layer_norm( + x_344, + (512,), + l_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_ + ) = None + x_346 = torch._C._nn.linear( + x_345, + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_345 = l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_347 = torch._C._nn.gelu(x_346, approximate="none") + x_346 = None + x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) + x_347 = None + x_349 = torch._C._nn.linear( + x_348, + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_348 = l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_350 = torch.nn.functional.dropout(x_349, 0.0, False, False) + x_349 = None + x_351 = x_344 + x_350 + x_344 = x_350 = None + cls_token_21 = x_351[(slice(None, None, None), slice(None, 1, None))] + img_tokens_21 = x_351[(slice(None, None, None), slice(1, None, None))] + x_351 = None + transpose_130 = img_tokens_21.transpose(1, 2) + img_tokens_21 = None + feat_21 = transpose_130.view(1, 512, 7, 7) + transpose_130 = None + conv2d_88 = torch.conv2d( + feat_21, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 512, + ) + x_352 = conv2d_88 + feat_21 + conv2d_88 = feat_21 = None + flatten_25 = x_352.flatten(2) + x_352 = None + x_353 = flatten_25.transpose(1, 2) + flatten_25 = None + x_354 = torch.cat((cls_token_21, x_353), dim=1) + cls_token_21 = x_353 = None + x_355 = torch.nn.functional.layer_norm( + x_354, + (512,), + l_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_84 = torch._C._nn.linear( + x_355, + l_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_355 = l_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_87 = linear_84.reshape(1, 50, 3, 8, 64) + linear_84 = None + qkv_21 = reshape_87.permute(2, 0, 3, 1, 4) + reshape_87 = None + unbind_21 = qkv_21.unbind(0) + qkv_21 = None + q_21 = unbind_21[0] + k_21 = unbind_21[1] + v_21 = unbind_21[2] + unbind_21 = None + k_softmax_21 = k_21.softmax(dim=2) + k_21 = None + transpose_132 = k_softmax_21.transpose(-1, -2) + k_softmax_21 = None + factor_att_42 = transpose_132 @ v_21 + transpose_132 = None + factor_att_43 = q_21 @ factor_att_42 + factor_att_42 = None + q_img_21 = q_21[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_21 = None + v_img_42 = v_21[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_21 = None + transpose_133 = v_img_42.transpose(-1, -2) + v_img_42 = None + v_img_43 = transpose_133.reshape(1, 512, 7, 7) + transpose_133 = None + split_21 = torch.functional.split(v_img_43, [128, 192, 192], dim=1) + v_img_43 = None + getitem_228 = split_21[0] + getitem_229 = split_21[1] + getitem_230 = split_21[2] + split_21 = None + conv2d_89 = torch.conv2d( + getitem_228, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 128, + ) + getitem_228 = None + conv2d_90 = torch.conv2d( + getitem_229, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 192, + ) + getitem_229 = None + conv2d_91 = torch.conv2d( + getitem_230, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + getitem_230 = None + conv_v_img_42 = torch.cat([conv2d_89, conv2d_90, conv2d_91], dim=1) + conv2d_89 = conv2d_90 = conv2d_91 = None + reshape_89 = conv_v_img_42.reshape(1, 8, 64, 49) + conv_v_img_42 = None + conv_v_img_43 = reshape_89.transpose(-1, -2) + reshape_89 = None + EV_hat_42 = q_img_21 * conv_v_img_43 + q_img_21 = conv_v_img_43 = None + EV_hat_43 = torch._C._nn.pad(EV_hat_42, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_42 = None + mul_43 = 0.125 * factor_att_43 + factor_att_43 = None + x_356 = mul_43 + EV_hat_43 + mul_43 = EV_hat_43 = None + transpose_135 = x_356.transpose(1, 2) + x_356 = None + x_357 = transpose_135.reshape(1, 50, 512) + transpose_135 = None + x_358 = torch._C._nn.linear( + x_357, + l_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_357 = l_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_359 = torch.nn.functional.dropout(x_358, 0.0, False, False) + x_358 = None + x_360 = x_354 + x_359 + x_354 = x_359 = None + x_361 = torch.nn.functional.layer_norm( + x_360, + (512,), + l_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_bias_ + ) = None + x_362 = torch._C._nn.linear( + x_361, + l_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_361 = l_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_363 = torch._C._nn.gelu(x_362, approximate="none") + x_362 = None + x_364 = torch.nn.functional.dropout(x_363, 0.0, False, False) + x_363 = None + x_365 = torch._C._nn.linear( + x_364, + l_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_364 = l_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_366 = torch.nn.functional.dropout(x_365, 0.0, False, False) + x_365 = None + x_367 = x_360 + x_366 + x_360 = x_366 = None + cls_token_22 = x_367[(slice(None, None, None), slice(None, 1, None))] + img_tokens_22 = x_367[(slice(None, None, None), slice(1, None, None))] + x_367 = None + transpose_136 = img_tokens_22.transpose(1, 2) + img_tokens_22 = None + feat_22 = transpose_136.view(1, 512, 7, 7) + transpose_136 = None + conv2d_92 = torch.conv2d( + feat_22, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 512, + ) + x_368 = conv2d_92 + feat_22 + conv2d_92 = feat_22 = None + flatten_26 = x_368.flatten(2) + x_368 = None + x_369 = flatten_26.transpose(1, 2) + flatten_26 = None + x_370 = torch.cat((cls_token_22, x_369), dim=1) + cls_token_22 = x_369 = None + x_371 = torch.nn.functional.layer_norm( + x_370, + (512,), + l_self_modules_serial_blocks4_modules_3_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks4_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_88 = torch._C._nn.linear( + x_371, + l_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_371 = l_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_91 = linear_88.reshape(1, 50, 3, 8, 64) + linear_88 = None + qkv_22 = reshape_91.permute(2, 0, 3, 1, 4) + reshape_91 = None + unbind_22 = qkv_22.unbind(0) + qkv_22 = None + q_22 = unbind_22[0] + k_22 = unbind_22[1] + v_22 = unbind_22[2] + unbind_22 = None + k_softmax_22 = k_22.softmax(dim=2) + k_22 = None + transpose_138 = k_softmax_22.transpose(-1, -2) + k_softmax_22 = None + factor_att_44 = transpose_138 @ v_22 + transpose_138 = None + factor_att_45 = q_22 @ factor_att_44 + factor_att_44 = None + q_img_22 = q_22[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_22 = None + v_img_44 = v_22[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_22 = None + transpose_139 = v_img_44.transpose(-1, -2) + v_img_44 = None + v_img_45 = transpose_139.reshape(1, 512, 7, 7) + transpose_139 = None + split_22 = torch.functional.split(v_img_45, [128, 192, 192], dim=1) + v_img_45 = None + getitem_238 = split_22[0] + getitem_239 = split_22[1] + getitem_240 = split_22[2] + split_22 = None + conv2d_93 = torch.conv2d( + getitem_238, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 128, + ) + getitem_238 = None + conv2d_94 = torch.conv2d( + getitem_239, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 192, + ) + getitem_239 = None + conv2d_95 = torch.conv2d( + getitem_240, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + getitem_240 = None + conv_v_img_44 = torch.cat([conv2d_93, conv2d_94, conv2d_95], dim=1) + conv2d_93 = conv2d_94 = conv2d_95 = None + reshape_93 = conv_v_img_44.reshape(1, 8, 64, 49) + conv_v_img_44 = None + conv_v_img_45 = reshape_93.transpose(-1, -2) + reshape_93 = None + EV_hat_44 = q_img_22 * conv_v_img_45 + q_img_22 = conv_v_img_45 = None + EV_hat_45 = torch._C._nn.pad(EV_hat_44, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_44 = None + mul_45 = 0.125 * factor_att_45 + factor_att_45 = None + x_372 = mul_45 + EV_hat_45 + mul_45 = EV_hat_45 = None + transpose_141 = x_372.transpose(1, 2) + x_372 = None + x_373 = transpose_141.reshape(1, 50, 512) + transpose_141 = None + x_374 = torch._C._nn.linear( + x_373, + l_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_373 = l_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_375 = torch.nn.functional.dropout(x_374, 0.0, False, False) + x_374 = None + x_376 = x_370 + x_375 + x_370 = x_375 = None + x_377 = torch.nn.functional.layer_norm( + x_376, + (512,), + l_self_modules_serial_blocks4_modules_3_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks4_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_3_modules_norm2_parameters_bias_ + ) = None + x_378 = torch._C._nn.linear( + x_377, + l_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_377 = l_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_379 = torch._C._nn.gelu(x_378, approximate="none") + x_378 = None + x_380 = torch.nn.functional.dropout(x_379, 0.0, False, False) + x_379 = None + x_381 = torch._C._nn.linear( + x_380, + l_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_380 = l_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_382 = torch.nn.functional.dropout(x_381, 0.0, False, False) + x_381 = None + x_383 = x_376 + x_382 + x_376 = x_382 = None + cls_token_23 = x_383[(slice(None, None, None), slice(None, 1, None))] + img_tokens_23 = x_383[(slice(None, None, None), slice(1, None, None))] + x_383 = None + transpose_142 = img_tokens_23.transpose(1, 2) + img_tokens_23 = None + feat_23 = transpose_142.view(1, 512, 7, 7) + transpose_142 = None + conv2d_96 = torch.conv2d( + feat_23, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 512, + ) + x_384 = conv2d_96 + feat_23 + conv2d_96 = feat_23 = None + flatten_27 = x_384.flatten(2) + x_384 = None + x_385 = flatten_27.transpose(1, 2) + flatten_27 = None + x_386 = torch.cat((cls_token_23, x_385), dim=1) + cls_token_23 = x_385 = None + x_387 = torch.nn.functional.layer_norm( + x_386, + (512,), + l_self_modules_serial_blocks4_modules_4_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks4_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_92 = torch._C._nn.linear( + x_387, + l_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_387 = l_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_95 = linear_92.reshape(1, 50, 3, 8, 64) + linear_92 = None + qkv_23 = reshape_95.permute(2, 0, 3, 1, 4) + reshape_95 = None + unbind_23 = qkv_23.unbind(0) + qkv_23 = None + q_23 = unbind_23[0] + k_23 = unbind_23[1] + v_23 = unbind_23[2] + unbind_23 = None + k_softmax_23 = k_23.softmax(dim=2) + k_23 = None + transpose_144 = k_softmax_23.transpose(-1, -2) + k_softmax_23 = None + factor_att_46 = transpose_144 @ v_23 + transpose_144 = None + factor_att_47 = q_23 @ factor_att_46 + factor_att_46 = None + q_img_23 = q_23[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_23 = None + v_img_46 = v_23[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_23 = None + transpose_145 = v_img_46.transpose(-1, -2) + v_img_46 = None + v_img_47 = transpose_145.reshape(1, 512, 7, 7) + transpose_145 = None + split_23 = torch.functional.split(v_img_47, [128, 192, 192], dim=1) + v_img_47 = None + getitem_248 = split_23[0] + getitem_249 = split_23[1] + getitem_250 = split_23[2] + split_23 = None + conv2d_97 = torch.conv2d( + getitem_248, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 128, + ) + getitem_248 = None + conv2d_98 = torch.conv2d( + getitem_249, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 192, + ) + getitem_249 = None + conv2d_99 = torch.conv2d( + getitem_250, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + getitem_250 = None + conv_v_img_46 = torch.cat([conv2d_97, conv2d_98, conv2d_99], dim=1) + conv2d_97 = conv2d_98 = conv2d_99 = None + reshape_97 = conv_v_img_46.reshape(1, 8, 64, 49) + conv_v_img_46 = None + conv_v_img_47 = reshape_97.transpose(-1, -2) + reshape_97 = None + EV_hat_46 = q_img_23 * conv_v_img_47 + q_img_23 = conv_v_img_47 = None + EV_hat_47 = torch._C._nn.pad(EV_hat_46, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_46 = None + mul_47 = 0.125 * factor_att_47 + factor_att_47 = None + x_388 = mul_47 + EV_hat_47 + mul_47 = EV_hat_47 = None + transpose_147 = x_388.transpose(1, 2) + x_388 = None + x_389 = transpose_147.reshape(1, 50, 512) + transpose_147 = None + x_390 = torch._C._nn.linear( + x_389, + l_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_389 = l_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_391 = torch.nn.functional.dropout(x_390, 0.0, False, False) + x_390 = None + x_392 = x_386 + x_391 + x_386 = x_391 = None + x_393 = torch.nn.functional.layer_norm( + x_392, + (512,), + l_self_modules_serial_blocks4_modules_4_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks4_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_4_modules_norm2_parameters_bias_ + ) = None + x_394 = torch._C._nn.linear( + x_393, + l_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_393 = l_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_395 = torch._C._nn.gelu(x_394, approximate="none") + x_394 = None + x_396 = torch.nn.functional.dropout(x_395, 0.0, False, False) + x_395 = None + x_397 = torch._C._nn.linear( + x_396, + l_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_396 = l_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_398 = torch.nn.functional.dropout(x_397, 0.0, False, False) + x_397 = None + x_399 = x_392 + x_398 + x_392 = x_398 = None + cls_token_24 = x_399[(slice(None, None, None), slice(None, 1, None))] + img_tokens_24 = x_399[(slice(None, None, None), slice(1, None, None))] + x_399 = None + transpose_148 = img_tokens_24.transpose(1, 2) + img_tokens_24 = None + feat_24 = transpose_148.view(1, 512, 7, 7) + transpose_148 = None + conv2d_100 = torch.conv2d( + feat_24, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 512, + ) + x_400 = conv2d_100 + feat_24 + conv2d_100 = feat_24 = None + flatten_28 = x_400.flatten(2) + x_400 = None + x_401 = flatten_28.transpose(1, 2) + flatten_28 = None + x_402 = torch.cat((cls_token_24, x_401), dim=1) + cls_token_24 = x_401 = None + x_403 = torch.nn.functional.layer_norm( + x_402, + (512,), + l_self_modules_serial_blocks4_modules_5_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks4_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_96 = torch._C._nn.linear( + x_403, + l_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_403 = l_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_99 = linear_96.reshape(1, 50, 3, 8, 64) + linear_96 = None + qkv_24 = reshape_99.permute(2, 0, 3, 1, 4) + reshape_99 = None + unbind_24 = qkv_24.unbind(0) + qkv_24 = None + q_24 = unbind_24[0] + k_24 = unbind_24[1] + v_24 = unbind_24[2] + unbind_24 = None + k_softmax_24 = k_24.softmax(dim=2) + k_24 = None + transpose_150 = k_softmax_24.transpose(-1, -2) + k_softmax_24 = None + factor_att_48 = transpose_150 @ v_24 + transpose_150 = None + factor_att_49 = q_24 @ factor_att_48 + factor_att_48 = None + q_img_24 = q_24[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_24 = None + v_img_48 = v_24[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_24 = None + transpose_151 = v_img_48.transpose(-1, -2) + v_img_48 = None + v_img_49 = transpose_151.reshape(1, 512, 7, 7) + transpose_151 = None + split_24 = torch.functional.split(v_img_49, [128, 192, 192], dim=1) + v_img_49 = None + getitem_258 = split_24[0] + getitem_259 = split_24[1] + getitem_260 = split_24[2] + split_24 = None + conv2d_101 = torch.conv2d( + getitem_258, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 128, + ) + getitem_258 = None + conv2d_102 = torch.conv2d( + getitem_259, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 192, + ) + getitem_259 = None + conv2d_103 = torch.conv2d( + getitem_260, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + getitem_260 = None + conv_v_img_48 = torch.cat([conv2d_101, conv2d_102, conv2d_103], dim=1) + conv2d_101 = conv2d_102 = conv2d_103 = None + reshape_101 = conv_v_img_48.reshape(1, 8, 64, 49) + conv_v_img_48 = None + conv_v_img_49 = reshape_101.transpose(-1, -2) + reshape_101 = None + EV_hat_48 = q_img_24 * conv_v_img_49 + q_img_24 = conv_v_img_49 = None + EV_hat_49 = torch._C._nn.pad(EV_hat_48, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_48 = None + mul_49 = 0.125 * factor_att_49 + factor_att_49 = None + x_404 = mul_49 + EV_hat_49 + mul_49 = EV_hat_49 = None + transpose_153 = x_404.transpose(1, 2) + x_404 = None + x_405 = transpose_153.reshape(1, 50, 512) + transpose_153 = None + x_406 = torch._C._nn.linear( + x_405, + l_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_405 = l_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_407 = torch.nn.functional.dropout(x_406, 0.0, False, False) + x_406 = None + x_408 = x_402 + x_407 + x_402 = x_407 = None + x_409 = torch.nn.functional.layer_norm( + x_408, + (512,), + l_self_modules_serial_blocks4_modules_5_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks4_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_5_modules_norm2_parameters_bias_ + ) = None + x_410 = torch._C._nn.linear( + x_409, + l_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_409 = l_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_411 = torch._C._nn.gelu(x_410, approximate="none") + x_410 = None + x_412 = torch.nn.functional.dropout(x_411, 0.0, False, False) + x_411 = None + x_413 = torch._C._nn.linear( + x_412, + l_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_412 = l_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_414 = torch.nn.functional.dropout(x_413, 0.0, False, False) + x_413 = None + x_415 = x_408 + x_414 + x_408 = x_414 = None + cls_token_25 = x_415[(slice(None, None, None), slice(None, 1, None))] + img_tokens_25 = x_415[(slice(None, None, None), slice(1, None, None))] + x_415 = None + transpose_154 = img_tokens_25.transpose(1, 2) + img_tokens_25 = None + feat_25 = transpose_154.view(1, 512, 7, 7) + transpose_154 = None + conv2d_104 = torch.conv2d( + feat_25, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 512, + ) + x_416 = conv2d_104 + feat_25 + conv2d_104 = feat_25 = None + flatten_29 = x_416.flatten(2) + x_416 = None + x_417 = flatten_29.transpose(1, 2) + flatten_29 = None + x_418 = torch.cat((cls_token_25, x_417), dim=1) + cls_token_25 = x_417 = None + x_419 = torch.nn.functional.layer_norm( + x_418, + (512,), + l_self_modules_serial_blocks4_modules_6_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks4_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_100 = torch._C._nn.linear( + x_419, + l_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_419 = l_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_103 = linear_100.reshape(1, 50, 3, 8, 64) + linear_100 = None + qkv_25 = reshape_103.permute(2, 0, 3, 1, 4) + reshape_103 = None + unbind_25 = qkv_25.unbind(0) + qkv_25 = None + q_25 = unbind_25[0] + k_25 = unbind_25[1] + v_25 = unbind_25[2] + unbind_25 = None + k_softmax_25 = k_25.softmax(dim=2) + k_25 = None + transpose_156 = k_softmax_25.transpose(-1, -2) + k_softmax_25 = None + factor_att_50 = transpose_156 @ v_25 + transpose_156 = None + factor_att_51 = q_25 @ factor_att_50 + factor_att_50 = None + q_img_25 = q_25[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_25 = None + v_img_50 = v_25[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_25 = None + transpose_157 = v_img_50.transpose(-1, -2) + v_img_50 = None + v_img_51 = transpose_157.reshape(1, 512, 7, 7) + transpose_157 = None + split_25 = torch.functional.split(v_img_51, [128, 192, 192], dim=1) + v_img_51 = None + getitem_268 = split_25[0] + getitem_269 = split_25[1] + getitem_270 = split_25[2] + split_25 = None + conv2d_105 = torch.conv2d( + getitem_268, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 128, + ) + getitem_268 = None + conv2d_106 = torch.conv2d( + getitem_269, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 192, + ) + getitem_269 = None + conv2d_107 = torch.conv2d( + getitem_270, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + getitem_270 = None + conv_v_img_50 = torch.cat([conv2d_105, conv2d_106, conv2d_107], dim=1) + conv2d_105 = conv2d_106 = conv2d_107 = None + reshape_105 = conv_v_img_50.reshape(1, 8, 64, 49) + conv_v_img_50 = None + conv_v_img_51 = reshape_105.transpose(-1, -2) + reshape_105 = None + EV_hat_50 = q_img_25 * conv_v_img_51 + q_img_25 = conv_v_img_51 = None + EV_hat_51 = torch._C._nn.pad(EV_hat_50, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_50 = None + mul_51 = 0.125 * factor_att_51 + factor_att_51 = None + x_420 = mul_51 + EV_hat_51 + mul_51 = EV_hat_51 = None + transpose_159 = x_420.transpose(1, 2) + x_420 = None + x_421 = transpose_159.reshape(1, 50, 512) + transpose_159 = None + x_422 = torch._C._nn.linear( + x_421, + l_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_421 = l_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_423 = torch.nn.functional.dropout(x_422, 0.0, False, False) + x_422 = None + x_424 = x_418 + x_423 + x_418 = x_423 = None + x_425 = torch.nn.functional.layer_norm( + x_424, + (512,), + l_self_modules_serial_blocks4_modules_6_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks4_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_6_modules_norm2_parameters_bias_ + ) = None + x_426 = torch._C._nn.linear( + x_425, + l_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_425 = l_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_427 = torch._C._nn.gelu(x_426, approximate="none") + x_426 = None + x_428 = torch.nn.functional.dropout(x_427, 0.0, False, False) + x_427 = None + x_429 = torch._C._nn.linear( + x_428, + l_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_428 = l_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_430 = torch.nn.functional.dropout(x_429, 0.0, False, False) + x_429 = None + x_431 = x_424 + x_430 + x_424 = x_430 = None + cls_token_26 = x_431[(slice(None, None, None), slice(None, 1, None))] + img_tokens_26 = x_431[(slice(None, None, None), slice(1, None, None))] + x_431 = None + transpose_160 = img_tokens_26.transpose(1, 2) + img_tokens_26 = None + feat_26 = transpose_160.view(1, 512, 7, 7) + transpose_160 = None + conv2d_108 = torch.conv2d( + feat_26, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 512, + ) + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_ = (None) + x_432 = conv2d_108 + feat_26 + conv2d_108 = feat_26 = None + flatten_30 = x_432.flatten(2) + x_432 = None + x_433 = flatten_30.transpose(1, 2) + flatten_30 = None + x_434 = torch.cat((cls_token_26, x_433), dim=1) + cls_token_26 = x_433 = None + x_435 = torch.nn.functional.layer_norm( + x_434, + (512,), + l_self_modules_serial_blocks4_modules_7_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks4_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_104 = torch._C._nn.linear( + x_435, + l_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_435 = l_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_107 = linear_104.reshape(1, 50, 3, 8, 64) + linear_104 = None + qkv_26 = reshape_107.permute(2, 0, 3, 1, 4) + reshape_107 = None + unbind_26 = qkv_26.unbind(0) + qkv_26 = None + q_26 = unbind_26[0] + k_26 = unbind_26[1] + v_26 = unbind_26[2] + unbind_26 = None + k_softmax_26 = k_26.softmax(dim=2) + k_26 = None + transpose_162 = k_softmax_26.transpose(-1, -2) + k_softmax_26 = None + factor_att_52 = transpose_162 @ v_26 + transpose_162 = None + factor_att_53 = q_26 @ factor_att_52 + factor_att_52 = None + q_img_26 = q_26[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_26 = None + v_img_52 = v_26[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_26 = None + transpose_163 = v_img_52.transpose(-1, -2) + v_img_52 = None + v_img_53 = transpose_163.reshape(1, 512, 7, 7) + transpose_163 = None + split_26 = torch.functional.split(v_img_53, [128, 192, 192], dim=1) + v_img_53 = None + getitem_278 = split_26[0] + getitem_279 = split_26[1] + getitem_280 = split_26[2] + split_26 = None + conv2d_109 = torch.conv2d( + getitem_278, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 128, + ) + getitem_278 = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = (None) + conv2d_110 = torch.conv2d( + getitem_279, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 192, + ) + getitem_279 = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = (None) + conv2d_111 = torch.conv2d( + getitem_280, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + getitem_280 = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = (None) + conv_v_img_52 = torch.cat([conv2d_109, conv2d_110, conv2d_111], dim=1) + conv2d_109 = conv2d_110 = conv2d_111 = None + reshape_109 = conv_v_img_52.reshape(1, 8, 64, 49) + conv_v_img_52 = None + conv_v_img_53 = reshape_109.transpose(-1, -2) + reshape_109 = None + EV_hat_52 = q_img_26 * conv_v_img_53 + q_img_26 = conv_v_img_53 = None + EV_hat_53 = torch._C._nn.pad(EV_hat_52, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_52 = None + mul_53 = 0.125 * factor_att_53 + factor_att_53 = None + x_436 = mul_53 + EV_hat_53 + mul_53 = EV_hat_53 = None + transpose_165 = x_436.transpose(1, 2) + x_436 = None + x_437 = transpose_165.reshape(1, 50, 512) + transpose_165 = None + x_438 = torch._C._nn.linear( + x_437, + l_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_437 = l_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_439 = torch.nn.functional.dropout(x_438, 0.0, False, False) + x_438 = None + x_440 = x_434 + x_439 + x_434 = x_439 = None + x_441 = torch.nn.functional.layer_norm( + x_440, + (512,), + l_self_modules_serial_blocks4_modules_7_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks4_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_7_modules_norm2_parameters_bias_ + ) = None + x_442 = torch._C._nn.linear( + x_441, + l_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_441 = l_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_443 = torch._C._nn.gelu(x_442, approximate="none") + x_442 = None + x_444 = torch.nn.functional.dropout(x_443, 0.0, False, False) + x_443 = None + x_445 = torch._C._nn.linear( + x_444, + l_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_444 = l_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_446 = torch.nn.functional.dropout(x_445, 0.0, False, False) + x_445 = None + x_447 = x_440 + x_446 + x_440 = x_446 = None + getitem_281 = x_447[ + (slice(None, None, None), slice(1, None, None), slice(None, None, None)) + ] + reshape_111 = getitem_281.reshape(1, 7, 7, -1) + getitem_281 = None + permute_30 = reshape_111.permute(0, 3, 1, 2) + reshape_111 = None + x4_nocls = permute_30.contiguous() + permute_30 = x4_nocls = None + x_448 = torch.nn.functional.layer_norm( + x_447, + (512,), + l_self_modules_norm4_parameters_weight_, + l_self_modules_norm4_parameters_bias_, + 1e-06, + ) + x_447 = ( + l_self_modules_norm4_parameters_weight_ + ) = l_self_modules_norm4_parameters_bias_ = None + x_449 = x_448[(slice(None, None, None), 0)] + x_448 = None + x_450 = torch.nn.functional.dropout(x_449, 0.0, False, False) + x_449 = None + x_451 = torch._C._nn.linear( + x_450, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_450 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_451,) diff --git a/samples/timm/coat_lite_medium.in1k/weight_meta.py b/samples/timm/coat_lite_medium.in1k/weight_meta.py new file mode 100644 index 0000000000..76285e6ec6 --- /dev/null +++ b/samples/timm/coat_lite_medium.in1k/weight_meta.py @@ -0,0 +1,3808 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.236 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed1_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed1_modules_proj_parameters_weight_" + shape = [128, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.084 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed1_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed1_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed1_modules_norm_parameters_weight_: + name = "L_self_modules_patch_embed1_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed1_modules_norm_parameters_bias_: + name = "L_self_modules_patch_embed1_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token1_: + name = "L_self_parameters_cls_token1_" + shape = [1, 1, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_" + shape = [128, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.191 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.011 + std = 0.191 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_" + shape = [32, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.010 + std = 0.188 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.029 + std = 0.207 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_" + shape = [48, 1, 5, 5] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.116 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.014 + std = 0.107 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_" + shape = [48, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.007 + std = 0.081 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed2_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed2_modules_proj_parameters_weight_" + shape = [256, 128, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.025 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed2_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed2_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed2_modules_norm_parameters_weight_: + name = "L_self_modules_patch_embed2_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed2_modules_norm_parameters_bias_: + name = "L_self_modules_patch_embed2_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token2_: + name = "L_self_parameters_cls_token2_" + shape = [1, 1, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_" + shape = [256, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.196 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.013 + std = 0.193 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_" + shape = [64, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.008 + std = 0.192 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.005 + std = 0.194 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_" + shape = [96, 1, 5, 5] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.116 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.008 + std = 0.122 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.007 + std = 0.085 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_4_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_4_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_4_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_4_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_5_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_5_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_5_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_5_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed3_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed3_modules_proj_parameters_weight_" + shape = [320, 256, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed3_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed3_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed3_modules_norm_parameters_weight_: + name = "L_self_modules_patch_embed3_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed3_modules_norm_parameters_bias_: + name = "L_self_modules_patch_embed3_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token3_: + name = "L_self_parameters_cls_token3_" + shape = [1, 1, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_" + shape = [320, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.008 + std = 0.193 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.197 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_" + shape = [80, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.193 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.022 + std = 0.191 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_" + shape = [120, 1, 5, 5] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.116 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.010 + std = 0.121 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_" + shape = [120, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.010 + std = 0.081 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_6_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_6_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_6_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_6_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_6_modules_norm2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_7_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_7_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_7_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_7_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_7_modules_norm2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_8_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_8_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_8_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_8_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_8_modules_norm2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_9_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_9_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_9_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_9_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_9_modules_norm2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed4_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed4_modules_proj_parameters_weight_" + shape = [512, 320, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.016 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed4_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed4_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.016 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed4_modules_norm_parameters_weight_: + name = "L_self_modules_patch_embed4_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed4_modules_norm_parameters_bias_: + name = "L_self_modules_patch_embed4_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token4_: + name = "L_self_parameters_cls_token4_" + shape = [1, 1, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_" + shape = [512, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.193 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.005 + std = 0.192 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_" + shape = [128, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.007 + std = 0.198 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.013 + std = 0.198 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_" + shape = [192, 1, 5, 5] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.116 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.012 + std = 0.120 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_3_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_3_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_3_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_3_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_4_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_4_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_4_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_4_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_5_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_5_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_5_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_5_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_6_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_6_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_6_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_6_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_6_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_7_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_7_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_7_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_7_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_7_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm4_parameters_weight_: + name = "L_self_modules_norm4_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm4_parameters_bias_: + name = "L_self_modules_norm4_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/coat_lite_mini.in1k/graph_hash.txt b/samples/timm/coat_lite_mini.in1k/graph_hash.txt new file mode 100644 index 0000000000..de0390b938 --- /dev/null +++ b/samples/timm/coat_lite_mini.in1k/graph_hash.txt @@ -0,0 +1 @@ +731518653d5bd2112ff4d03f41383bfd9ada5b116521be8ee6ff2a86b0651993 \ No newline at end of file diff --git a/samples/timm/coat_lite_mini.in1k/graph_net.json b/samples/timm/coat_lite_mini.in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/coat_lite_mini.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/coat_lite_mini.in1k/input_meta.py b/samples/timm/coat_lite_mini.in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/coat_lite_mini.in1k/input_tensor_constraints.py b/samples/timm/coat_lite_mini.in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/coat_lite_mini.in1k/model.py b/samples/timm/coat_lite_mini.in1k/model.py new file mode 100644 index 0000000000..08cda1d446 --- /dev/null +++ b/samples/timm/coat_lite_mini.in1k/model.py @@ -0,0 +1,1989 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed1_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed1_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token1_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed2_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed2_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token2_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed3_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed3_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token3_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed4_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed4_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token4_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm4_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm4_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed1_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed1_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed1_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed1_modules_proj_parameters_bias_ + ) + l_self_modules_patch_embed1_modules_norm_parameters_weight_ = ( + L_self_modules_patch_embed1_modules_norm_parameters_weight_ + ) + l_self_modules_patch_embed1_modules_norm_parameters_bias_ = ( + L_self_modules_patch_embed1_modules_norm_parameters_bias_ + ) + l_self_parameters_cls_token1_ = L_self_parameters_cls_token1_ + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_patch_embed2_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed2_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed2_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed2_modules_proj_parameters_bias_ + ) + l_self_modules_patch_embed2_modules_norm_parameters_weight_ = ( + L_self_modules_patch_embed2_modules_norm_parameters_weight_ + ) + l_self_modules_patch_embed2_modules_norm_parameters_bias_ = ( + L_self_modules_patch_embed2_modules_norm_parameters_bias_ + ) + l_self_parameters_cls_token2_ = L_self_parameters_cls_token2_ + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_patch_embed3_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed3_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed3_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed3_modules_proj_parameters_bias_ + ) + l_self_modules_patch_embed3_modules_norm_parameters_weight_ = ( + L_self_modules_patch_embed3_modules_norm_parameters_weight_ + ) + l_self_modules_patch_embed3_modules_norm_parameters_bias_ = ( + L_self_modules_patch_embed3_modules_norm_parameters_bias_ + ) + l_self_parameters_cls_token3_ = L_self_parameters_cls_token3_ + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_patch_embed4_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed4_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed4_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed4_modules_proj_parameters_bias_ + ) + l_self_modules_patch_embed4_modules_norm_parameters_weight_ = ( + L_self_modules_patch_embed4_modules_norm_parameters_weight_ + ) + l_self_modules_patch_embed4_modules_norm_parameters_bias_ = ( + L_self_modules_patch_embed4_modules_norm_parameters_bias_ + ) + l_self_parameters_cls_token4_ = L_self_parameters_cls_token4_ + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_norm4_parameters_weight_ = ( + L_self_modules_norm4_parameters_weight_ + ) + l_self_modules_norm4_parameters_bias_ = L_self_modules_norm4_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed1_modules_proj_parameters_weight_, + l_self_modules_patch_embed1_modules_proj_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed1_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed1_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + x_2 = torch.nn.functional.layer_norm( + x_1, + (64,), + l_self_modules_patch_embed1_modules_norm_parameters_weight_, + l_self_modules_patch_embed1_modules_norm_parameters_bias_, + 1e-05, + ) + x_1 = ( + l_self_modules_patch_embed1_modules_norm_parameters_weight_ + ) = l_self_modules_patch_embed1_modules_norm_parameters_bias_ = None + cls_tokens = l_self_parameters_cls_token1_.expand(1, -1, -1) + l_self_parameters_cls_token1_ = None + x_3 = torch.cat((cls_tokens, x_2), dim=1) + cls_tokens = x_2 = None + cls_token = x_3[(slice(None, None, None), slice(None, 1, None))] + img_tokens = x_3[(slice(None, None, None), slice(1, None, None))] + x_3 = None + transpose_1 = img_tokens.transpose(1, 2) + img_tokens = None + feat = transpose_1.view(1, 64, 56, 56) + transpose_1 = None + conv2d_1 = torch.conv2d( + feat, + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 64, + ) + x_4 = conv2d_1 + feat + conv2d_1 = feat = None + flatten_1 = x_4.flatten(2) + x_4 = None + x_5 = flatten_1.transpose(1, 2) + flatten_1 = None + x_6 = torch.cat((cls_token, x_5), dim=1) + cls_token = x_5 = None + x_7 = torch.nn.functional.layer_norm( + x_6, + (64,), + l_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_7, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_7 = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 3137, 3, 8, 8) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + k_softmax = k.softmax(dim=2) + k = None + transpose_3 = k_softmax.transpose(-1, -2) + k_softmax = None + factor_att = transpose_3 @ v + transpose_3 = None + factor_att_1 = q @ factor_att + factor_att = None + q_img = q[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q = None + v_img = v[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v = None + transpose_4 = v_img.transpose(-1, -2) + v_img = None + v_img_1 = transpose_4.reshape(1, 64, 56, 56) + transpose_4 = None + split = torch.functional.split(v_img_1, [16, 24, 24], dim=1) + v_img_1 = None + getitem_15 = split[0] + getitem_16 = split[1] + getitem_17 = split[2] + split = None + conv2d_2 = torch.conv2d( + getitem_15, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 16, + ) + getitem_15 = None + conv2d_3 = torch.conv2d( + getitem_16, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 24, + ) + getitem_16 = None + conv2d_4 = torch.conv2d( + getitem_17, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 24, + ) + getitem_17 = None + conv_v_img = torch.cat([conv2d_2, conv2d_3, conv2d_4], dim=1) + conv2d_2 = conv2d_3 = conv2d_4 = None + reshape_2 = conv_v_img.reshape(1, 8, 8, 3136) + conv_v_img = None + conv_v_img_1 = reshape_2.transpose(-1, -2) + reshape_2 = None + EV_hat = q_img * conv_v_img_1 + q_img = conv_v_img_1 = None + EV_hat_1 = torch._C._nn.pad(EV_hat, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat = None + mul_1 = 0.3535533905932738 * factor_att_1 + factor_att_1 = None + x_8 = mul_1 + EV_hat_1 + mul_1 = EV_hat_1 = None + transpose_6 = x_8.transpose(1, 2) + x_8 = None + x_9 = transpose_6.reshape(1, 3137, 64) + transpose_6 = None + x_10 = torch._C._nn.linear( + x_9, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_9 = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = x_6 + x_11 + x_6 = x_11 = None + x_13 = torch.nn.functional.layer_norm( + x_12, + (64,), + l_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_ + ) = None + x_14 = torch._C._nn.linear( + x_13, + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_13 = l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_15 = torch._C._nn.gelu(x_14, approximate="none") + x_14 = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_16 = l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_18 = torch.nn.functional.dropout(x_17, 0.0, False, False) + x_17 = None + x_19 = x_12 + x_18 + x_12 = x_18 = None + cls_token_1 = x_19[(slice(None, None, None), slice(None, 1, None))] + img_tokens_1 = x_19[(slice(None, None, None), slice(1, None, None))] + x_19 = None + transpose_7 = img_tokens_1.transpose(1, 2) + img_tokens_1 = None + feat_1 = transpose_7.view(1, 64, 56, 56) + transpose_7 = None + conv2d_5 = torch.conv2d( + feat_1, + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 64, + ) + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_ = (None) + x_20 = conv2d_5 + feat_1 + conv2d_5 = feat_1 = None + flatten_2 = x_20.flatten(2) + x_20 = None + x_21 = flatten_2.transpose(1, 2) + flatten_2 = None + x_22 = torch.cat((cls_token_1, x_21), dim=1) + cls_token_1 = x_21 = None + x_23 = torch.nn.functional.layer_norm( + x_22, + (64,), + l_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_23, + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_23 = l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_4.reshape(1, 3137, 3, 8, 8) + linear_4 = None + qkv_1 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + k_softmax_1 = k_1.softmax(dim=2) + k_1 = None + transpose_9 = k_softmax_1.transpose(-1, -2) + k_softmax_1 = None + factor_att_2 = transpose_9 @ v_1 + transpose_9 = None + factor_att_3 = q_1 @ factor_att_2 + factor_att_2 = None + q_img_1 = q_1[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_1 = None + v_img_2 = v_1[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_1 = None + transpose_10 = v_img_2.transpose(-1, -2) + v_img_2 = None + v_img_3 = transpose_10.reshape(1, 64, 56, 56) + transpose_10 = None + split_1 = torch.functional.split(v_img_3, [16, 24, 24], dim=1) + v_img_3 = None + getitem_25 = split_1[0] + getitem_26 = split_1[1] + getitem_27 = split_1[2] + split_1 = None + conv2d_6 = torch.conv2d( + getitem_25, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 16, + ) + getitem_25 = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = (None) + conv2d_7 = torch.conv2d( + getitem_26, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 24, + ) + getitem_26 = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = (None) + conv2d_8 = torch.conv2d( + getitem_27, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 24, + ) + getitem_27 = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = (None) + conv_v_img_2 = torch.cat([conv2d_6, conv2d_7, conv2d_8], dim=1) + conv2d_6 = conv2d_7 = conv2d_8 = None + reshape_6 = conv_v_img_2.reshape(1, 8, 8, 3136) + conv_v_img_2 = None + conv_v_img_3 = reshape_6.transpose(-1, -2) + reshape_6 = None + EV_hat_2 = q_img_1 * conv_v_img_3 + q_img_1 = conv_v_img_3 = None + EV_hat_3 = torch._C._nn.pad(EV_hat_2, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_2 = None + mul_3 = 0.3535533905932738 * factor_att_3 + factor_att_3 = None + x_24 = mul_3 + EV_hat_3 + mul_3 = EV_hat_3 = None + transpose_12 = x_24.transpose(1, 2) + x_24 = None + x_25 = transpose_12.reshape(1, 3137, 64) + transpose_12 = None + x_26 = torch._C._nn.linear( + x_25, + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_25 = l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = x_22 + x_27 + x_22 = x_27 = None + x_29 = torch.nn.functional.layer_norm( + x_28, + (64,), + l_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_ + ) = None + x_30 = torch._C._nn.linear( + x_29, + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_29 = l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_31 = torch._C._nn.gelu(x_30, approximate="none") + x_30 = None + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = torch._C._nn.linear( + x_32, + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_32 = l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = x_28 + x_34 + x_28 = x_34 = None + getitem_28 = x_35[ + (slice(None, None, None), slice(1, None, None), slice(None, None, None)) + ] + x_35 = None + reshape_8 = getitem_28.reshape(1, 56, 56, -1) + getitem_28 = None + permute_2 = reshape_8.permute(0, 3, 1, 2) + reshape_8 = None + x1_nocls = permute_2.contiguous() + permute_2 = None + x_36 = torch.conv2d( + x1_nocls, + l_self_modules_patch_embed2_modules_proj_parameters_weight_, + l_self_modules_patch_embed2_modules_proj_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x1_nocls = ( + l_self_modules_patch_embed2_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed2_modules_proj_parameters_bias_ = None + flatten_3 = x_36.flatten(2) + x_36 = None + x_37 = flatten_3.transpose(1, 2) + flatten_3 = None + x_38 = torch.nn.functional.layer_norm( + x_37, + (128,), + l_self_modules_patch_embed2_modules_norm_parameters_weight_, + l_self_modules_patch_embed2_modules_norm_parameters_bias_, + 1e-05, + ) + x_37 = ( + l_self_modules_patch_embed2_modules_norm_parameters_weight_ + ) = l_self_modules_patch_embed2_modules_norm_parameters_bias_ = None + cls_tokens_1 = l_self_parameters_cls_token2_.expand(1, -1, -1) + l_self_parameters_cls_token2_ = None + x_39 = torch.cat((cls_tokens_1, x_38), dim=1) + cls_tokens_1 = x_38 = None + cls_token_2 = x_39[(slice(None, None, None), slice(None, 1, None))] + img_tokens_2 = x_39[(slice(None, None, None), slice(1, None, None))] + x_39 = None + transpose_14 = img_tokens_2.transpose(1, 2) + img_tokens_2 = None + feat_2 = transpose_14.view(1, 128, 28, 28) + transpose_14 = None + conv2d_10 = torch.conv2d( + feat_2, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 128, + ) + x_40 = conv2d_10 + feat_2 + conv2d_10 = feat_2 = None + flatten_4 = x_40.flatten(2) + x_40 = None + x_41 = flatten_4.transpose(1, 2) + flatten_4 = None + x_42 = torch.cat((cls_token_2, x_41), dim=1) + cls_token_2 = x_41 = None + x_43 = torch.nn.functional.layer_norm( + x_42, + (128,), + l_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_43, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_43 = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_9 = linear_8.reshape(1, 785, 3, 8, 16) + linear_8 = None + qkv_2 = reshape_9.permute(2, 0, 3, 1, 4) + reshape_9 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + k_softmax_2 = k_2.softmax(dim=2) + k_2 = None + transpose_16 = k_softmax_2.transpose(-1, -2) + k_softmax_2 = None + factor_att_4 = transpose_16 @ v_2 + transpose_16 = None + factor_att_5 = q_2 @ factor_att_4 + factor_att_4 = None + q_img_2 = q_2[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_2 = None + v_img_4 = v_2[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_2 = None + transpose_17 = v_img_4.transpose(-1, -2) + v_img_4 = None + v_img_5 = transpose_17.reshape(1, 128, 28, 28) + transpose_17 = None + split_2 = torch.functional.split(v_img_5, [32, 48, 48], dim=1) + v_img_5 = None + getitem_36 = split_2[0] + getitem_37 = split_2[1] + getitem_38 = split_2[2] + split_2 = None + conv2d_11 = torch.conv2d( + getitem_36, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + getitem_36 = None + conv2d_12 = torch.conv2d( + getitem_37, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 48, + ) + getitem_37 = None + conv2d_13 = torch.conv2d( + getitem_38, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 48, + ) + getitem_38 = None + conv_v_img_4 = torch.cat([conv2d_11, conv2d_12, conv2d_13], dim=1) + conv2d_11 = conv2d_12 = conv2d_13 = None + reshape_11 = conv_v_img_4.reshape(1, 8, 16, 784) + conv_v_img_4 = None + conv_v_img_5 = reshape_11.transpose(-1, -2) + reshape_11 = None + EV_hat_4 = q_img_2 * conv_v_img_5 + q_img_2 = conv_v_img_5 = None + EV_hat_5 = torch._C._nn.pad(EV_hat_4, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_4 = None + mul_5 = 0.25 * factor_att_5 + factor_att_5 = None + x_44 = mul_5 + EV_hat_5 + mul_5 = EV_hat_5 = None + transpose_19 = x_44.transpose(1, 2) + x_44 = None + x_45 = transpose_19.reshape(1, 785, 128) + transpose_19 = None + x_46 = torch._C._nn.linear( + x_45, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_45 = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + x_48 = x_42 + x_47 + x_42 = x_47 = None + x_49 = torch.nn.functional.layer_norm( + x_48, + (128,), + l_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_ + ) = None + x_50 = torch._C._nn.linear( + x_49, + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_49 = l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_51 = torch._C._nn.gelu(x_50, approximate="none") + x_50 = None + x_52 = torch.nn.functional.dropout(x_51, 0.0, False, False) + x_51 = None + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_52 = l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_54 = torch.nn.functional.dropout(x_53, 0.0, False, False) + x_53 = None + x_55 = x_48 + x_54 + x_48 = x_54 = None + cls_token_3 = x_55[(slice(None, None, None), slice(None, 1, None))] + img_tokens_3 = x_55[(slice(None, None, None), slice(1, None, None))] + x_55 = None + transpose_20 = img_tokens_3.transpose(1, 2) + img_tokens_3 = None + feat_3 = transpose_20.view(1, 128, 28, 28) + transpose_20 = None + conv2d_14 = torch.conv2d( + feat_3, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 128, + ) + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_ = (None) + x_56 = conv2d_14 + feat_3 + conv2d_14 = feat_3 = None + flatten_5 = x_56.flatten(2) + x_56 = None + x_57 = flatten_5.transpose(1, 2) + flatten_5 = None + x_58 = torch.cat((cls_token_3, x_57), dim=1) + cls_token_3 = x_57 = None + x_59 = torch.nn.functional.layer_norm( + x_58, + (128,), + l_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_59, + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_59 = l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_13 = linear_12.reshape(1, 785, 3, 8, 16) + linear_12 = None + qkv_3 = reshape_13.permute(2, 0, 3, 1, 4) + reshape_13 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + k_softmax_3 = k_3.softmax(dim=2) + k_3 = None + transpose_22 = k_softmax_3.transpose(-1, -2) + k_softmax_3 = None + factor_att_6 = transpose_22 @ v_3 + transpose_22 = None + factor_att_7 = q_3 @ factor_att_6 + factor_att_6 = None + q_img_3 = q_3[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_3 = None + v_img_6 = v_3[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_3 = None + transpose_23 = v_img_6.transpose(-1, -2) + v_img_6 = None + v_img_7 = transpose_23.reshape(1, 128, 28, 28) + transpose_23 = None + split_3 = torch.functional.split(v_img_7, [32, 48, 48], dim=1) + v_img_7 = None + getitem_46 = split_3[0] + getitem_47 = split_3[1] + getitem_48 = split_3[2] + split_3 = None + conv2d_15 = torch.conv2d( + getitem_46, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + getitem_46 = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = (None) + conv2d_16 = torch.conv2d( + getitem_47, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 48, + ) + getitem_47 = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = (None) + conv2d_17 = torch.conv2d( + getitem_48, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 48, + ) + getitem_48 = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = (None) + conv_v_img_6 = torch.cat([conv2d_15, conv2d_16, conv2d_17], dim=1) + conv2d_15 = conv2d_16 = conv2d_17 = None + reshape_15 = conv_v_img_6.reshape(1, 8, 16, 784) + conv_v_img_6 = None + conv_v_img_7 = reshape_15.transpose(-1, -2) + reshape_15 = None + EV_hat_6 = q_img_3 * conv_v_img_7 + q_img_3 = conv_v_img_7 = None + EV_hat_7 = torch._C._nn.pad(EV_hat_6, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_6 = None + mul_7 = 0.25 * factor_att_7 + factor_att_7 = None + x_60 = mul_7 + EV_hat_7 + mul_7 = EV_hat_7 = None + transpose_25 = x_60.transpose(1, 2) + x_60 = None + x_61 = transpose_25.reshape(1, 785, 128) + transpose_25 = None + x_62 = torch._C._nn.linear( + x_61, + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_61 = l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_63 = torch.nn.functional.dropout(x_62, 0.0, False, False) + x_62 = None + x_64 = x_58 + x_63 + x_58 = x_63 = None + x_65 = torch.nn.functional.layer_norm( + x_64, + (128,), + l_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_ + ) = None + x_66 = torch._C._nn.linear( + x_65, + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_65 = l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_67 = torch._C._nn.gelu(x_66, approximate="none") + x_66 = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = torch._C._nn.linear( + x_68, + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_68 = l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_71 = x_64 + x_70 + x_64 = x_70 = None + getitem_49 = x_71[ + (slice(None, None, None), slice(1, None, None), slice(None, None, None)) + ] + x_71 = None + reshape_17 = getitem_49.reshape(1, 28, 28, -1) + getitem_49 = None + permute_5 = reshape_17.permute(0, 3, 1, 2) + reshape_17 = None + x2_nocls = permute_5.contiguous() + permute_5 = None + x_72 = torch.conv2d( + x2_nocls, + l_self_modules_patch_embed3_modules_proj_parameters_weight_, + l_self_modules_patch_embed3_modules_proj_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x2_nocls = ( + l_self_modules_patch_embed3_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed3_modules_proj_parameters_bias_ = None + flatten_6 = x_72.flatten(2) + x_72 = None + x_73 = flatten_6.transpose(1, 2) + flatten_6 = None + x_74 = torch.nn.functional.layer_norm( + x_73, + (320,), + l_self_modules_patch_embed3_modules_norm_parameters_weight_, + l_self_modules_patch_embed3_modules_norm_parameters_bias_, + 1e-05, + ) + x_73 = ( + l_self_modules_patch_embed3_modules_norm_parameters_weight_ + ) = l_self_modules_patch_embed3_modules_norm_parameters_bias_ = None + cls_tokens_2 = l_self_parameters_cls_token3_.expand(1, -1, -1) + l_self_parameters_cls_token3_ = None + x_75 = torch.cat((cls_tokens_2, x_74), dim=1) + cls_tokens_2 = x_74 = None + cls_token_4 = x_75[(slice(None, None, None), slice(None, 1, None))] + img_tokens_4 = x_75[(slice(None, None, None), slice(1, None, None))] + x_75 = None + transpose_27 = img_tokens_4.transpose(1, 2) + img_tokens_4 = None + feat_4 = transpose_27.view(1, 320, 14, 14) + transpose_27 = None + conv2d_19 = torch.conv2d( + feat_4, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 320, + ) + x_76 = conv2d_19 + feat_4 + conv2d_19 = feat_4 = None + flatten_7 = x_76.flatten(2) + x_76 = None + x_77 = flatten_7.transpose(1, 2) + flatten_7 = None + x_78 = torch.cat((cls_token_4, x_77), dim=1) + cls_token_4 = x_77 = None + x_79 = torch.nn.functional.layer_norm( + x_78, + (320,), + l_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_79, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_79 = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_18 = linear_16.reshape(1, 197, 3, 8, 40) + linear_16 = None + qkv_4 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + k_softmax_4 = k_4.softmax(dim=2) + k_4 = None + transpose_29 = k_softmax_4.transpose(-1, -2) + k_softmax_4 = None + factor_att_8 = transpose_29 @ v_4 + transpose_29 = None + factor_att_9 = q_4 @ factor_att_8 + factor_att_8 = None + q_img_4 = q_4[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_4 = None + v_img_8 = v_4[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_4 = None + transpose_30 = v_img_8.transpose(-1, -2) + v_img_8 = None + v_img_9 = transpose_30.reshape(1, 320, 14, 14) + transpose_30 = None + split_4 = torch.functional.split(v_img_9, [80, 120, 120], dim=1) + v_img_9 = None + getitem_57 = split_4[0] + getitem_58 = split_4[1] + getitem_59 = split_4[2] + split_4 = None + conv2d_20 = torch.conv2d( + getitem_57, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 80, + ) + getitem_57 = None + conv2d_21 = torch.conv2d( + getitem_58, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 120, + ) + getitem_58 = None + conv2d_22 = torch.conv2d( + getitem_59, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 120, + ) + getitem_59 = None + conv_v_img_8 = torch.cat([conv2d_20, conv2d_21, conv2d_22], dim=1) + conv2d_20 = conv2d_21 = conv2d_22 = None + reshape_20 = conv_v_img_8.reshape(1, 8, 40, 196) + conv_v_img_8 = None + conv_v_img_9 = reshape_20.transpose(-1, -2) + reshape_20 = None + EV_hat_8 = q_img_4 * conv_v_img_9 + q_img_4 = conv_v_img_9 = None + EV_hat_9 = torch._C._nn.pad(EV_hat_8, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_8 = None + mul_9 = 0.15811388300841897 * factor_att_9 + factor_att_9 = None + x_80 = mul_9 + EV_hat_9 + mul_9 = EV_hat_9 = None + transpose_32 = x_80.transpose(1, 2) + x_80 = None + x_81 = transpose_32.reshape(1, 197, 320) + transpose_32 = None + x_82 = torch._C._nn.linear( + x_81, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_81 = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + x_84 = x_78 + x_83 + x_78 = x_83 = None + x_85 = torch.nn.functional.layer_norm( + x_84, + (320,), + l_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_ + ) = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_85 = l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_87 = torch._C._nn.gelu(x_86, approximate="none") + x_86 = None + x_88 = torch.nn.functional.dropout(x_87, 0.0, False, False) + x_87 = None + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_88 = l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_90 = torch.nn.functional.dropout(x_89, 0.0, False, False) + x_89 = None + x_91 = x_84 + x_90 + x_84 = x_90 = None + cls_token_5 = x_91[(slice(None, None, None), slice(None, 1, None))] + img_tokens_5 = x_91[(slice(None, None, None), slice(1, None, None))] + x_91 = None + transpose_33 = img_tokens_5.transpose(1, 2) + img_tokens_5 = None + feat_5 = transpose_33.view(1, 320, 14, 14) + transpose_33 = None + conv2d_23 = torch.conv2d( + feat_5, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 320, + ) + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_ = (None) + x_92 = conv2d_23 + feat_5 + conv2d_23 = feat_5 = None + flatten_8 = x_92.flatten(2) + x_92 = None + x_93 = flatten_8.transpose(1, 2) + flatten_8 = None + x_94 = torch.cat((cls_token_5, x_93), dim=1) + cls_token_5 = x_93 = None + x_95 = torch.nn.functional.layer_norm( + x_94, + (320,), + l_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_95, + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_95 = l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_22 = linear_20.reshape(1, 197, 3, 8, 40) + linear_20 = None + qkv_5 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + k_softmax_5 = k_5.softmax(dim=2) + k_5 = None + transpose_35 = k_softmax_5.transpose(-1, -2) + k_softmax_5 = None + factor_att_10 = transpose_35 @ v_5 + transpose_35 = None + factor_att_11 = q_5 @ factor_att_10 + factor_att_10 = None + q_img_5 = q_5[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_5 = None + v_img_10 = v_5[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_5 = None + transpose_36 = v_img_10.transpose(-1, -2) + v_img_10 = None + v_img_11 = transpose_36.reshape(1, 320, 14, 14) + transpose_36 = None + split_5 = torch.functional.split(v_img_11, [80, 120, 120], dim=1) + v_img_11 = None + getitem_67 = split_5[0] + getitem_68 = split_5[1] + getitem_69 = split_5[2] + split_5 = None + conv2d_24 = torch.conv2d( + getitem_67, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 80, + ) + getitem_67 = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = (None) + conv2d_25 = torch.conv2d( + getitem_68, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 120, + ) + getitem_68 = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = (None) + conv2d_26 = torch.conv2d( + getitem_69, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 120, + ) + getitem_69 = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = (None) + conv_v_img_10 = torch.cat([conv2d_24, conv2d_25, conv2d_26], dim=1) + conv2d_24 = conv2d_25 = conv2d_26 = None + reshape_24 = conv_v_img_10.reshape(1, 8, 40, 196) + conv_v_img_10 = None + conv_v_img_11 = reshape_24.transpose(-1, -2) + reshape_24 = None + EV_hat_10 = q_img_5 * conv_v_img_11 + q_img_5 = conv_v_img_11 = None + EV_hat_11 = torch._C._nn.pad(EV_hat_10, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_10 = None + mul_11 = 0.15811388300841897 * factor_att_11 + factor_att_11 = None + x_96 = mul_11 + EV_hat_11 + mul_11 = EV_hat_11 = None + transpose_38 = x_96.transpose(1, 2) + x_96 = None + x_97 = transpose_38.reshape(1, 197, 320) + transpose_38 = None + x_98 = torch._C._nn.linear( + x_97, + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_97 = l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_99 = torch.nn.functional.dropout(x_98, 0.0, False, False) + x_98 = None + x_100 = x_94 + x_99 + x_94 = x_99 = None + x_101 = torch.nn.functional.layer_norm( + x_100, + (320,), + l_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_ + ) = None + x_102 = torch._C._nn.linear( + x_101, + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_101 = l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_103 = torch._C._nn.gelu(x_102, approximate="none") + x_102 = None + x_104 = torch.nn.functional.dropout(x_103, 0.0, False, False) + x_103 = None + x_105 = torch._C._nn.linear( + x_104, + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_104 = l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = x_100 + x_106 + x_100 = x_106 = None + getitem_70 = x_107[ + (slice(None, None, None), slice(1, None, None), slice(None, None, None)) + ] + x_107 = None + reshape_26 = getitem_70.reshape(1, 14, 14, -1) + getitem_70 = None + permute_8 = reshape_26.permute(0, 3, 1, 2) + reshape_26 = None + x3_nocls = permute_8.contiguous() + permute_8 = None + x_108 = torch.conv2d( + x3_nocls, + l_self_modules_patch_embed4_modules_proj_parameters_weight_, + l_self_modules_patch_embed4_modules_proj_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x3_nocls = ( + l_self_modules_patch_embed4_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed4_modules_proj_parameters_bias_ = None + flatten_9 = x_108.flatten(2) + x_108 = None + x_109 = flatten_9.transpose(1, 2) + flatten_9 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (512,), + l_self_modules_patch_embed4_modules_norm_parameters_weight_, + l_self_modules_patch_embed4_modules_norm_parameters_bias_, + 1e-05, + ) + x_109 = ( + l_self_modules_patch_embed4_modules_norm_parameters_weight_ + ) = l_self_modules_patch_embed4_modules_norm_parameters_bias_ = None + cls_tokens_3 = l_self_parameters_cls_token4_.expand(1, -1, -1) + l_self_parameters_cls_token4_ = None + x_111 = torch.cat((cls_tokens_3, x_110), dim=1) + cls_tokens_3 = x_110 = None + cls_token_6 = x_111[(slice(None, None, None), slice(None, 1, None))] + img_tokens_6 = x_111[(slice(None, None, None), slice(1, None, None))] + x_111 = None + transpose_40 = img_tokens_6.transpose(1, 2) + img_tokens_6 = None + feat_6 = transpose_40.view(1, 512, 7, 7) + transpose_40 = None + conv2d_28 = torch.conv2d( + feat_6, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 512, + ) + x_112 = conv2d_28 + feat_6 + conv2d_28 = feat_6 = None + flatten_10 = x_112.flatten(2) + x_112 = None + x_113 = flatten_10.transpose(1, 2) + flatten_10 = None + x_114 = torch.cat((cls_token_6, x_113), dim=1) + cls_token_6 = x_113 = None + x_115 = torch.nn.functional.layer_norm( + x_114, + (512,), + l_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_115, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_115 = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_27 = linear_24.reshape(1, 50, 3, 8, 64) + linear_24 = None + qkv_6 = reshape_27.permute(2, 0, 3, 1, 4) + reshape_27 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + k_softmax_6 = k_6.softmax(dim=2) + k_6 = None + transpose_42 = k_softmax_6.transpose(-1, -2) + k_softmax_6 = None + factor_att_12 = transpose_42 @ v_6 + transpose_42 = None + factor_att_13 = q_6 @ factor_att_12 + factor_att_12 = None + q_img_6 = q_6[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_6 = None + v_img_12 = v_6[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_6 = None + transpose_43 = v_img_12.transpose(-1, -2) + v_img_12 = None + v_img_13 = transpose_43.reshape(1, 512, 7, 7) + transpose_43 = None + split_6 = torch.functional.split(v_img_13, [128, 192, 192], dim=1) + v_img_13 = None + getitem_78 = split_6[0] + getitem_79 = split_6[1] + getitem_80 = split_6[2] + split_6 = None + conv2d_29 = torch.conv2d( + getitem_78, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 128, + ) + getitem_78 = None + conv2d_30 = torch.conv2d( + getitem_79, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 192, + ) + getitem_79 = None + conv2d_31 = torch.conv2d( + getitem_80, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + getitem_80 = None + conv_v_img_12 = torch.cat([conv2d_29, conv2d_30, conv2d_31], dim=1) + conv2d_29 = conv2d_30 = conv2d_31 = None + reshape_29 = conv_v_img_12.reshape(1, 8, 64, 49) + conv_v_img_12 = None + conv_v_img_13 = reshape_29.transpose(-1, -2) + reshape_29 = None + EV_hat_12 = q_img_6 * conv_v_img_13 + q_img_6 = conv_v_img_13 = None + EV_hat_13 = torch._C._nn.pad(EV_hat_12, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_12 = None + mul_13 = 0.125 * factor_att_13 + factor_att_13 = None + x_116 = mul_13 + EV_hat_13 + mul_13 = EV_hat_13 = None + transpose_45 = x_116.transpose(1, 2) + x_116 = None + x_117 = transpose_45.reshape(1, 50, 512) + transpose_45 = None + x_118 = torch._C._nn.linear( + x_117, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_117 = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_119 = torch.nn.functional.dropout(x_118, 0.0, False, False) + x_118 = None + x_120 = x_114 + x_119 + x_114 = x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (512,), + l_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_ + ) = None + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122, approximate="none") + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_124 = l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_120 + x_126 + x_120 = x_126 = None + cls_token_7 = x_127[(slice(None, None, None), slice(None, 1, None))] + img_tokens_7 = x_127[(slice(None, None, None), slice(1, None, None))] + x_127 = None + transpose_46 = img_tokens_7.transpose(1, 2) + img_tokens_7 = None + feat_7 = transpose_46.view(1, 512, 7, 7) + transpose_46 = None + conv2d_32 = torch.conv2d( + feat_7, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 512, + ) + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_ = (None) + x_128 = conv2d_32 + feat_7 + conv2d_32 = feat_7 = None + flatten_11 = x_128.flatten(2) + x_128 = None + x_129 = flatten_11.transpose(1, 2) + flatten_11 = None + x_130 = torch.cat((cls_token_7, x_129), dim=1) + cls_token_7 = x_129 = None + x_131 = torch.nn.functional.layer_norm( + x_130, + (512,), + l_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_131, + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_131 = l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_31 = linear_28.reshape(1, 50, 3, 8, 64) + linear_28 = None + qkv_7 = reshape_31.permute(2, 0, 3, 1, 4) + reshape_31 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + k_softmax_7 = k_7.softmax(dim=2) + k_7 = None + transpose_48 = k_softmax_7.transpose(-1, -2) + k_softmax_7 = None + factor_att_14 = transpose_48 @ v_7 + transpose_48 = None + factor_att_15 = q_7 @ factor_att_14 + factor_att_14 = None + q_img_7 = q_7[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_7 = None + v_img_14 = v_7[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_7 = None + transpose_49 = v_img_14.transpose(-1, -2) + v_img_14 = None + v_img_15 = transpose_49.reshape(1, 512, 7, 7) + transpose_49 = None + split_7 = torch.functional.split(v_img_15, [128, 192, 192], dim=1) + v_img_15 = None + getitem_88 = split_7[0] + getitem_89 = split_7[1] + getitem_90 = split_7[2] + split_7 = None + conv2d_33 = torch.conv2d( + getitem_88, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 128, + ) + getitem_88 = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = (None) + conv2d_34 = torch.conv2d( + getitem_89, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 192, + ) + getitem_89 = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = (None) + conv2d_35 = torch.conv2d( + getitem_90, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + getitem_90 = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = (None) + conv_v_img_14 = torch.cat([conv2d_33, conv2d_34, conv2d_35], dim=1) + conv2d_33 = conv2d_34 = conv2d_35 = None + reshape_33 = conv_v_img_14.reshape(1, 8, 64, 49) + conv_v_img_14 = None + conv_v_img_15 = reshape_33.transpose(-1, -2) + reshape_33 = None + EV_hat_14 = q_img_7 * conv_v_img_15 + q_img_7 = conv_v_img_15 = None + EV_hat_15 = torch._C._nn.pad(EV_hat_14, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_14 = None + mul_15 = 0.125 * factor_att_15 + factor_att_15 = None + x_132 = mul_15 + EV_hat_15 + mul_15 = EV_hat_15 = None + transpose_51 = x_132.transpose(1, 2) + x_132 = None + x_133 = transpose_51.reshape(1, 50, 512) + transpose_51 = None + x_134 = torch._C._nn.linear( + x_133, + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_133 = l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = x_130 + x_135 + x_130 = x_135 = None + x_137 = torch.nn.functional.layer_norm( + x_136, + (512,), + l_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_ + ) = None + x_138 = torch._C._nn.linear( + x_137, + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_137 = l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_139 = torch._C._nn.gelu(x_138, approximate="none") + x_138 = None + x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) + x_139 = None + x_141 = torch._C._nn.linear( + x_140, + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_140 = l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_136 + x_142 + x_136 = x_142 = None + getitem_91 = x_143[ + (slice(None, None, None), slice(1, None, None), slice(None, None, None)) + ] + reshape_35 = getitem_91.reshape(1, 7, 7, -1) + getitem_91 = None + permute_11 = reshape_35.permute(0, 3, 1, 2) + reshape_35 = None + x4_nocls = permute_11.contiguous() + permute_11 = x4_nocls = None + x_144 = torch.nn.functional.layer_norm( + x_143, + (512,), + l_self_modules_norm4_parameters_weight_, + l_self_modules_norm4_parameters_bias_, + 1e-06, + ) + x_143 = ( + l_self_modules_norm4_parameters_weight_ + ) = l_self_modules_norm4_parameters_bias_ = None + x_145 = x_144[(slice(None, None, None), 0)] + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = torch._C._nn.linear( + x_146, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_146 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_147,) diff --git a/samples/timm/coat_lite_mini.in1k/weight_meta.py b/samples/timm/coat_lite_mini.in1k/weight_meta.py new file mode 100644 index 0000000000..f45c7960a8 --- /dev/null +++ b/samples/timm/coat_lite_mini.in1k/weight_meta.py @@ -0,0 +1,1528 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.223 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed1_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed1_modules_proj_parameters_weight_" + shape = [64, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed1_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed1_modules_proj_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.004 + std = 0.081 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed1_modules_norm_parameters_weight_: + name = "L_self_modules_patch_embed1_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed1_modules_norm_parameters_bias_: + name = "L_self_modules_patch_embed1_modules_norm_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token1_: + name = "L_self_parameters_cls_token1_" + shape = [1, 1, 64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_" + shape = [64, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.018 + std = 0.189 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.059 + std = 0.193 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [192, 64] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_" + shape = [16, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.012 + std = 0.199 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_" + shape = [16] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.007 + std = 0.222 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_" + shape = [24, 1, 5, 5] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.016 + std = 0.109 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_" + shape = [24, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.006 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [64, 64] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 64] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [192, 64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [64, 64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed2_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed2_modules_proj_parameters_weight_" + shape = [128, 64, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed2_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed2_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed2_modules_norm_parameters_weight_: + name = "L_self_modules_patch_embed2_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed2_modules_norm_parameters_bias_: + name = "L_self_modules_patch_embed2_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token2_: + name = "L_self_parameters_cls_token2_" + shape = [1, 1, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_" + shape = [128, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.004 + std = 0.189 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.004 + std = 0.210 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_" + shape = [32, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.011 + std = 0.190 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.060 + std = 0.186 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_" + shape = [48, 1, 5, 5] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.113 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.007 + std = 0.119 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_" + shape = [48, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.081 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.085 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed3_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed3_modules_proj_parameters_weight_" + shape = [320, 128, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed3_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed3_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed3_modules_norm_parameters_weight_: + name = "L_self_modules_patch_embed3_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed3_modules_norm_parameters_bias_: + name = "L_self_modules_patch_embed3_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token3_: + name = "L_self_parameters_cls_token3_" + shape = [1, 1, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_" + shape = [320, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.192 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.006 + std = 0.192 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_" + shape = [80, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.197 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.194 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_" + shape = [120, 1, 5, 5] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.115 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.009 + std = 0.110 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_" + shape = [120, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.008 + std = 0.084 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed4_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed4_modules_proj_parameters_weight_" + shape = [512, 320, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.016 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed4_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed4_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.016 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed4_modules_norm_parameters_weight_: + name = "L_self_modules_patch_embed4_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed4_modules_norm_parameters_bias_: + name = "L_self_modules_patch_embed4_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token4_: + name = "L_self_parameters_cls_token4_" + shape = [1, 1, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_" + shape = [512, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.193 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.015 + std = 0.192 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_" + shape = [128, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.192 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.178 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_" + shape = [192, 1, 5, 5] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.117 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.014 + std = 0.115 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.084 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm4_parameters_weight_: + name = "L_self_modules_norm4_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm4_parameters_bias_: + name = "L_self_modules_norm4_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/coat_lite_small.in1k/graph_hash.txt b/samples/timm/coat_lite_small.in1k/graph_hash.txt new file mode 100644 index 0000000000..4b3af640e1 --- /dev/null +++ b/samples/timm/coat_lite_small.in1k/graph_hash.txt @@ -0,0 +1 @@ +490850194b83b142b6f65c861665b9cd1b46288297fdafeeb5a8eec6473a0f99 \ No newline at end of file diff --git a/samples/timm/coat_lite_small.in1k/graph_net.json b/samples/timm/coat_lite_small.in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/coat_lite_small.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/coat_lite_small.in1k/input_meta.py b/samples/timm/coat_lite_small.in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/coat_lite_small.in1k/input_tensor_constraints.py b/samples/timm/coat_lite_small.in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/coat_lite_small.in1k/model.py b/samples/timm/coat_lite_small.in1k/model.py new file mode 100644 index 0000000000..b9c9651bf3 --- /dev/null +++ b/samples/timm/coat_lite_small.in1k/model.py @@ -0,0 +1,3637 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed1_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed1_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token1_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed2_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed2_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token2_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed3_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed3_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token3_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed4_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed4_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token4_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm4_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm4_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed1_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed1_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed1_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed1_modules_proj_parameters_bias_ + ) + l_self_modules_patch_embed1_modules_norm_parameters_weight_ = ( + L_self_modules_patch_embed1_modules_norm_parameters_weight_ + ) + l_self_modules_patch_embed1_modules_norm_parameters_bias_ = ( + L_self_modules_patch_embed1_modules_norm_parameters_bias_ + ) + l_self_parameters_cls_token1_ = L_self_parameters_cls_token1_ + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_patch_embed2_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed2_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed2_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed2_modules_proj_parameters_bias_ + ) + l_self_modules_patch_embed2_modules_norm_parameters_weight_ = ( + L_self_modules_patch_embed2_modules_norm_parameters_weight_ + ) + l_self_modules_patch_embed2_modules_norm_parameters_bias_ = ( + L_self_modules_patch_embed2_modules_norm_parameters_bias_ + ) + l_self_parameters_cls_token2_ = L_self_parameters_cls_token2_ + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_patch_embed3_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed3_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed3_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed3_modules_proj_parameters_bias_ + ) + l_self_modules_patch_embed3_modules_norm_parameters_weight_ = ( + L_self_modules_patch_embed3_modules_norm_parameters_weight_ + ) + l_self_modules_patch_embed3_modules_norm_parameters_bias_ = ( + L_self_modules_patch_embed3_modules_norm_parameters_bias_ + ) + l_self_parameters_cls_token3_ = L_self_parameters_cls_token3_ + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_patch_embed4_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed4_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed4_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed4_modules_proj_parameters_bias_ + ) + l_self_modules_patch_embed4_modules_norm_parameters_weight_ = ( + L_self_modules_patch_embed4_modules_norm_parameters_weight_ + ) + l_self_modules_patch_embed4_modules_norm_parameters_bias_ = ( + L_self_modules_patch_embed4_modules_norm_parameters_bias_ + ) + l_self_parameters_cls_token4_ = L_self_parameters_cls_token4_ + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_norm4_parameters_weight_ = ( + L_self_modules_norm4_parameters_weight_ + ) + l_self_modules_norm4_parameters_bias_ = L_self_modules_norm4_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed1_modules_proj_parameters_weight_, + l_self_modules_patch_embed1_modules_proj_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed1_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed1_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + x_2 = torch.nn.functional.layer_norm( + x_1, + (64,), + l_self_modules_patch_embed1_modules_norm_parameters_weight_, + l_self_modules_patch_embed1_modules_norm_parameters_bias_, + 1e-05, + ) + x_1 = ( + l_self_modules_patch_embed1_modules_norm_parameters_weight_ + ) = l_self_modules_patch_embed1_modules_norm_parameters_bias_ = None + cls_tokens = l_self_parameters_cls_token1_.expand(1, -1, -1) + l_self_parameters_cls_token1_ = None + x_3 = torch.cat((cls_tokens, x_2), dim=1) + cls_tokens = x_2 = None + cls_token = x_3[(slice(None, None, None), slice(None, 1, None))] + img_tokens = x_3[(slice(None, None, None), slice(1, None, None))] + x_3 = None + transpose_1 = img_tokens.transpose(1, 2) + img_tokens = None + feat = transpose_1.view(1, 64, 56, 56) + transpose_1 = None + conv2d_1 = torch.conv2d( + feat, + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 64, + ) + x_4 = conv2d_1 + feat + conv2d_1 = feat = None + flatten_1 = x_4.flatten(2) + x_4 = None + x_5 = flatten_1.transpose(1, 2) + flatten_1 = None + x_6 = torch.cat((cls_token, x_5), dim=1) + cls_token = x_5 = None + x_7 = torch.nn.functional.layer_norm( + x_6, + (64,), + l_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_7, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_7 = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 3137, 3, 8, 8) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + k_softmax = k.softmax(dim=2) + k = None + transpose_3 = k_softmax.transpose(-1, -2) + k_softmax = None + factor_att = transpose_3 @ v + transpose_3 = None + factor_att_1 = q @ factor_att + factor_att = None + q_img = q[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q = None + v_img = v[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v = None + transpose_4 = v_img.transpose(-1, -2) + v_img = None + v_img_1 = transpose_4.reshape(1, 64, 56, 56) + transpose_4 = None + split = torch.functional.split(v_img_1, [16, 24, 24], dim=1) + v_img_1 = None + getitem_15 = split[0] + getitem_16 = split[1] + getitem_17 = split[2] + split = None + conv2d_2 = torch.conv2d( + getitem_15, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 16, + ) + getitem_15 = None + conv2d_3 = torch.conv2d( + getitem_16, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 24, + ) + getitem_16 = None + conv2d_4 = torch.conv2d( + getitem_17, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 24, + ) + getitem_17 = None + conv_v_img = torch.cat([conv2d_2, conv2d_3, conv2d_4], dim=1) + conv2d_2 = conv2d_3 = conv2d_4 = None + reshape_2 = conv_v_img.reshape(1, 8, 8, 3136) + conv_v_img = None + conv_v_img_1 = reshape_2.transpose(-1, -2) + reshape_2 = None + EV_hat = q_img * conv_v_img_1 + q_img = conv_v_img_1 = None + EV_hat_1 = torch._C._nn.pad(EV_hat, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat = None + mul_1 = 0.3535533905932738 * factor_att_1 + factor_att_1 = None + x_8 = mul_1 + EV_hat_1 + mul_1 = EV_hat_1 = None + transpose_6 = x_8.transpose(1, 2) + x_8 = None + x_9 = transpose_6.reshape(1, 3137, 64) + transpose_6 = None + x_10 = torch._C._nn.linear( + x_9, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_9 = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = x_6 + x_11 + x_6 = x_11 = None + x_13 = torch.nn.functional.layer_norm( + x_12, + (64,), + l_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_ + ) = None + x_14 = torch._C._nn.linear( + x_13, + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_13 = l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_15 = torch._C._nn.gelu(x_14, approximate="none") + x_14 = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_16 = l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_18 = torch.nn.functional.dropout(x_17, 0.0, False, False) + x_17 = None + x_19 = x_12 + x_18 + x_12 = x_18 = None + cls_token_1 = x_19[(slice(None, None, None), slice(None, 1, None))] + img_tokens_1 = x_19[(slice(None, None, None), slice(1, None, None))] + x_19 = None + transpose_7 = img_tokens_1.transpose(1, 2) + img_tokens_1 = None + feat_1 = transpose_7.view(1, 64, 56, 56) + transpose_7 = None + conv2d_5 = torch.conv2d( + feat_1, + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 64, + ) + x_20 = conv2d_5 + feat_1 + conv2d_5 = feat_1 = None + flatten_2 = x_20.flatten(2) + x_20 = None + x_21 = flatten_2.transpose(1, 2) + flatten_2 = None + x_22 = torch.cat((cls_token_1, x_21), dim=1) + cls_token_1 = x_21 = None + x_23 = torch.nn.functional.layer_norm( + x_22, + (64,), + l_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_23, + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_23 = l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_4.reshape(1, 3137, 3, 8, 8) + linear_4 = None + qkv_1 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + k_softmax_1 = k_1.softmax(dim=2) + k_1 = None + transpose_9 = k_softmax_1.transpose(-1, -2) + k_softmax_1 = None + factor_att_2 = transpose_9 @ v_1 + transpose_9 = None + factor_att_3 = q_1 @ factor_att_2 + factor_att_2 = None + q_img_1 = q_1[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_1 = None + v_img_2 = v_1[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_1 = None + transpose_10 = v_img_2.transpose(-1, -2) + v_img_2 = None + v_img_3 = transpose_10.reshape(1, 64, 56, 56) + transpose_10 = None + split_1 = torch.functional.split(v_img_3, [16, 24, 24], dim=1) + v_img_3 = None + getitem_25 = split_1[0] + getitem_26 = split_1[1] + getitem_27 = split_1[2] + split_1 = None + conv2d_6 = torch.conv2d( + getitem_25, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 16, + ) + getitem_25 = None + conv2d_7 = torch.conv2d( + getitem_26, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 24, + ) + getitem_26 = None + conv2d_8 = torch.conv2d( + getitem_27, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 24, + ) + getitem_27 = None + conv_v_img_2 = torch.cat([conv2d_6, conv2d_7, conv2d_8], dim=1) + conv2d_6 = conv2d_7 = conv2d_8 = None + reshape_6 = conv_v_img_2.reshape(1, 8, 8, 3136) + conv_v_img_2 = None + conv_v_img_3 = reshape_6.transpose(-1, -2) + reshape_6 = None + EV_hat_2 = q_img_1 * conv_v_img_3 + q_img_1 = conv_v_img_3 = None + EV_hat_3 = torch._C._nn.pad(EV_hat_2, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_2 = None + mul_3 = 0.3535533905932738 * factor_att_3 + factor_att_3 = None + x_24 = mul_3 + EV_hat_3 + mul_3 = EV_hat_3 = None + transpose_12 = x_24.transpose(1, 2) + x_24 = None + x_25 = transpose_12.reshape(1, 3137, 64) + transpose_12 = None + x_26 = torch._C._nn.linear( + x_25, + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_25 = l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = x_22 + x_27 + x_22 = x_27 = None + x_29 = torch.nn.functional.layer_norm( + x_28, + (64,), + l_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_ + ) = None + x_30 = torch._C._nn.linear( + x_29, + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_29 = l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_31 = torch._C._nn.gelu(x_30, approximate="none") + x_30 = None + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = torch._C._nn.linear( + x_32, + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_32 = l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = x_28 + x_34 + x_28 = x_34 = None + cls_token_2 = x_35[(slice(None, None, None), slice(None, 1, None))] + img_tokens_2 = x_35[(slice(None, None, None), slice(1, None, None))] + x_35 = None + transpose_13 = img_tokens_2.transpose(1, 2) + img_tokens_2 = None + feat_2 = transpose_13.view(1, 64, 56, 56) + transpose_13 = None + conv2d_9 = torch.conv2d( + feat_2, + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 64, + ) + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_ = (None) + x_36 = conv2d_9 + feat_2 + conv2d_9 = feat_2 = None + flatten_3 = x_36.flatten(2) + x_36 = None + x_37 = flatten_3.transpose(1, 2) + flatten_3 = None + x_38 = torch.cat((cls_token_2, x_37), dim=1) + cls_token_2 = x_37 = None + x_39 = torch.nn.functional.layer_norm( + x_38, + (64,), + l_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_39, + l_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_39 = l_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_8 = linear_8.reshape(1, 3137, 3, 8, 8) + linear_8 = None + qkv_2 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + k_softmax_2 = k_2.softmax(dim=2) + k_2 = None + transpose_15 = k_softmax_2.transpose(-1, -2) + k_softmax_2 = None + factor_att_4 = transpose_15 @ v_2 + transpose_15 = None + factor_att_5 = q_2 @ factor_att_4 + factor_att_4 = None + q_img_2 = q_2[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_2 = None + v_img_4 = v_2[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_2 = None + transpose_16 = v_img_4.transpose(-1, -2) + v_img_4 = None + v_img_5 = transpose_16.reshape(1, 64, 56, 56) + transpose_16 = None + split_2 = torch.functional.split(v_img_5, [16, 24, 24], dim=1) + v_img_5 = None + getitem_35 = split_2[0] + getitem_36 = split_2[1] + getitem_37 = split_2[2] + split_2 = None + conv2d_10 = torch.conv2d( + getitem_35, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 16, + ) + getitem_35 = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = (None) + conv2d_11 = torch.conv2d( + getitem_36, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 24, + ) + getitem_36 = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = (None) + conv2d_12 = torch.conv2d( + getitem_37, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 24, + ) + getitem_37 = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = (None) + conv_v_img_4 = torch.cat([conv2d_10, conv2d_11, conv2d_12], dim=1) + conv2d_10 = conv2d_11 = conv2d_12 = None + reshape_10 = conv_v_img_4.reshape(1, 8, 8, 3136) + conv_v_img_4 = None + conv_v_img_5 = reshape_10.transpose(-1, -2) + reshape_10 = None + EV_hat_4 = q_img_2 * conv_v_img_5 + q_img_2 = conv_v_img_5 = None + EV_hat_5 = torch._C._nn.pad(EV_hat_4, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_4 = None + mul_5 = 0.3535533905932738 * factor_att_5 + factor_att_5 = None + x_40 = mul_5 + EV_hat_5 + mul_5 = EV_hat_5 = None + transpose_18 = x_40.transpose(1, 2) + x_40 = None + x_41 = transpose_18.reshape(1, 3137, 64) + transpose_18 = None + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_41 = l_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_43 = torch.nn.functional.dropout(x_42, 0.0, False, False) + x_42 = None + x_44 = x_38 + x_43 + x_38 = x_43 = None + x_45 = torch.nn.functional.layer_norm( + x_44, + (64,), + l_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_bias_ + ) = None + x_46 = torch._C._nn.linear( + x_45, + l_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_45 = l_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_47 = torch._C._nn.gelu(x_46, approximate="none") + x_46 = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = torch._C._nn.linear( + x_48, + l_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_48 = l_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_50 = torch.nn.functional.dropout(x_49, 0.0, False, False) + x_49 = None + x_51 = x_44 + x_50 + x_44 = x_50 = None + getitem_38 = x_51[ + (slice(None, None, None), slice(1, None, None), slice(None, None, None)) + ] + x_51 = None + reshape_12 = getitem_38.reshape(1, 56, 56, -1) + getitem_38 = None + permute_3 = reshape_12.permute(0, 3, 1, 2) + reshape_12 = None + x1_nocls = permute_3.contiguous() + permute_3 = None + x_52 = torch.conv2d( + x1_nocls, + l_self_modules_patch_embed2_modules_proj_parameters_weight_, + l_self_modules_patch_embed2_modules_proj_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x1_nocls = ( + l_self_modules_patch_embed2_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed2_modules_proj_parameters_bias_ = None + flatten_4 = x_52.flatten(2) + x_52 = None + x_53 = flatten_4.transpose(1, 2) + flatten_4 = None + x_54 = torch.nn.functional.layer_norm( + x_53, + (128,), + l_self_modules_patch_embed2_modules_norm_parameters_weight_, + l_self_modules_patch_embed2_modules_norm_parameters_bias_, + 1e-05, + ) + x_53 = ( + l_self_modules_patch_embed2_modules_norm_parameters_weight_ + ) = l_self_modules_patch_embed2_modules_norm_parameters_bias_ = None + cls_tokens_1 = l_self_parameters_cls_token2_.expand(1, -1, -1) + l_self_parameters_cls_token2_ = None + x_55 = torch.cat((cls_tokens_1, x_54), dim=1) + cls_tokens_1 = x_54 = None + cls_token_3 = x_55[(slice(None, None, None), slice(None, 1, None))] + img_tokens_3 = x_55[(slice(None, None, None), slice(1, None, None))] + x_55 = None + transpose_20 = img_tokens_3.transpose(1, 2) + img_tokens_3 = None + feat_3 = transpose_20.view(1, 128, 28, 28) + transpose_20 = None + conv2d_14 = torch.conv2d( + feat_3, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 128, + ) + x_56 = conv2d_14 + feat_3 + conv2d_14 = feat_3 = None + flatten_5 = x_56.flatten(2) + x_56 = None + x_57 = flatten_5.transpose(1, 2) + flatten_5 = None + x_58 = torch.cat((cls_token_3, x_57), dim=1) + cls_token_3 = x_57 = None + x_59 = torch.nn.functional.layer_norm( + x_58, + (128,), + l_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_59, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_59 = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_13 = linear_12.reshape(1, 785, 3, 8, 16) + linear_12 = None + qkv_3 = reshape_13.permute(2, 0, 3, 1, 4) + reshape_13 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + k_softmax_3 = k_3.softmax(dim=2) + k_3 = None + transpose_22 = k_softmax_3.transpose(-1, -2) + k_softmax_3 = None + factor_att_6 = transpose_22 @ v_3 + transpose_22 = None + factor_att_7 = q_3 @ factor_att_6 + factor_att_6 = None + q_img_3 = q_3[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_3 = None + v_img_6 = v_3[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_3 = None + transpose_23 = v_img_6.transpose(-1, -2) + v_img_6 = None + v_img_7 = transpose_23.reshape(1, 128, 28, 28) + transpose_23 = None + split_3 = torch.functional.split(v_img_7, [32, 48, 48], dim=1) + v_img_7 = None + getitem_46 = split_3[0] + getitem_47 = split_3[1] + getitem_48 = split_3[2] + split_3 = None + conv2d_15 = torch.conv2d( + getitem_46, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + getitem_46 = None + conv2d_16 = torch.conv2d( + getitem_47, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 48, + ) + getitem_47 = None + conv2d_17 = torch.conv2d( + getitem_48, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 48, + ) + getitem_48 = None + conv_v_img_6 = torch.cat([conv2d_15, conv2d_16, conv2d_17], dim=1) + conv2d_15 = conv2d_16 = conv2d_17 = None + reshape_15 = conv_v_img_6.reshape(1, 8, 16, 784) + conv_v_img_6 = None + conv_v_img_7 = reshape_15.transpose(-1, -2) + reshape_15 = None + EV_hat_6 = q_img_3 * conv_v_img_7 + q_img_3 = conv_v_img_7 = None + EV_hat_7 = torch._C._nn.pad(EV_hat_6, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_6 = None + mul_7 = 0.25 * factor_att_7 + factor_att_7 = None + x_60 = mul_7 + EV_hat_7 + mul_7 = EV_hat_7 = None + transpose_25 = x_60.transpose(1, 2) + x_60 = None + x_61 = transpose_25.reshape(1, 785, 128) + transpose_25 = None + x_62 = torch._C._nn.linear( + x_61, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_61 = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_63 = torch.nn.functional.dropout(x_62, 0.0, False, False) + x_62 = None + x_64 = x_58 + x_63 + x_58 = x_63 = None + x_65 = torch.nn.functional.layer_norm( + x_64, + (128,), + l_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_ + ) = None + x_66 = torch._C._nn.linear( + x_65, + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_65 = l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_67 = torch._C._nn.gelu(x_66, approximate="none") + x_66 = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = torch._C._nn.linear( + x_68, + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_68 = l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_71 = x_64 + x_70 + x_64 = x_70 = None + cls_token_4 = x_71[(slice(None, None, None), slice(None, 1, None))] + img_tokens_4 = x_71[(slice(None, None, None), slice(1, None, None))] + x_71 = None + transpose_26 = img_tokens_4.transpose(1, 2) + img_tokens_4 = None + feat_4 = transpose_26.view(1, 128, 28, 28) + transpose_26 = None + conv2d_18 = torch.conv2d( + feat_4, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 128, + ) + x_72 = conv2d_18 + feat_4 + conv2d_18 = feat_4 = None + flatten_6 = x_72.flatten(2) + x_72 = None + x_73 = flatten_6.transpose(1, 2) + flatten_6 = None + x_74 = torch.cat((cls_token_4, x_73), dim=1) + cls_token_4 = x_73 = None + x_75 = torch.nn.functional.layer_norm( + x_74, + (128,), + l_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_75, + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_75 = l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_17 = linear_16.reshape(1, 785, 3, 8, 16) + linear_16 = None + qkv_4 = reshape_17.permute(2, 0, 3, 1, 4) + reshape_17 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + k_softmax_4 = k_4.softmax(dim=2) + k_4 = None + transpose_28 = k_softmax_4.transpose(-1, -2) + k_softmax_4 = None + factor_att_8 = transpose_28 @ v_4 + transpose_28 = None + factor_att_9 = q_4 @ factor_att_8 + factor_att_8 = None + q_img_4 = q_4[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_4 = None + v_img_8 = v_4[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_4 = None + transpose_29 = v_img_8.transpose(-1, -2) + v_img_8 = None + v_img_9 = transpose_29.reshape(1, 128, 28, 28) + transpose_29 = None + split_4 = torch.functional.split(v_img_9, [32, 48, 48], dim=1) + v_img_9 = None + getitem_56 = split_4[0] + getitem_57 = split_4[1] + getitem_58 = split_4[2] + split_4 = None + conv2d_19 = torch.conv2d( + getitem_56, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + getitem_56 = None + conv2d_20 = torch.conv2d( + getitem_57, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 48, + ) + getitem_57 = None + conv2d_21 = torch.conv2d( + getitem_58, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 48, + ) + getitem_58 = None + conv_v_img_8 = torch.cat([conv2d_19, conv2d_20, conv2d_21], dim=1) + conv2d_19 = conv2d_20 = conv2d_21 = None + reshape_19 = conv_v_img_8.reshape(1, 8, 16, 784) + conv_v_img_8 = None + conv_v_img_9 = reshape_19.transpose(-1, -2) + reshape_19 = None + EV_hat_8 = q_img_4 * conv_v_img_9 + q_img_4 = conv_v_img_9 = None + EV_hat_9 = torch._C._nn.pad(EV_hat_8, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_8 = None + mul_9 = 0.25 * factor_att_9 + factor_att_9 = None + x_76 = mul_9 + EV_hat_9 + mul_9 = EV_hat_9 = None + transpose_31 = x_76.transpose(1, 2) + x_76 = None + x_77 = transpose_31.reshape(1, 785, 128) + transpose_31 = None + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_77 = l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = x_74 + x_79 + x_74 = x_79 = None + x_81 = torch.nn.functional.layer_norm( + x_80, + (128,), + l_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_ + ) = None + x_82 = torch._C._nn.linear( + x_81, + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_81 = l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_83 = torch._C._nn.gelu(x_82, approximate="none") + x_82 = None + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = torch._C._nn.linear( + x_84, + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_84 = l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_86 = torch.nn.functional.dropout(x_85, 0.0, False, False) + x_85 = None + x_87 = x_80 + x_86 + x_80 = x_86 = None + cls_token_5 = x_87[(slice(None, None, None), slice(None, 1, None))] + img_tokens_5 = x_87[(slice(None, None, None), slice(1, None, None))] + x_87 = None + transpose_32 = img_tokens_5.transpose(1, 2) + img_tokens_5 = None + feat_5 = transpose_32.view(1, 128, 28, 28) + transpose_32 = None + conv2d_22 = torch.conv2d( + feat_5, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 128, + ) + x_88 = conv2d_22 + feat_5 + conv2d_22 = feat_5 = None + flatten_7 = x_88.flatten(2) + x_88 = None + x_89 = flatten_7.transpose(1, 2) + flatten_7 = None + x_90 = torch.cat((cls_token_5, x_89), dim=1) + cls_token_5 = x_89 = None + x_91 = torch.nn.functional.layer_norm( + x_90, + (128,), + l_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_91, + l_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_91 = l_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_21 = linear_20.reshape(1, 785, 3, 8, 16) + linear_20 = None + qkv_5 = reshape_21.permute(2, 0, 3, 1, 4) + reshape_21 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + k_softmax_5 = k_5.softmax(dim=2) + k_5 = None + transpose_34 = k_softmax_5.transpose(-1, -2) + k_softmax_5 = None + factor_att_10 = transpose_34 @ v_5 + transpose_34 = None + factor_att_11 = q_5 @ factor_att_10 + factor_att_10 = None + q_img_5 = q_5[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_5 = None + v_img_10 = v_5[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_5 = None + transpose_35 = v_img_10.transpose(-1, -2) + v_img_10 = None + v_img_11 = transpose_35.reshape(1, 128, 28, 28) + transpose_35 = None + split_5 = torch.functional.split(v_img_11, [32, 48, 48], dim=1) + v_img_11 = None + getitem_66 = split_5[0] + getitem_67 = split_5[1] + getitem_68 = split_5[2] + split_5 = None + conv2d_23 = torch.conv2d( + getitem_66, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + getitem_66 = None + conv2d_24 = torch.conv2d( + getitem_67, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 48, + ) + getitem_67 = None + conv2d_25 = torch.conv2d( + getitem_68, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 48, + ) + getitem_68 = None + conv_v_img_10 = torch.cat([conv2d_23, conv2d_24, conv2d_25], dim=1) + conv2d_23 = conv2d_24 = conv2d_25 = None + reshape_23 = conv_v_img_10.reshape(1, 8, 16, 784) + conv_v_img_10 = None + conv_v_img_11 = reshape_23.transpose(-1, -2) + reshape_23 = None + EV_hat_10 = q_img_5 * conv_v_img_11 + q_img_5 = conv_v_img_11 = None + EV_hat_11 = torch._C._nn.pad(EV_hat_10, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_10 = None + mul_11 = 0.25 * factor_att_11 + factor_att_11 = None + x_92 = mul_11 + EV_hat_11 + mul_11 = EV_hat_11 = None + transpose_37 = x_92.transpose(1, 2) + x_92 = None + x_93 = transpose_37.reshape(1, 785, 128) + transpose_37 = None + x_94 = torch._C._nn.linear( + x_93, + l_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_93 = l_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = x_90 + x_95 + x_90 = x_95 = None + x_97 = torch.nn.functional.layer_norm( + x_96, + (128,), + l_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_bias_ + ) = None + x_98 = torch._C._nn.linear( + x_97, + l_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_97 = l_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_99 = torch._C._nn.gelu(x_98, approximate="none") + x_98 = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + x_101 = torch._C._nn.linear( + x_100, + l_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_100 = l_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = x_96 + x_102 + x_96 = x_102 = None + cls_token_6 = x_103[(slice(None, None, None), slice(None, 1, None))] + img_tokens_6 = x_103[(slice(None, None, None), slice(1, None, None))] + x_103 = None + transpose_38 = img_tokens_6.transpose(1, 2) + img_tokens_6 = None + feat_6 = transpose_38.view(1, 128, 28, 28) + transpose_38 = None + conv2d_26 = torch.conv2d( + feat_6, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 128, + ) + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_ = (None) + x_104 = conv2d_26 + feat_6 + conv2d_26 = feat_6 = None + flatten_8 = x_104.flatten(2) + x_104 = None + x_105 = flatten_8.transpose(1, 2) + flatten_8 = None + x_106 = torch.cat((cls_token_6, x_105), dim=1) + cls_token_6 = x_105 = None + x_107 = torch.nn.functional.layer_norm( + x_106, + (128,), + l_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_107, + l_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_107 = l_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_25 = linear_24.reshape(1, 785, 3, 8, 16) + linear_24 = None + qkv_6 = reshape_25.permute(2, 0, 3, 1, 4) + reshape_25 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + k_softmax_6 = k_6.softmax(dim=2) + k_6 = None + transpose_40 = k_softmax_6.transpose(-1, -2) + k_softmax_6 = None + factor_att_12 = transpose_40 @ v_6 + transpose_40 = None + factor_att_13 = q_6 @ factor_att_12 + factor_att_12 = None + q_img_6 = q_6[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_6 = None + v_img_12 = v_6[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_6 = None + transpose_41 = v_img_12.transpose(-1, -2) + v_img_12 = None + v_img_13 = transpose_41.reshape(1, 128, 28, 28) + transpose_41 = None + split_6 = torch.functional.split(v_img_13, [32, 48, 48], dim=1) + v_img_13 = None + getitem_76 = split_6[0] + getitem_77 = split_6[1] + getitem_78 = split_6[2] + split_6 = None + conv2d_27 = torch.conv2d( + getitem_76, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + getitem_76 = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = (None) + conv2d_28 = torch.conv2d( + getitem_77, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 48, + ) + getitem_77 = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = (None) + conv2d_29 = torch.conv2d( + getitem_78, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 48, + ) + getitem_78 = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = (None) + conv_v_img_12 = torch.cat([conv2d_27, conv2d_28, conv2d_29], dim=1) + conv2d_27 = conv2d_28 = conv2d_29 = None + reshape_27 = conv_v_img_12.reshape(1, 8, 16, 784) + conv_v_img_12 = None + conv_v_img_13 = reshape_27.transpose(-1, -2) + reshape_27 = None + EV_hat_12 = q_img_6 * conv_v_img_13 + q_img_6 = conv_v_img_13 = None + EV_hat_13 = torch._C._nn.pad(EV_hat_12, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_12 = None + mul_13 = 0.25 * factor_att_13 + factor_att_13 = None + x_108 = mul_13 + EV_hat_13 + mul_13 = EV_hat_13 = None + transpose_43 = x_108.transpose(1, 2) + x_108 = None + x_109 = transpose_43.reshape(1, 785, 128) + transpose_43 = None + x_110 = torch._C._nn.linear( + x_109, + l_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_109 = l_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_111 = torch.nn.functional.dropout(x_110, 0.0, False, False) + x_110 = None + x_112 = x_106 + x_111 + x_106 = x_111 = None + x_113 = torch.nn.functional.layer_norm( + x_112, + (128,), + l_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_bias_ + ) = None + x_114 = torch._C._nn.linear( + x_113, + l_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_113 = l_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_115 = torch._C._nn.gelu(x_114, approximate="none") + x_114 = None + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + x_117 = torch._C._nn.linear( + x_116, + l_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_116 = l_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = x_112 + x_118 + x_112 = x_118 = None + getitem_79 = x_119[ + (slice(None, None, None), slice(1, None, None), slice(None, None, None)) + ] + x_119 = None + reshape_29 = getitem_79.reshape(1, 28, 28, -1) + getitem_79 = None + permute_8 = reshape_29.permute(0, 3, 1, 2) + reshape_29 = None + x2_nocls = permute_8.contiguous() + permute_8 = None + x_120 = torch.conv2d( + x2_nocls, + l_self_modules_patch_embed3_modules_proj_parameters_weight_, + l_self_modules_patch_embed3_modules_proj_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x2_nocls = ( + l_self_modules_patch_embed3_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed3_modules_proj_parameters_bias_ = None + flatten_9 = x_120.flatten(2) + x_120 = None + x_121 = flatten_9.transpose(1, 2) + flatten_9 = None + x_122 = torch.nn.functional.layer_norm( + x_121, + (320,), + l_self_modules_patch_embed3_modules_norm_parameters_weight_, + l_self_modules_patch_embed3_modules_norm_parameters_bias_, + 1e-05, + ) + x_121 = ( + l_self_modules_patch_embed3_modules_norm_parameters_weight_ + ) = l_self_modules_patch_embed3_modules_norm_parameters_bias_ = None + cls_tokens_2 = l_self_parameters_cls_token3_.expand(1, -1, -1) + l_self_parameters_cls_token3_ = None + x_123 = torch.cat((cls_tokens_2, x_122), dim=1) + cls_tokens_2 = x_122 = None + cls_token_7 = x_123[(slice(None, None, None), slice(None, 1, None))] + img_tokens_7 = x_123[(slice(None, None, None), slice(1, None, None))] + x_123 = None + transpose_45 = img_tokens_7.transpose(1, 2) + img_tokens_7 = None + feat_7 = transpose_45.view(1, 320, 14, 14) + transpose_45 = None + conv2d_31 = torch.conv2d( + feat_7, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 320, + ) + x_124 = conv2d_31 + feat_7 + conv2d_31 = feat_7 = None + flatten_10 = x_124.flatten(2) + x_124 = None + x_125 = flatten_10.transpose(1, 2) + flatten_10 = None + x_126 = torch.cat((cls_token_7, x_125), dim=1) + cls_token_7 = x_125 = None + x_127 = torch.nn.functional.layer_norm( + x_126, + (320,), + l_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_127, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_127 = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_30 = linear_28.reshape(1, 197, 3, 8, 40) + linear_28 = None + qkv_7 = reshape_30.permute(2, 0, 3, 1, 4) + reshape_30 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + k_softmax_7 = k_7.softmax(dim=2) + k_7 = None + transpose_47 = k_softmax_7.transpose(-1, -2) + k_softmax_7 = None + factor_att_14 = transpose_47 @ v_7 + transpose_47 = None + factor_att_15 = q_7 @ factor_att_14 + factor_att_14 = None + q_img_7 = q_7[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_7 = None + v_img_14 = v_7[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_7 = None + transpose_48 = v_img_14.transpose(-1, -2) + v_img_14 = None + v_img_15 = transpose_48.reshape(1, 320, 14, 14) + transpose_48 = None + split_7 = torch.functional.split(v_img_15, [80, 120, 120], dim=1) + v_img_15 = None + getitem_87 = split_7[0] + getitem_88 = split_7[1] + getitem_89 = split_7[2] + split_7 = None + conv2d_32 = torch.conv2d( + getitem_87, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 80, + ) + getitem_87 = None + conv2d_33 = torch.conv2d( + getitem_88, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 120, + ) + getitem_88 = None + conv2d_34 = torch.conv2d( + getitem_89, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 120, + ) + getitem_89 = None + conv_v_img_14 = torch.cat([conv2d_32, conv2d_33, conv2d_34], dim=1) + conv2d_32 = conv2d_33 = conv2d_34 = None + reshape_32 = conv_v_img_14.reshape(1, 8, 40, 196) + conv_v_img_14 = None + conv_v_img_15 = reshape_32.transpose(-1, -2) + reshape_32 = None + EV_hat_14 = q_img_7 * conv_v_img_15 + q_img_7 = conv_v_img_15 = None + EV_hat_15 = torch._C._nn.pad(EV_hat_14, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_14 = None + mul_15 = 0.15811388300841897 * factor_att_15 + factor_att_15 = None + x_128 = mul_15 + EV_hat_15 + mul_15 = EV_hat_15 = None + transpose_50 = x_128.transpose(1, 2) + x_128 = None + x_129 = transpose_50.reshape(1, 197, 320) + transpose_50 = None + x_130 = torch._C._nn.linear( + x_129, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_129 = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = x_126 + x_131 + x_126 = x_131 = None + x_133 = torch.nn.functional.layer_norm( + x_132, + (320,), + l_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_ + ) = None + x_134 = torch._C._nn.linear( + x_133, + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_133 = l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_135 = torch._C._nn.gelu(x_134, approximate="none") + x_134 = None + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_132 + x_138 + x_132 = x_138 = None + cls_token_8 = x_139[(slice(None, None, None), slice(None, 1, None))] + img_tokens_8 = x_139[(slice(None, None, None), slice(1, None, None))] + x_139 = None + transpose_51 = img_tokens_8.transpose(1, 2) + img_tokens_8 = None + feat_8 = transpose_51.view(1, 320, 14, 14) + transpose_51 = None + conv2d_35 = torch.conv2d( + feat_8, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 320, + ) + x_140 = conv2d_35 + feat_8 + conv2d_35 = feat_8 = None + flatten_11 = x_140.flatten(2) + x_140 = None + x_141 = flatten_11.transpose(1, 2) + flatten_11 = None + x_142 = torch.cat((cls_token_8, x_141), dim=1) + cls_token_8 = x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (320,), + l_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_143, + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_143 = l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_34 = linear_32.reshape(1, 197, 3, 8, 40) + linear_32 = None + qkv_8 = reshape_34.permute(2, 0, 3, 1, 4) + reshape_34 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + k_softmax_8 = k_8.softmax(dim=2) + k_8 = None + transpose_53 = k_softmax_8.transpose(-1, -2) + k_softmax_8 = None + factor_att_16 = transpose_53 @ v_8 + transpose_53 = None + factor_att_17 = q_8 @ factor_att_16 + factor_att_16 = None + q_img_8 = q_8[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_8 = None + v_img_16 = v_8[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_8 = None + transpose_54 = v_img_16.transpose(-1, -2) + v_img_16 = None + v_img_17 = transpose_54.reshape(1, 320, 14, 14) + transpose_54 = None + split_8 = torch.functional.split(v_img_17, [80, 120, 120], dim=1) + v_img_17 = None + getitem_97 = split_8[0] + getitem_98 = split_8[1] + getitem_99 = split_8[2] + split_8 = None + conv2d_36 = torch.conv2d( + getitem_97, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 80, + ) + getitem_97 = None + conv2d_37 = torch.conv2d( + getitem_98, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 120, + ) + getitem_98 = None + conv2d_38 = torch.conv2d( + getitem_99, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 120, + ) + getitem_99 = None + conv_v_img_16 = torch.cat([conv2d_36, conv2d_37, conv2d_38], dim=1) + conv2d_36 = conv2d_37 = conv2d_38 = None + reshape_36 = conv_v_img_16.reshape(1, 8, 40, 196) + conv_v_img_16 = None + conv_v_img_17 = reshape_36.transpose(-1, -2) + reshape_36 = None + EV_hat_16 = q_img_8 * conv_v_img_17 + q_img_8 = conv_v_img_17 = None + EV_hat_17 = torch._C._nn.pad(EV_hat_16, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_16 = None + mul_17 = 0.15811388300841897 * factor_att_17 + factor_att_17 = None + x_144 = mul_17 + EV_hat_17 + mul_17 = EV_hat_17 = None + transpose_56 = x_144.transpose(1, 2) + x_144 = None + x_145 = transpose_56.reshape(1, 197, 320) + transpose_56 = None + x_146 = torch._C._nn.linear( + x_145, + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_145 = l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_147 = torch.nn.functional.dropout(x_146, 0.0, False, False) + x_146 = None + x_148 = x_142 + x_147 + x_142 = x_147 = None + x_149 = torch.nn.functional.layer_norm( + x_148, + (320,), + l_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_ + ) = None + x_150 = torch._C._nn.linear( + x_149, + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_149 = l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_151 = torch._C._nn.gelu(x_150, approximate="none") + x_150 = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + x_153 = torch._C._nn.linear( + x_152, + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_152 = l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_154 = torch.nn.functional.dropout(x_153, 0.0, False, False) + x_153 = None + x_155 = x_148 + x_154 + x_148 = x_154 = None + cls_token_9 = x_155[(slice(None, None, None), slice(None, 1, None))] + img_tokens_9 = x_155[(slice(None, None, None), slice(1, None, None))] + x_155 = None + transpose_57 = img_tokens_9.transpose(1, 2) + img_tokens_9 = None + feat_9 = transpose_57.view(1, 320, 14, 14) + transpose_57 = None + conv2d_39 = torch.conv2d( + feat_9, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 320, + ) + x_156 = conv2d_39 + feat_9 + conv2d_39 = feat_9 = None + flatten_12 = x_156.flatten(2) + x_156 = None + x_157 = flatten_12.transpose(1, 2) + flatten_12 = None + x_158 = torch.cat((cls_token_9, x_157), dim=1) + cls_token_9 = x_157 = None + x_159 = torch.nn.functional.layer_norm( + x_158, + (320,), + l_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_159, + l_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_159 = l_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_38 = linear_36.reshape(1, 197, 3, 8, 40) + linear_36 = None + qkv_9 = reshape_38.permute(2, 0, 3, 1, 4) + reshape_38 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + k_softmax_9 = k_9.softmax(dim=2) + k_9 = None + transpose_59 = k_softmax_9.transpose(-1, -2) + k_softmax_9 = None + factor_att_18 = transpose_59 @ v_9 + transpose_59 = None + factor_att_19 = q_9 @ factor_att_18 + factor_att_18 = None + q_img_9 = q_9[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_9 = None + v_img_18 = v_9[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_9 = None + transpose_60 = v_img_18.transpose(-1, -2) + v_img_18 = None + v_img_19 = transpose_60.reshape(1, 320, 14, 14) + transpose_60 = None + split_9 = torch.functional.split(v_img_19, [80, 120, 120], dim=1) + v_img_19 = None + getitem_107 = split_9[0] + getitem_108 = split_9[1] + getitem_109 = split_9[2] + split_9 = None + conv2d_40 = torch.conv2d( + getitem_107, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 80, + ) + getitem_107 = None + conv2d_41 = torch.conv2d( + getitem_108, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 120, + ) + getitem_108 = None + conv2d_42 = torch.conv2d( + getitem_109, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 120, + ) + getitem_109 = None + conv_v_img_18 = torch.cat([conv2d_40, conv2d_41, conv2d_42], dim=1) + conv2d_40 = conv2d_41 = conv2d_42 = None + reshape_40 = conv_v_img_18.reshape(1, 8, 40, 196) + conv_v_img_18 = None + conv_v_img_19 = reshape_40.transpose(-1, -2) + reshape_40 = None + EV_hat_18 = q_img_9 * conv_v_img_19 + q_img_9 = conv_v_img_19 = None + EV_hat_19 = torch._C._nn.pad(EV_hat_18, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_18 = None + mul_19 = 0.15811388300841897 * factor_att_19 + factor_att_19 = None + x_160 = mul_19 + EV_hat_19 + mul_19 = EV_hat_19 = None + transpose_62 = x_160.transpose(1, 2) + x_160 = None + x_161 = transpose_62.reshape(1, 197, 320) + transpose_62 = None + x_162 = torch._C._nn.linear( + x_161, + l_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_161 = l_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_163 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + x_164 = x_158 + x_163 + x_158 = x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (320,), + l_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_bias_ + ) = None + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166, approximate="none") + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_169 = torch._C._nn.linear( + x_168, + l_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_168 = l_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_170 = torch.nn.functional.dropout(x_169, 0.0, False, False) + x_169 = None + x_171 = x_164 + x_170 + x_164 = x_170 = None + cls_token_10 = x_171[(slice(None, None, None), slice(None, 1, None))] + img_tokens_10 = x_171[(slice(None, None, None), slice(1, None, None))] + x_171 = None + transpose_63 = img_tokens_10.transpose(1, 2) + img_tokens_10 = None + feat_10 = transpose_63.view(1, 320, 14, 14) + transpose_63 = None + conv2d_43 = torch.conv2d( + feat_10, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 320, + ) + x_172 = conv2d_43 + feat_10 + conv2d_43 = feat_10 = None + flatten_13 = x_172.flatten(2) + x_172 = None + x_173 = flatten_13.transpose(1, 2) + flatten_13 = None + x_174 = torch.cat((cls_token_10, x_173), dim=1) + cls_token_10 = x_173 = None + x_175 = torch.nn.functional.layer_norm( + x_174, + (320,), + l_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_175, + l_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_175 = l_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_42 = linear_40.reshape(1, 197, 3, 8, 40) + linear_40 = None + qkv_10 = reshape_42.permute(2, 0, 3, 1, 4) + reshape_42 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + k_softmax_10 = k_10.softmax(dim=2) + k_10 = None + transpose_65 = k_softmax_10.transpose(-1, -2) + k_softmax_10 = None + factor_att_20 = transpose_65 @ v_10 + transpose_65 = None + factor_att_21 = q_10 @ factor_att_20 + factor_att_20 = None + q_img_10 = q_10[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_10 = None + v_img_20 = v_10[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_10 = None + transpose_66 = v_img_20.transpose(-1, -2) + v_img_20 = None + v_img_21 = transpose_66.reshape(1, 320, 14, 14) + transpose_66 = None + split_10 = torch.functional.split(v_img_21, [80, 120, 120], dim=1) + v_img_21 = None + getitem_117 = split_10[0] + getitem_118 = split_10[1] + getitem_119 = split_10[2] + split_10 = None + conv2d_44 = torch.conv2d( + getitem_117, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 80, + ) + getitem_117 = None + conv2d_45 = torch.conv2d( + getitem_118, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 120, + ) + getitem_118 = None + conv2d_46 = torch.conv2d( + getitem_119, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 120, + ) + getitem_119 = None + conv_v_img_20 = torch.cat([conv2d_44, conv2d_45, conv2d_46], dim=1) + conv2d_44 = conv2d_45 = conv2d_46 = None + reshape_44 = conv_v_img_20.reshape(1, 8, 40, 196) + conv_v_img_20 = None + conv_v_img_21 = reshape_44.transpose(-1, -2) + reshape_44 = None + EV_hat_20 = q_img_10 * conv_v_img_21 + q_img_10 = conv_v_img_21 = None + EV_hat_21 = torch._C._nn.pad(EV_hat_20, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_20 = None + mul_21 = 0.15811388300841897 * factor_att_21 + factor_att_21 = None + x_176 = mul_21 + EV_hat_21 + mul_21 = EV_hat_21 = None + transpose_68 = x_176.transpose(1, 2) + x_176 = None + x_177 = transpose_68.reshape(1, 197, 320) + transpose_68 = None + x_178 = torch._C._nn.linear( + x_177, + l_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_177 = l_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_180 = x_174 + x_179 + x_174 = x_179 = None + x_181 = torch.nn.functional.layer_norm( + x_180, + (320,), + l_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_bias_ + ) = None + x_182 = torch._C._nn.linear( + x_181, + l_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_181 = l_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_183 = torch._C._nn.gelu(x_182, approximate="none") + x_182 = None + x_184 = torch.nn.functional.dropout(x_183, 0.0, False, False) + x_183 = None + x_185 = torch._C._nn.linear( + x_184, + l_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_184 = l_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_186 = torch.nn.functional.dropout(x_185, 0.0, False, False) + x_185 = None + x_187 = x_180 + x_186 + x_180 = x_186 = None + cls_token_11 = x_187[(slice(None, None, None), slice(None, 1, None))] + img_tokens_11 = x_187[(slice(None, None, None), slice(1, None, None))] + x_187 = None + transpose_69 = img_tokens_11.transpose(1, 2) + img_tokens_11 = None + feat_11 = transpose_69.view(1, 320, 14, 14) + transpose_69 = None + conv2d_47 = torch.conv2d( + feat_11, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 320, + ) + x_188 = conv2d_47 + feat_11 + conv2d_47 = feat_11 = None + flatten_14 = x_188.flatten(2) + x_188 = None + x_189 = flatten_14.transpose(1, 2) + flatten_14 = None + x_190 = torch.cat((cls_token_11, x_189), dim=1) + cls_token_11 = x_189 = None + x_191 = torch.nn.functional.layer_norm( + x_190, + (320,), + l_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_191, + l_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_191 = l_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_46 = linear_44.reshape(1, 197, 3, 8, 40) + linear_44 = None + qkv_11 = reshape_46.permute(2, 0, 3, 1, 4) + reshape_46 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + k_softmax_11 = k_11.softmax(dim=2) + k_11 = None + transpose_71 = k_softmax_11.transpose(-1, -2) + k_softmax_11 = None + factor_att_22 = transpose_71 @ v_11 + transpose_71 = None + factor_att_23 = q_11 @ factor_att_22 + factor_att_22 = None + q_img_11 = q_11[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_11 = None + v_img_22 = v_11[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_11 = None + transpose_72 = v_img_22.transpose(-1, -2) + v_img_22 = None + v_img_23 = transpose_72.reshape(1, 320, 14, 14) + transpose_72 = None + split_11 = torch.functional.split(v_img_23, [80, 120, 120], dim=1) + v_img_23 = None + getitem_127 = split_11[0] + getitem_128 = split_11[1] + getitem_129 = split_11[2] + split_11 = None + conv2d_48 = torch.conv2d( + getitem_127, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 80, + ) + getitem_127 = None + conv2d_49 = torch.conv2d( + getitem_128, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 120, + ) + getitem_128 = None + conv2d_50 = torch.conv2d( + getitem_129, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 120, + ) + getitem_129 = None + conv_v_img_22 = torch.cat([conv2d_48, conv2d_49, conv2d_50], dim=1) + conv2d_48 = conv2d_49 = conv2d_50 = None + reshape_48 = conv_v_img_22.reshape(1, 8, 40, 196) + conv_v_img_22 = None + conv_v_img_23 = reshape_48.transpose(-1, -2) + reshape_48 = None + EV_hat_22 = q_img_11 * conv_v_img_23 + q_img_11 = conv_v_img_23 = None + EV_hat_23 = torch._C._nn.pad(EV_hat_22, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_22 = None + mul_23 = 0.15811388300841897 * factor_att_23 + factor_att_23 = None + x_192 = mul_23 + EV_hat_23 + mul_23 = EV_hat_23 = None + transpose_74 = x_192.transpose(1, 2) + x_192 = None + x_193 = transpose_74.reshape(1, 197, 320) + transpose_74 = None + x_194 = torch._C._nn.linear( + x_193, + l_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_193 = l_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_195 = torch.nn.functional.dropout(x_194, 0.0, False, False) + x_194 = None + x_196 = x_190 + x_195 + x_190 = x_195 = None + x_197 = torch.nn.functional.layer_norm( + x_196, + (320,), + l_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_bias_ + ) = None + x_198 = torch._C._nn.linear( + x_197, + l_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_197 = l_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_199 = torch._C._nn.gelu(x_198, approximate="none") + x_198 = None + x_200 = torch.nn.functional.dropout(x_199, 0.0, False, False) + x_199 = None + x_201 = torch._C._nn.linear( + x_200, + l_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_200 = l_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_202 = torch.nn.functional.dropout(x_201, 0.0, False, False) + x_201 = None + x_203 = x_196 + x_202 + x_196 = x_202 = None + cls_token_12 = x_203[(slice(None, None, None), slice(None, 1, None))] + img_tokens_12 = x_203[(slice(None, None, None), slice(1, None, None))] + x_203 = None + transpose_75 = img_tokens_12.transpose(1, 2) + img_tokens_12 = None + feat_12 = transpose_75.view(1, 320, 14, 14) + transpose_75 = None + conv2d_51 = torch.conv2d( + feat_12, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 320, + ) + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_ = (None) + x_204 = conv2d_51 + feat_12 + conv2d_51 = feat_12 = None + flatten_15 = x_204.flatten(2) + x_204 = None + x_205 = flatten_15.transpose(1, 2) + flatten_15 = None + x_206 = torch.cat((cls_token_12, x_205), dim=1) + cls_token_12 = x_205 = None + x_207 = torch.nn.functional.layer_norm( + x_206, + (320,), + l_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_48 = torch._C._nn.linear( + x_207, + l_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_207 = l_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_50 = linear_48.reshape(1, 197, 3, 8, 40) + linear_48 = None + qkv_12 = reshape_50.permute(2, 0, 3, 1, 4) + reshape_50 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_12 = unbind_12[0] + k_12 = unbind_12[1] + v_12 = unbind_12[2] + unbind_12 = None + k_softmax_12 = k_12.softmax(dim=2) + k_12 = None + transpose_77 = k_softmax_12.transpose(-1, -2) + k_softmax_12 = None + factor_att_24 = transpose_77 @ v_12 + transpose_77 = None + factor_att_25 = q_12 @ factor_att_24 + factor_att_24 = None + q_img_12 = q_12[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_12 = None + v_img_24 = v_12[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_12 = None + transpose_78 = v_img_24.transpose(-1, -2) + v_img_24 = None + v_img_25 = transpose_78.reshape(1, 320, 14, 14) + transpose_78 = None + split_12 = torch.functional.split(v_img_25, [80, 120, 120], dim=1) + v_img_25 = None + getitem_137 = split_12[0] + getitem_138 = split_12[1] + getitem_139 = split_12[2] + split_12 = None + conv2d_52 = torch.conv2d( + getitem_137, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 80, + ) + getitem_137 = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = (None) + conv2d_53 = torch.conv2d( + getitem_138, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 120, + ) + getitem_138 = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = (None) + conv2d_54 = torch.conv2d( + getitem_139, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 120, + ) + getitem_139 = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = (None) + conv_v_img_24 = torch.cat([conv2d_52, conv2d_53, conv2d_54], dim=1) + conv2d_52 = conv2d_53 = conv2d_54 = None + reshape_52 = conv_v_img_24.reshape(1, 8, 40, 196) + conv_v_img_24 = None + conv_v_img_25 = reshape_52.transpose(-1, -2) + reshape_52 = None + EV_hat_24 = q_img_12 * conv_v_img_25 + q_img_12 = conv_v_img_25 = None + EV_hat_25 = torch._C._nn.pad(EV_hat_24, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_24 = None + mul_25 = 0.15811388300841897 * factor_att_25 + factor_att_25 = None + x_208 = mul_25 + EV_hat_25 + mul_25 = EV_hat_25 = None + transpose_80 = x_208.transpose(1, 2) + x_208 = None + x_209 = transpose_80.reshape(1, 197, 320) + transpose_80 = None + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_209 = l_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_211 = torch.nn.functional.dropout(x_210, 0.0, False, False) + x_210 = None + x_212 = x_206 + x_211 + x_206 = x_211 = None + x_213 = torch.nn.functional.layer_norm( + x_212, + (320,), + l_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_bias_ + ) = None + x_214 = torch._C._nn.linear( + x_213, + l_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_213 = l_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_215 = torch._C._nn.gelu(x_214, approximate="none") + x_214 = None + x_216 = torch.nn.functional.dropout(x_215, 0.0, False, False) + x_215 = None + x_217 = torch._C._nn.linear( + x_216, + l_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_216 = l_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_218 = torch.nn.functional.dropout(x_217, 0.0, False, False) + x_217 = None + x_219 = x_212 + x_218 + x_212 = x_218 = None + getitem_140 = x_219[ + (slice(None, None, None), slice(1, None, None), slice(None, None, None)) + ] + x_219 = None + reshape_54 = getitem_140.reshape(1, 14, 14, -1) + getitem_140 = None + permute_15 = reshape_54.permute(0, 3, 1, 2) + reshape_54 = None + x3_nocls = permute_15.contiguous() + permute_15 = None + x_220 = torch.conv2d( + x3_nocls, + l_self_modules_patch_embed4_modules_proj_parameters_weight_, + l_self_modules_patch_embed4_modules_proj_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x3_nocls = ( + l_self_modules_patch_embed4_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed4_modules_proj_parameters_bias_ = None + flatten_16 = x_220.flatten(2) + x_220 = None + x_221 = flatten_16.transpose(1, 2) + flatten_16 = None + x_222 = torch.nn.functional.layer_norm( + x_221, + (512,), + l_self_modules_patch_embed4_modules_norm_parameters_weight_, + l_self_modules_patch_embed4_modules_norm_parameters_bias_, + 1e-05, + ) + x_221 = ( + l_self_modules_patch_embed4_modules_norm_parameters_weight_ + ) = l_self_modules_patch_embed4_modules_norm_parameters_bias_ = None + cls_tokens_3 = l_self_parameters_cls_token4_.expand(1, -1, -1) + l_self_parameters_cls_token4_ = None + x_223 = torch.cat((cls_tokens_3, x_222), dim=1) + cls_tokens_3 = x_222 = None + cls_token_13 = x_223[(slice(None, None, None), slice(None, 1, None))] + img_tokens_13 = x_223[(slice(None, None, None), slice(1, None, None))] + x_223 = None + transpose_82 = img_tokens_13.transpose(1, 2) + img_tokens_13 = None + feat_13 = transpose_82.view(1, 512, 7, 7) + transpose_82 = None + conv2d_56 = torch.conv2d( + feat_13, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 512, + ) + x_224 = conv2d_56 + feat_13 + conv2d_56 = feat_13 = None + flatten_17 = x_224.flatten(2) + x_224 = None + x_225 = flatten_17.transpose(1, 2) + flatten_17 = None + x_226 = torch.cat((cls_token_13, x_225), dim=1) + cls_token_13 = x_225 = None + x_227 = torch.nn.functional.layer_norm( + x_226, + (512,), + l_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_ + ) = None + linear_52 = torch._C._nn.linear( + x_227, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_227 = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_55 = linear_52.reshape(1, 50, 3, 8, 64) + linear_52 = None + qkv_13 = reshape_55.permute(2, 0, 3, 1, 4) + reshape_55 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_13 = unbind_13[0] + k_13 = unbind_13[1] + v_13 = unbind_13[2] + unbind_13 = None + k_softmax_13 = k_13.softmax(dim=2) + k_13 = None + transpose_84 = k_softmax_13.transpose(-1, -2) + k_softmax_13 = None + factor_att_26 = transpose_84 @ v_13 + transpose_84 = None + factor_att_27 = q_13 @ factor_att_26 + factor_att_26 = None + q_img_13 = q_13[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_13 = None + v_img_26 = v_13[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_13 = None + transpose_85 = v_img_26.transpose(-1, -2) + v_img_26 = None + v_img_27 = transpose_85.reshape(1, 512, 7, 7) + transpose_85 = None + split_13 = torch.functional.split(v_img_27, [128, 192, 192], dim=1) + v_img_27 = None + getitem_148 = split_13[0] + getitem_149 = split_13[1] + getitem_150 = split_13[2] + split_13 = None + conv2d_57 = torch.conv2d( + getitem_148, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 128, + ) + getitem_148 = None + conv2d_58 = torch.conv2d( + getitem_149, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 192, + ) + getitem_149 = None + conv2d_59 = torch.conv2d( + getitem_150, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + getitem_150 = None + conv_v_img_26 = torch.cat([conv2d_57, conv2d_58, conv2d_59], dim=1) + conv2d_57 = conv2d_58 = conv2d_59 = None + reshape_57 = conv_v_img_26.reshape(1, 8, 64, 49) + conv_v_img_26 = None + conv_v_img_27 = reshape_57.transpose(-1, -2) + reshape_57 = None + EV_hat_26 = q_img_13 * conv_v_img_27 + q_img_13 = conv_v_img_27 = None + EV_hat_27 = torch._C._nn.pad(EV_hat_26, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_26 = None + mul_27 = 0.125 * factor_att_27 + factor_att_27 = None + x_228 = mul_27 + EV_hat_27 + mul_27 = EV_hat_27 = None + transpose_87 = x_228.transpose(1, 2) + x_228 = None + x_229 = transpose_87.reshape(1, 50, 512) + transpose_87 = None + x_230 = torch._C._nn.linear( + x_229, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_229 = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_231 = torch.nn.functional.dropout(x_230, 0.0, False, False) + x_230 = None + x_232 = x_226 + x_231 + x_226 = x_231 = None + x_233 = torch.nn.functional.layer_norm( + x_232, + (512,), + l_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_ + ) = None + x_234 = torch._C._nn.linear( + x_233, + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_233 = l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_235 = torch._C._nn.gelu(x_234, approximate="none") + x_234 = None + x_236 = torch.nn.functional.dropout(x_235, 0.0, False, False) + x_235 = None + x_237 = torch._C._nn.linear( + x_236, + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_236 = l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_238 = torch.nn.functional.dropout(x_237, 0.0, False, False) + x_237 = None + x_239 = x_232 + x_238 + x_232 = x_238 = None + cls_token_14 = x_239[(slice(None, None, None), slice(None, 1, None))] + img_tokens_14 = x_239[(slice(None, None, None), slice(1, None, None))] + x_239 = None + transpose_88 = img_tokens_14.transpose(1, 2) + img_tokens_14 = None + feat_14 = transpose_88.view(1, 512, 7, 7) + transpose_88 = None + conv2d_60 = torch.conv2d( + feat_14, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 512, + ) + x_240 = conv2d_60 + feat_14 + conv2d_60 = feat_14 = None + flatten_18 = x_240.flatten(2) + x_240 = None + x_241 = flatten_18.transpose(1, 2) + flatten_18 = None + x_242 = torch.cat((cls_token_14, x_241), dim=1) + cls_token_14 = x_241 = None + x_243 = torch.nn.functional.layer_norm( + x_242, + (512,), + l_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_56 = torch._C._nn.linear( + x_243, + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_243 = l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_59 = linear_56.reshape(1, 50, 3, 8, 64) + linear_56 = None + qkv_14 = reshape_59.permute(2, 0, 3, 1, 4) + reshape_59 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_14 = unbind_14[0] + k_14 = unbind_14[1] + v_14 = unbind_14[2] + unbind_14 = None + k_softmax_14 = k_14.softmax(dim=2) + k_14 = None + transpose_90 = k_softmax_14.transpose(-1, -2) + k_softmax_14 = None + factor_att_28 = transpose_90 @ v_14 + transpose_90 = None + factor_att_29 = q_14 @ factor_att_28 + factor_att_28 = None + q_img_14 = q_14[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_14 = None + v_img_28 = v_14[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_14 = None + transpose_91 = v_img_28.transpose(-1, -2) + v_img_28 = None + v_img_29 = transpose_91.reshape(1, 512, 7, 7) + transpose_91 = None + split_14 = torch.functional.split(v_img_29, [128, 192, 192], dim=1) + v_img_29 = None + getitem_158 = split_14[0] + getitem_159 = split_14[1] + getitem_160 = split_14[2] + split_14 = None + conv2d_61 = torch.conv2d( + getitem_158, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 128, + ) + getitem_158 = None + conv2d_62 = torch.conv2d( + getitem_159, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 192, + ) + getitem_159 = None + conv2d_63 = torch.conv2d( + getitem_160, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + getitem_160 = None + conv_v_img_28 = torch.cat([conv2d_61, conv2d_62, conv2d_63], dim=1) + conv2d_61 = conv2d_62 = conv2d_63 = None + reshape_61 = conv_v_img_28.reshape(1, 8, 64, 49) + conv_v_img_28 = None + conv_v_img_29 = reshape_61.transpose(-1, -2) + reshape_61 = None + EV_hat_28 = q_img_14 * conv_v_img_29 + q_img_14 = conv_v_img_29 = None + EV_hat_29 = torch._C._nn.pad(EV_hat_28, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_28 = None + mul_29 = 0.125 * factor_att_29 + factor_att_29 = None + x_244 = mul_29 + EV_hat_29 + mul_29 = EV_hat_29 = None + transpose_93 = x_244.transpose(1, 2) + x_244 = None + x_245 = transpose_93.reshape(1, 50, 512) + transpose_93 = None + x_246 = torch._C._nn.linear( + x_245, + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_245 = l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_247 = torch.nn.functional.dropout(x_246, 0.0, False, False) + x_246 = None + x_248 = x_242 + x_247 + x_242 = x_247 = None + x_249 = torch.nn.functional.layer_norm( + x_248, + (512,), + l_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_ + ) = None + x_250 = torch._C._nn.linear( + x_249, + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_249 = l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_251 = torch._C._nn.gelu(x_250, approximate="none") + x_250 = None + x_252 = torch.nn.functional.dropout(x_251, 0.0, False, False) + x_251 = None + x_253 = torch._C._nn.linear( + x_252, + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_252 = l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_254 = torch.nn.functional.dropout(x_253, 0.0, False, False) + x_253 = None + x_255 = x_248 + x_254 + x_248 = x_254 = None + cls_token_15 = x_255[(slice(None, None, None), slice(None, 1, None))] + img_tokens_15 = x_255[(slice(None, None, None), slice(1, None, None))] + x_255 = None + transpose_94 = img_tokens_15.transpose(1, 2) + img_tokens_15 = None + feat_15 = transpose_94.view(1, 512, 7, 7) + transpose_94 = None + conv2d_64 = torch.conv2d( + feat_15, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 512, + ) + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_ = (None) + x_256 = conv2d_64 + feat_15 + conv2d_64 = feat_15 = None + flatten_19 = x_256.flatten(2) + x_256 = None + x_257 = flatten_19.transpose(1, 2) + flatten_19 = None + x_258 = torch.cat((cls_token_15, x_257), dim=1) + cls_token_15 = x_257 = None + x_259 = torch.nn.functional.layer_norm( + x_258, + (512,), + l_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_60 = torch._C._nn.linear( + x_259, + l_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_259 = l_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_63 = linear_60.reshape(1, 50, 3, 8, 64) + linear_60 = None + qkv_15 = reshape_63.permute(2, 0, 3, 1, 4) + reshape_63 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_15 = unbind_15[0] + k_15 = unbind_15[1] + v_15 = unbind_15[2] + unbind_15 = None + k_softmax_15 = k_15.softmax(dim=2) + k_15 = None + transpose_96 = k_softmax_15.transpose(-1, -2) + k_softmax_15 = None + factor_att_30 = transpose_96 @ v_15 + transpose_96 = None + factor_att_31 = q_15 @ factor_att_30 + factor_att_30 = None + q_img_15 = q_15[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_15 = None + v_img_30 = v_15[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_15 = None + transpose_97 = v_img_30.transpose(-1, -2) + v_img_30 = None + v_img_31 = transpose_97.reshape(1, 512, 7, 7) + transpose_97 = None + split_15 = torch.functional.split(v_img_31, [128, 192, 192], dim=1) + v_img_31 = None + getitem_168 = split_15[0] + getitem_169 = split_15[1] + getitem_170 = split_15[2] + split_15 = None + conv2d_65 = torch.conv2d( + getitem_168, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 128, + ) + getitem_168 = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = (None) + conv2d_66 = torch.conv2d( + getitem_169, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 192, + ) + getitem_169 = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = (None) + conv2d_67 = torch.conv2d( + getitem_170, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + getitem_170 = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = (None) + conv_v_img_30 = torch.cat([conv2d_65, conv2d_66, conv2d_67], dim=1) + conv2d_65 = conv2d_66 = conv2d_67 = None + reshape_65 = conv_v_img_30.reshape(1, 8, 64, 49) + conv_v_img_30 = None + conv_v_img_31 = reshape_65.transpose(-1, -2) + reshape_65 = None + EV_hat_30 = q_img_15 * conv_v_img_31 + q_img_15 = conv_v_img_31 = None + EV_hat_31 = torch._C._nn.pad(EV_hat_30, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_30 = None + mul_31 = 0.125 * factor_att_31 + factor_att_31 = None + x_260 = mul_31 + EV_hat_31 + mul_31 = EV_hat_31 = None + transpose_99 = x_260.transpose(1, 2) + x_260 = None + x_261 = transpose_99.reshape(1, 50, 512) + transpose_99 = None + x_262 = torch._C._nn.linear( + x_261, + l_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_261 = l_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_263 = torch.nn.functional.dropout(x_262, 0.0, False, False) + x_262 = None + x_264 = x_258 + x_263 + x_258 = x_263 = None + x_265 = torch.nn.functional.layer_norm( + x_264, + (512,), + l_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_bias_ + ) = None + x_266 = torch._C._nn.linear( + x_265, + l_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_265 = l_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_267 = torch._C._nn.gelu(x_266, approximate="none") + x_266 = None + x_268 = torch.nn.functional.dropout(x_267, 0.0, False, False) + x_267 = None + x_269 = torch._C._nn.linear( + x_268, + l_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_268 = l_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_270 = torch.nn.functional.dropout(x_269, 0.0, False, False) + x_269 = None + x_271 = x_264 + x_270 + x_264 = x_270 = None + getitem_171 = x_271[ + (slice(None, None, None), slice(1, None, None), slice(None, None, None)) + ] + reshape_67 = getitem_171.reshape(1, 7, 7, -1) + getitem_171 = None + permute_19 = reshape_67.permute(0, 3, 1, 2) + reshape_67 = None + x4_nocls = permute_19.contiguous() + permute_19 = x4_nocls = None + x_272 = torch.nn.functional.layer_norm( + x_271, + (512,), + l_self_modules_norm4_parameters_weight_, + l_self_modules_norm4_parameters_bias_, + 1e-06, + ) + x_271 = ( + l_self_modules_norm4_parameters_weight_ + ) = l_self_modules_norm4_parameters_bias_ = None + x_273 = x_272[(slice(None, None, None), 0)] + x_272 = None + x_274 = torch.nn.functional.dropout(x_273, 0.0, False, False) + x_273 = None + x_275 = torch._C._nn.linear( + x_274, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_274 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_275,) diff --git a/samples/timm/coat_lite_small.in1k/weight_meta.py b/samples/timm/coat_lite_small.in1k/weight_meta.py new file mode 100644 index 0000000000..221e94d318 --- /dev/null +++ b/samples/timm/coat_lite_small.in1k/weight_meta.py @@ -0,0 +1,2488 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.225 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed1_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed1_modules_proj_parameters_weight_" + shape = [64, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.085 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed1_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed1_modules_proj_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed1_modules_norm_parameters_weight_: + name = "L_self_modules_patch_embed1_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed1_modules_norm_parameters_bias_: + name = "L_self_modules_patch_embed1_modules_norm_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token1_: + name = "L_self_parameters_cls_token1_" + shape = [1, 1, 64] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_" + shape = [64, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.007 + std = 0.186 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.205 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [192, 64] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_" + shape = [16, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.192 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_" + shape = [16] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.092 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_" + shape = [24, 1, 5, 5] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.007 + std = 0.118 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.031 + std = 0.108 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_" + shape = [24, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.005 + std = 0.095 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [64, 64] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [192, 64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [64, 64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 64] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_2_modules_norm1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [192, 64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [64, 64] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_2_modules_norm2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed2_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed2_modules_proj_parameters_weight_" + shape = [128, 64, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed2_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed2_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed2_modules_norm_parameters_weight_: + name = "L_self_modules_patch_embed2_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed2_modules_norm_parameters_bias_: + name = "L_self_modules_patch_embed2_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token2_: + name = "L_self_parameters_cls_token2_" + shape = [1, 1, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_" + shape = [128, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.007 + std = 0.196 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.199 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_" + shape = [32, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.012 + std = 0.190 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.078 + std = 0.164 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_" + shape = [48, 1, 5, 5] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.114 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.010 + std = 0.116 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_" + shape = [48, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.009 + std = 0.075 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_2_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_3_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed3_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed3_modules_proj_parameters_weight_" + shape = [320, 128, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed3_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed3_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed3_modules_norm_parameters_weight_: + name = "L_self_modules_patch_embed3_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed3_modules_norm_parameters_bias_: + name = "L_self_modules_patch_embed3_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token3_: + name = "L_self_parameters_cls_token3_" + shape = [1, 1, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_" + shape = [320, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.190 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.012 + std = 0.193 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_" + shape = [80, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.194 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.047 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_" + shape = [120, 1, 5, 5] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.115 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.116 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_" + shape = [120, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.077 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_2_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_2_modules_norm2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_3_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_3_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_3_modules_norm2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_4_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_4_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_4_modules_norm2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_5_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_5_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_5_modules_norm2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed4_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed4_modules_proj_parameters_weight_" + shape = [512, 320, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.016 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed4_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed4_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.016 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed4_modules_norm_parameters_weight_: + name = "L_self_modules_patch_embed4_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed4_modules_norm_parameters_bias_: + name = "L_self_modules_patch_embed4_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token4_: + name = "L_self_parameters_cls_token4_" + shape = [1, 1, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_" + shape = [512, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.193 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.013 + std = 0.194 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_" + shape = [128, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.190 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.051 + std = 0.194 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_" + shape = [192, 1, 5, 5] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.117 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.116 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.085 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_2_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_2_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_2_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm4_parameters_weight_: + name = "L_self_modules_norm4_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm4_parameters_bias_: + name = "L_self_modules_norm4_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/coat_lite_tiny.in1k/graph_hash.txt b/samples/timm/coat_lite_tiny.in1k/graph_hash.txt new file mode 100644 index 0000000000..2aed4edeea --- /dev/null +++ b/samples/timm/coat_lite_tiny.in1k/graph_hash.txt @@ -0,0 +1 @@ +0f835b2acd918ac86f94f98d8976832e84263a568562660d3fe87debf158b23d \ No newline at end of file diff --git a/samples/timm/coat_lite_tiny.in1k/graph_net.json b/samples/timm/coat_lite_tiny.in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/coat_lite_tiny.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/coat_lite_tiny.in1k/input_meta.py b/samples/timm/coat_lite_tiny.in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/coat_lite_tiny.in1k/input_tensor_constraints.py b/samples/timm/coat_lite_tiny.in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/coat_lite_tiny.in1k/model.py b/samples/timm/coat_lite_tiny.in1k/model.py new file mode 100644 index 0000000000..bfb2593034 --- /dev/null +++ b/samples/timm/coat_lite_tiny.in1k/model.py @@ -0,0 +1,1989 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed1_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed1_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token1_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed2_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed2_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token2_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed3_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed3_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token3_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed4_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed4_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token4_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm4_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm4_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed1_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed1_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed1_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed1_modules_proj_parameters_bias_ + ) + l_self_modules_patch_embed1_modules_norm_parameters_weight_ = ( + L_self_modules_patch_embed1_modules_norm_parameters_weight_ + ) + l_self_modules_patch_embed1_modules_norm_parameters_bias_ = ( + L_self_modules_patch_embed1_modules_norm_parameters_bias_ + ) + l_self_parameters_cls_token1_ = L_self_parameters_cls_token1_ + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_patch_embed2_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed2_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed2_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed2_modules_proj_parameters_bias_ + ) + l_self_modules_patch_embed2_modules_norm_parameters_weight_ = ( + L_self_modules_patch_embed2_modules_norm_parameters_weight_ + ) + l_self_modules_patch_embed2_modules_norm_parameters_bias_ = ( + L_self_modules_patch_embed2_modules_norm_parameters_bias_ + ) + l_self_parameters_cls_token2_ = L_self_parameters_cls_token2_ + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_patch_embed3_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed3_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed3_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed3_modules_proj_parameters_bias_ + ) + l_self_modules_patch_embed3_modules_norm_parameters_weight_ = ( + L_self_modules_patch_embed3_modules_norm_parameters_weight_ + ) + l_self_modules_patch_embed3_modules_norm_parameters_bias_ = ( + L_self_modules_patch_embed3_modules_norm_parameters_bias_ + ) + l_self_parameters_cls_token3_ = L_self_parameters_cls_token3_ + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_patch_embed4_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed4_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed4_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed4_modules_proj_parameters_bias_ + ) + l_self_modules_patch_embed4_modules_norm_parameters_weight_ = ( + L_self_modules_patch_embed4_modules_norm_parameters_weight_ + ) + l_self_modules_patch_embed4_modules_norm_parameters_bias_ = ( + L_self_modules_patch_embed4_modules_norm_parameters_bias_ + ) + l_self_parameters_cls_token4_ = L_self_parameters_cls_token4_ + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ + l_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_norm4_parameters_weight_ = ( + L_self_modules_norm4_parameters_weight_ + ) + l_self_modules_norm4_parameters_bias_ = L_self_modules_norm4_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed1_modules_proj_parameters_weight_, + l_self_modules_patch_embed1_modules_proj_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed1_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed1_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + x_2 = torch.nn.functional.layer_norm( + x_1, + (64,), + l_self_modules_patch_embed1_modules_norm_parameters_weight_, + l_self_modules_patch_embed1_modules_norm_parameters_bias_, + 1e-05, + ) + x_1 = ( + l_self_modules_patch_embed1_modules_norm_parameters_weight_ + ) = l_self_modules_patch_embed1_modules_norm_parameters_bias_ = None + cls_tokens = l_self_parameters_cls_token1_.expand(1, -1, -1) + l_self_parameters_cls_token1_ = None + x_3 = torch.cat((cls_tokens, x_2), dim=1) + cls_tokens = x_2 = None + cls_token = x_3[(slice(None, None, None), slice(None, 1, None))] + img_tokens = x_3[(slice(None, None, None), slice(1, None, None))] + x_3 = None + transpose_1 = img_tokens.transpose(1, 2) + img_tokens = None + feat = transpose_1.view(1, 64, 56, 56) + transpose_1 = None + conv2d_1 = torch.conv2d( + feat, + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 64, + ) + x_4 = conv2d_1 + feat + conv2d_1 = feat = None + flatten_1 = x_4.flatten(2) + x_4 = None + x_5 = flatten_1.transpose(1, 2) + flatten_1 = None + x_6 = torch.cat((cls_token, x_5), dim=1) + cls_token = x_5 = None + x_7 = torch.nn.functional.layer_norm( + x_6, + (64,), + l_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_7, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_7 = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 3137, 3, 8, 8) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + k_softmax = k.softmax(dim=2) + k = None + transpose_3 = k_softmax.transpose(-1, -2) + k_softmax = None + factor_att = transpose_3 @ v + transpose_3 = None + factor_att_1 = q @ factor_att + factor_att = None + q_img = q[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q = None + v_img = v[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v = None + transpose_4 = v_img.transpose(-1, -2) + v_img = None + v_img_1 = transpose_4.reshape(1, 64, 56, 56) + transpose_4 = None + split = torch.functional.split(v_img_1, [16, 24, 24], dim=1) + v_img_1 = None + getitem_15 = split[0] + getitem_16 = split[1] + getitem_17 = split[2] + split = None + conv2d_2 = torch.conv2d( + getitem_15, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 16, + ) + getitem_15 = None + conv2d_3 = torch.conv2d( + getitem_16, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 24, + ) + getitem_16 = None + conv2d_4 = torch.conv2d( + getitem_17, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 24, + ) + getitem_17 = None + conv_v_img = torch.cat([conv2d_2, conv2d_3, conv2d_4], dim=1) + conv2d_2 = conv2d_3 = conv2d_4 = None + reshape_2 = conv_v_img.reshape(1, 8, 8, 3136) + conv_v_img = None + conv_v_img_1 = reshape_2.transpose(-1, -2) + reshape_2 = None + EV_hat = q_img * conv_v_img_1 + q_img = conv_v_img_1 = None + EV_hat_1 = torch._C._nn.pad(EV_hat, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat = None + mul_1 = 0.3535533905932738 * factor_att_1 + factor_att_1 = None + x_8 = mul_1 + EV_hat_1 + mul_1 = EV_hat_1 = None + transpose_6 = x_8.transpose(1, 2) + x_8 = None + x_9 = transpose_6.reshape(1, 3137, 64) + transpose_6 = None + x_10 = torch._C._nn.linear( + x_9, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_9 = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = x_6 + x_11 + x_6 = x_11 = None + x_13 = torch.nn.functional.layer_norm( + x_12, + (64,), + l_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_ + ) = None + x_14 = torch._C._nn.linear( + x_13, + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_13 = l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_15 = torch._C._nn.gelu(x_14, approximate="none") + x_14 = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_16 = l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_18 = torch.nn.functional.dropout(x_17, 0.0, False, False) + x_17 = None + x_19 = x_12 + x_18 + x_12 = x_18 = None + cls_token_1 = x_19[(slice(None, None, None), slice(None, 1, None))] + img_tokens_1 = x_19[(slice(None, None, None), slice(1, None, None))] + x_19 = None + transpose_7 = img_tokens_1.transpose(1, 2) + img_tokens_1 = None + feat_1 = transpose_7.view(1, 64, 56, 56) + transpose_7 = None + conv2d_5 = torch.conv2d( + feat_1, + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 64, + ) + l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_ = (None) + x_20 = conv2d_5 + feat_1 + conv2d_5 = feat_1 = None + flatten_2 = x_20.flatten(2) + x_20 = None + x_21 = flatten_2.transpose(1, 2) + flatten_2 = None + x_22 = torch.cat((cls_token_1, x_21), dim=1) + cls_token_1 = x_21 = None + x_23 = torch.nn.functional.layer_norm( + x_22, + (64,), + l_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_23, + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_23 = l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_4.reshape(1, 3137, 3, 8, 8) + linear_4 = None + qkv_1 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + k_softmax_1 = k_1.softmax(dim=2) + k_1 = None + transpose_9 = k_softmax_1.transpose(-1, -2) + k_softmax_1 = None + factor_att_2 = transpose_9 @ v_1 + transpose_9 = None + factor_att_3 = q_1 @ factor_att_2 + factor_att_2 = None + q_img_1 = q_1[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_1 = None + v_img_2 = v_1[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_1 = None + transpose_10 = v_img_2.transpose(-1, -2) + v_img_2 = None + v_img_3 = transpose_10.reshape(1, 64, 56, 56) + transpose_10 = None + split_1 = torch.functional.split(v_img_3, [16, 24, 24], dim=1) + v_img_3 = None + getitem_25 = split_1[0] + getitem_26 = split_1[1] + getitem_27 = split_1[2] + split_1 = None + conv2d_6 = torch.conv2d( + getitem_25, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 16, + ) + getitem_25 = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = (None) + conv2d_7 = torch.conv2d( + getitem_26, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 24, + ) + getitem_26 = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = (None) + conv2d_8 = torch.conv2d( + getitem_27, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 24, + ) + getitem_27 = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = l_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = (None) + conv_v_img_2 = torch.cat([conv2d_6, conv2d_7, conv2d_8], dim=1) + conv2d_6 = conv2d_7 = conv2d_8 = None + reshape_6 = conv_v_img_2.reshape(1, 8, 8, 3136) + conv_v_img_2 = None + conv_v_img_3 = reshape_6.transpose(-1, -2) + reshape_6 = None + EV_hat_2 = q_img_1 * conv_v_img_3 + q_img_1 = conv_v_img_3 = None + EV_hat_3 = torch._C._nn.pad(EV_hat_2, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_2 = None + mul_3 = 0.3535533905932738 * factor_att_3 + factor_att_3 = None + x_24 = mul_3 + EV_hat_3 + mul_3 = EV_hat_3 = None + transpose_12 = x_24.transpose(1, 2) + x_24 = None + x_25 = transpose_12.reshape(1, 3137, 64) + transpose_12 = None + x_26 = torch._C._nn.linear( + x_25, + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_25 = l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = x_22 + x_27 + x_22 = x_27 = None + x_29 = torch.nn.functional.layer_norm( + x_28, + (64,), + l_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_ + ) = None + x_30 = torch._C._nn.linear( + x_29, + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_29 = l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_31 = torch._C._nn.gelu(x_30, approximate="none") + x_30 = None + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = torch._C._nn.linear( + x_32, + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_32 = l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = x_28 + x_34 + x_28 = x_34 = None + getitem_28 = x_35[ + (slice(None, None, None), slice(1, None, None), slice(None, None, None)) + ] + x_35 = None + reshape_8 = getitem_28.reshape(1, 56, 56, -1) + getitem_28 = None + permute_2 = reshape_8.permute(0, 3, 1, 2) + reshape_8 = None + x1_nocls = permute_2.contiguous() + permute_2 = None + x_36 = torch.conv2d( + x1_nocls, + l_self_modules_patch_embed2_modules_proj_parameters_weight_, + l_self_modules_patch_embed2_modules_proj_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x1_nocls = ( + l_self_modules_patch_embed2_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed2_modules_proj_parameters_bias_ = None + flatten_3 = x_36.flatten(2) + x_36 = None + x_37 = flatten_3.transpose(1, 2) + flatten_3 = None + x_38 = torch.nn.functional.layer_norm( + x_37, + (128,), + l_self_modules_patch_embed2_modules_norm_parameters_weight_, + l_self_modules_patch_embed2_modules_norm_parameters_bias_, + 1e-05, + ) + x_37 = ( + l_self_modules_patch_embed2_modules_norm_parameters_weight_ + ) = l_self_modules_patch_embed2_modules_norm_parameters_bias_ = None + cls_tokens_1 = l_self_parameters_cls_token2_.expand(1, -1, -1) + l_self_parameters_cls_token2_ = None + x_39 = torch.cat((cls_tokens_1, x_38), dim=1) + cls_tokens_1 = x_38 = None + cls_token_2 = x_39[(slice(None, None, None), slice(None, 1, None))] + img_tokens_2 = x_39[(slice(None, None, None), slice(1, None, None))] + x_39 = None + transpose_14 = img_tokens_2.transpose(1, 2) + img_tokens_2 = None + feat_2 = transpose_14.view(1, 128, 28, 28) + transpose_14 = None + conv2d_10 = torch.conv2d( + feat_2, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 128, + ) + x_40 = conv2d_10 + feat_2 + conv2d_10 = feat_2 = None + flatten_4 = x_40.flatten(2) + x_40 = None + x_41 = flatten_4.transpose(1, 2) + flatten_4 = None + x_42 = torch.cat((cls_token_2, x_41), dim=1) + cls_token_2 = x_41 = None + x_43 = torch.nn.functional.layer_norm( + x_42, + (128,), + l_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_43, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_43 = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_9 = linear_8.reshape(1, 785, 3, 8, 16) + linear_8 = None + qkv_2 = reshape_9.permute(2, 0, 3, 1, 4) + reshape_9 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + k_softmax_2 = k_2.softmax(dim=2) + k_2 = None + transpose_16 = k_softmax_2.transpose(-1, -2) + k_softmax_2 = None + factor_att_4 = transpose_16 @ v_2 + transpose_16 = None + factor_att_5 = q_2 @ factor_att_4 + factor_att_4 = None + q_img_2 = q_2[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_2 = None + v_img_4 = v_2[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_2 = None + transpose_17 = v_img_4.transpose(-1, -2) + v_img_4 = None + v_img_5 = transpose_17.reshape(1, 128, 28, 28) + transpose_17 = None + split_2 = torch.functional.split(v_img_5, [32, 48, 48], dim=1) + v_img_5 = None + getitem_36 = split_2[0] + getitem_37 = split_2[1] + getitem_38 = split_2[2] + split_2 = None + conv2d_11 = torch.conv2d( + getitem_36, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + getitem_36 = None + conv2d_12 = torch.conv2d( + getitem_37, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 48, + ) + getitem_37 = None + conv2d_13 = torch.conv2d( + getitem_38, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 48, + ) + getitem_38 = None + conv_v_img_4 = torch.cat([conv2d_11, conv2d_12, conv2d_13], dim=1) + conv2d_11 = conv2d_12 = conv2d_13 = None + reshape_11 = conv_v_img_4.reshape(1, 8, 16, 784) + conv_v_img_4 = None + conv_v_img_5 = reshape_11.transpose(-1, -2) + reshape_11 = None + EV_hat_4 = q_img_2 * conv_v_img_5 + q_img_2 = conv_v_img_5 = None + EV_hat_5 = torch._C._nn.pad(EV_hat_4, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_4 = None + mul_5 = 0.25 * factor_att_5 + factor_att_5 = None + x_44 = mul_5 + EV_hat_5 + mul_5 = EV_hat_5 = None + transpose_19 = x_44.transpose(1, 2) + x_44 = None + x_45 = transpose_19.reshape(1, 785, 128) + transpose_19 = None + x_46 = torch._C._nn.linear( + x_45, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_45 = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + x_48 = x_42 + x_47 + x_42 = x_47 = None + x_49 = torch.nn.functional.layer_norm( + x_48, + (128,), + l_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_ + ) = None + x_50 = torch._C._nn.linear( + x_49, + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_49 = l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_51 = torch._C._nn.gelu(x_50, approximate="none") + x_50 = None + x_52 = torch.nn.functional.dropout(x_51, 0.0, False, False) + x_51 = None + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_52 = l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_54 = torch.nn.functional.dropout(x_53, 0.0, False, False) + x_53 = None + x_55 = x_48 + x_54 + x_48 = x_54 = None + cls_token_3 = x_55[(slice(None, None, None), slice(None, 1, None))] + img_tokens_3 = x_55[(slice(None, None, None), slice(1, None, None))] + x_55 = None + transpose_20 = img_tokens_3.transpose(1, 2) + img_tokens_3 = None + feat_3 = transpose_20.view(1, 128, 28, 28) + transpose_20 = None + conv2d_14 = torch.conv2d( + feat_3, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 128, + ) + l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_ = (None) + x_56 = conv2d_14 + feat_3 + conv2d_14 = feat_3 = None + flatten_5 = x_56.flatten(2) + x_56 = None + x_57 = flatten_5.transpose(1, 2) + flatten_5 = None + x_58 = torch.cat((cls_token_3, x_57), dim=1) + cls_token_3 = x_57 = None + x_59 = torch.nn.functional.layer_norm( + x_58, + (128,), + l_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_59, + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_59 = l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_13 = linear_12.reshape(1, 785, 3, 8, 16) + linear_12 = None + qkv_3 = reshape_13.permute(2, 0, 3, 1, 4) + reshape_13 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + k_softmax_3 = k_3.softmax(dim=2) + k_3 = None + transpose_22 = k_softmax_3.transpose(-1, -2) + k_softmax_3 = None + factor_att_6 = transpose_22 @ v_3 + transpose_22 = None + factor_att_7 = q_3 @ factor_att_6 + factor_att_6 = None + q_img_3 = q_3[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_3 = None + v_img_6 = v_3[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_3 = None + transpose_23 = v_img_6.transpose(-1, -2) + v_img_6 = None + v_img_7 = transpose_23.reshape(1, 128, 28, 28) + transpose_23 = None + split_3 = torch.functional.split(v_img_7, [32, 48, 48], dim=1) + v_img_7 = None + getitem_46 = split_3[0] + getitem_47 = split_3[1] + getitem_48 = split_3[2] + split_3 = None + conv2d_15 = torch.conv2d( + getitem_46, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + getitem_46 = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = (None) + conv2d_16 = torch.conv2d( + getitem_47, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 48, + ) + getitem_47 = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = (None) + conv2d_17 = torch.conv2d( + getitem_48, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 48, + ) + getitem_48 = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = l_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = (None) + conv_v_img_6 = torch.cat([conv2d_15, conv2d_16, conv2d_17], dim=1) + conv2d_15 = conv2d_16 = conv2d_17 = None + reshape_15 = conv_v_img_6.reshape(1, 8, 16, 784) + conv_v_img_6 = None + conv_v_img_7 = reshape_15.transpose(-1, -2) + reshape_15 = None + EV_hat_6 = q_img_3 * conv_v_img_7 + q_img_3 = conv_v_img_7 = None + EV_hat_7 = torch._C._nn.pad(EV_hat_6, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_6 = None + mul_7 = 0.25 * factor_att_7 + factor_att_7 = None + x_60 = mul_7 + EV_hat_7 + mul_7 = EV_hat_7 = None + transpose_25 = x_60.transpose(1, 2) + x_60 = None + x_61 = transpose_25.reshape(1, 785, 128) + transpose_25 = None + x_62 = torch._C._nn.linear( + x_61, + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_61 = l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_63 = torch.nn.functional.dropout(x_62, 0.0, False, False) + x_62 = None + x_64 = x_58 + x_63 + x_58 = x_63 = None + x_65 = torch.nn.functional.layer_norm( + x_64, + (128,), + l_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_ + ) = None + x_66 = torch._C._nn.linear( + x_65, + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_65 = l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_67 = torch._C._nn.gelu(x_66, approximate="none") + x_66 = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = torch._C._nn.linear( + x_68, + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_68 = l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_71 = x_64 + x_70 + x_64 = x_70 = None + getitem_49 = x_71[ + (slice(None, None, None), slice(1, None, None), slice(None, None, None)) + ] + x_71 = None + reshape_17 = getitem_49.reshape(1, 28, 28, -1) + getitem_49 = None + permute_5 = reshape_17.permute(0, 3, 1, 2) + reshape_17 = None + x2_nocls = permute_5.contiguous() + permute_5 = None + x_72 = torch.conv2d( + x2_nocls, + l_self_modules_patch_embed3_modules_proj_parameters_weight_, + l_self_modules_patch_embed3_modules_proj_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x2_nocls = ( + l_self_modules_patch_embed3_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed3_modules_proj_parameters_bias_ = None + flatten_6 = x_72.flatten(2) + x_72 = None + x_73 = flatten_6.transpose(1, 2) + flatten_6 = None + x_74 = torch.nn.functional.layer_norm( + x_73, + (256,), + l_self_modules_patch_embed3_modules_norm_parameters_weight_, + l_self_modules_patch_embed3_modules_norm_parameters_bias_, + 1e-05, + ) + x_73 = ( + l_self_modules_patch_embed3_modules_norm_parameters_weight_ + ) = l_self_modules_patch_embed3_modules_norm_parameters_bias_ = None + cls_tokens_2 = l_self_parameters_cls_token3_.expand(1, -1, -1) + l_self_parameters_cls_token3_ = None + x_75 = torch.cat((cls_tokens_2, x_74), dim=1) + cls_tokens_2 = x_74 = None + cls_token_4 = x_75[(slice(None, None, None), slice(None, 1, None))] + img_tokens_4 = x_75[(slice(None, None, None), slice(1, None, None))] + x_75 = None + transpose_27 = img_tokens_4.transpose(1, 2) + img_tokens_4 = None + feat_4 = transpose_27.view(1, 256, 14, 14) + transpose_27 = None + conv2d_19 = torch.conv2d( + feat_4, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 256, + ) + x_76 = conv2d_19 + feat_4 + conv2d_19 = feat_4 = None + flatten_7 = x_76.flatten(2) + x_76 = None + x_77 = flatten_7.transpose(1, 2) + flatten_7 = None + x_78 = torch.cat((cls_token_4, x_77), dim=1) + cls_token_4 = x_77 = None + x_79 = torch.nn.functional.layer_norm( + x_78, + (256,), + l_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_79, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_79 = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_18 = linear_16.reshape(1, 197, 3, 8, 32) + linear_16 = None + qkv_4 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + k_softmax_4 = k_4.softmax(dim=2) + k_4 = None + transpose_29 = k_softmax_4.transpose(-1, -2) + k_softmax_4 = None + factor_att_8 = transpose_29 @ v_4 + transpose_29 = None + factor_att_9 = q_4 @ factor_att_8 + factor_att_8 = None + q_img_4 = q_4[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_4 = None + v_img_8 = v_4[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_4 = None + transpose_30 = v_img_8.transpose(-1, -2) + v_img_8 = None + v_img_9 = transpose_30.reshape(1, 256, 14, 14) + transpose_30 = None + split_4 = torch.functional.split(v_img_9, [64, 96, 96], dim=1) + v_img_9 = None + getitem_57 = split_4[0] + getitem_58 = split_4[1] + getitem_59 = split_4[2] + split_4 = None + conv2d_20 = torch.conv2d( + getitem_57, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 64, + ) + getitem_57 = None + conv2d_21 = torch.conv2d( + getitem_58, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 96, + ) + getitem_58 = None + conv2d_22 = torch.conv2d( + getitem_59, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + getitem_59 = None + conv_v_img_8 = torch.cat([conv2d_20, conv2d_21, conv2d_22], dim=1) + conv2d_20 = conv2d_21 = conv2d_22 = None + reshape_20 = conv_v_img_8.reshape(1, 8, 32, 196) + conv_v_img_8 = None + conv_v_img_9 = reshape_20.transpose(-1, -2) + reshape_20 = None + EV_hat_8 = q_img_4 * conv_v_img_9 + q_img_4 = conv_v_img_9 = None + EV_hat_9 = torch._C._nn.pad(EV_hat_8, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_8 = None + mul_9 = 0.1767766952966369 * factor_att_9 + factor_att_9 = None + x_80 = mul_9 + EV_hat_9 + mul_9 = EV_hat_9 = None + transpose_32 = x_80.transpose(1, 2) + x_80 = None + x_81 = transpose_32.reshape(1, 197, 256) + transpose_32 = None + x_82 = torch._C._nn.linear( + x_81, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_81 = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + x_84 = x_78 + x_83 + x_78 = x_83 = None + x_85 = torch.nn.functional.layer_norm( + x_84, + (256,), + l_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_ + ) = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_85 = l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_87 = torch._C._nn.gelu(x_86, approximate="none") + x_86 = None + x_88 = torch.nn.functional.dropout(x_87, 0.0, False, False) + x_87 = None + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_88 = l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_90 = torch.nn.functional.dropout(x_89, 0.0, False, False) + x_89 = None + x_91 = x_84 + x_90 + x_84 = x_90 = None + cls_token_5 = x_91[(slice(None, None, None), slice(None, 1, None))] + img_tokens_5 = x_91[(slice(None, None, None), slice(1, None, None))] + x_91 = None + transpose_33 = img_tokens_5.transpose(1, 2) + img_tokens_5 = None + feat_5 = transpose_33.view(1, 256, 14, 14) + transpose_33 = None + conv2d_23 = torch.conv2d( + feat_5, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 256, + ) + l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_ = (None) + x_92 = conv2d_23 + feat_5 + conv2d_23 = feat_5 = None + flatten_8 = x_92.flatten(2) + x_92 = None + x_93 = flatten_8.transpose(1, 2) + flatten_8 = None + x_94 = torch.cat((cls_token_5, x_93), dim=1) + cls_token_5 = x_93 = None + x_95 = torch.nn.functional.layer_norm( + x_94, + (256,), + l_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_95, + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_95 = l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_22 = linear_20.reshape(1, 197, 3, 8, 32) + linear_20 = None + qkv_5 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + k_softmax_5 = k_5.softmax(dim=2) + k_5 = None + transpose_35 = k_softmax_5.transpose(-1, -2) + k_softmax_5 = None + factor_att_10 = transpose_35 @ v_5 + transpose_35 = None + factor_att_11 = q_5 @ factor_att_10 + factor_att_10 = None + q_img_5 = q_5[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_5 = None + v_img_10 = v_5[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_5 = None + transpose_36 = v_img_10.transpose(-1, -2) + v_img_10 = None + v_img_11 = transpose_36.reshape(1, 256, 14, 14) + transpose_36 = None + split_5 = torch.functional.split(v_img_11, [64, 96, 96], dim=1) + v_img_11 = None + getitem_67 = split_5[0] + getitem_68 = split_5[1] + getitem_69 = split_5[2] + split_5 = None + conv2d_24 = torch.conv2d( + getitem_67, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 64, + ) + getitem_67 = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = (None) + conv2d_25 = torch.conv2d( + getitem_68, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 96, + ) + getitem_68 = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = (None) + conv2d_26 = torch.conv2d( + getitem_69, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + getitem_69 = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = l_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = (None) + conv_v_img_10 = torch.cat([conv2d_24, conv2d_25, conv2d_26], dim=1) + conv2d_24 = conv2d_25 = conv2d_26 = None + reshape_24 = conv_v_img_10.reshape(1, 8, 32, 196) + conv_v_img_10 = None + conv_v_img_11 = reshape_24.transpose(-1, -2) + reshape_24 = None + EV_hat_10 = q_img_5 * conv_v_img_11 + q_img_5 = conv_v_img_11 = None + EV_hat_11 = torch._C._nn.pad(EV_hat_10, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_10 = None + mul_11 = 0.1767766952966369 * factor_att_11 + factor_att_11 = None + x_96 = mul_11 + EV_hat_11 + mul_11 = EV_hat_11 = None + transpose_38 = x_96.transpose(1, 2) + x_96 = None + x_97 = transpose_38.reshape(1, 197, 256) + transpose_38 = None + x_98 = torch._C._nn.linear( + x_97, + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_97 = l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_99 = torch.nn.functional.dropout(x_98, 0.0, False, False) + x_98 = None + x_100 = x_94 + x_99 + x_94 = x_99 = None + x_101 = torch.nn.functional.layer_norm( + x_100, + (256,), + l_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_ + ) = None + x_102 = torch._C._nn.linear( + x_101, + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_101 = l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_103 = torch._C._nn.gelu(x_102, approximate="none") + x_102 = None + x_104 = torch.nn.functional.dropout(x_103, 0.0, False, False) + x_103 = None + x_105 = torch._C._nn.linear( + x_104, + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_104 = l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = x_100 + x_106 + x_100 = x_106 = None + getitem_70 = x_107[ + (slice(None, None, None), slice(1, None, None), slice(None, None, None)) + ] + x_107 = None + reshape_26 = getitem_70.reshape(1, 14, 14, -1) + getitem_70 = None + permute_8 = reshape_26.permute(0, 3, 1, 2) + reshape_26 = None + x3_nocls = permute_8.contiguous() + permute_8 = None + x_108 = torch.conv2d( + x3_nocls, + l_self_modules_patch_embed4_modules_proj_parameters_weight_, + l_self_modules_patch_embed4_modules_proj_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x3_nocls = ( + l_self_modules_patch_embed4_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed4_modules_proj_parameters_bias_ = None + flatten_9 = x_108.flatten(2) + x_108 = None + x_109 = flatten_9.transpose(1, 2) + flatten_9 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (320,), + l_self_modules_patch_embed4_modules_norm_parameters_weight_, + l_self_modules_patch_embed4_modules_norm_parameters_bias_, + 1e-05, + ) + x_109 = ( + l_self_modules_patch_embed4_modules_norm_parameters_weight_ + ) = l_self_modules_patch_embed4_modules_norm_parameters_bias_ = None + cls_tokens_3 = l_self_parameters_cls_token4_.expand(1, -1, -1) + l_self_parameters_cls_token4_ = None + x_111 = torch.cat((cls_tokens_3, x_110), dim=1) + cls_tokens_3 = x_110 = None + cls_token_6 = x_111[(slice(None, None, None), slice(None, 1, None))] + img_tokens_6 = x_111[(slice(None, None, None), slice(1, None, None))] + x_111 = None + transpose_40 = img_tokens_6.transpose(1, 2) + img_tokens_6 = None + feat_6 = transpose_40.view(1, 320, 7, 7) + transpose_40 = None + conv2d_28 = torch.conv2d( + feat_6, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 320, + ) + x_112 = conv2d_28 + feat_6 + conv2d_28 = feat_6 = None + flatten_10 = x_112.flatten(2) + x_112 = None + x_113 = flatten_10.transpose(1, 2) + flatten_10 = None + x_114 = torch.cat((cls_token_6, x_113), dim=1) + cls_token_6 = x_113 = None + x_115 = torch.nn.functional.layer_norm( + x_114, + (320,), + l_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_115, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_115 = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_27 = linear_24.reshape(1, 50, 3, 8, 40) + linear_24 = None + qkv_6 = reshape_27.permute(2, 0, 3, 1, 4) + reshape_27 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + k_softmax_6 = k_6.softmax(dim=2) + k_6 = None + transpose_42 = k_softmax_6.transpose(-1, -2) + k_softmax_6 = None + factor_att_12 = transpose_42 @ v_6 + transpose_42 = None + factor_att_13 = q_6 @ factor_att_12 + factor_att_12 = None + q_img_6 = q_6[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_6 = None + v_img_12 = v_6[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_6 = None + transpose_43 = v_img_12.transpose(-1, -2) + v_img_12 = None + v_img_13 = transpose_43.reshape(1, 320, 7, 7) + transpose_43 = None + split_6 = torch.functional.split(v_img_13, [80, 120, 120], dim=1) + v_img_13 = None + getitem_78 = split_6[0] + getitem_79 = split_6[1] + getitem_80 = split_6[2] + split_6 = None + conv2d_29 = torch.conv2d( + getitem_78, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 80, + ) + getitem_78 = None + conv2d_30 = torch.conv2d( + getitem_79, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 120, + ) + getitem_79 = None + conv2d_31 = torch.conv2d( + getitem_80, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 120, + ) + getitem_80 = None + conv_v_img_12 = torch.cat([conv2d_29, conv2d_30, conv2d_31], dim=1) + conv2d_29 = conv2d_30 = conv2d_31 = None + reshape_29 = conv_v_img_12.reshape(1, 8, 40, 49) + conv_v_img_12 = None + conv_v_img_13 = reshape_29.transpose(-1, -2) + reshape_29 = None + EV_hat_12 = q_img_6 * conv_v_img_13 + q_img_6 = conv_v_img_13 = None + EV_hat_13 = torch._C._nn.pad(EV_hat_12, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_12 = None + mul_13 = 0.15811388300841897 * factor_att_13 + factor_att_13 = None + x_116 = mul_13 + EV_hat_13 + mul_13 = EV_hat_13 = None + transpose_45 = x_116.transpose(1, 2) + x_116 = None + x_117 = transpose_45.reshape(1, 50, 320) + transpose_45 = None + x_118 = torch._C._nn.linear( + x_117, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_117 = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_119 = torch.nn.functional.dropout(x_118, 0.0, False, False) + x_118 = None + x_120 = x_114 + x_119 + x_114 = x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (320,), + l_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_ + ) = None + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122, approximate="none") + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_124 = l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_120 + x_126 + x_120 = x_126 = None + cls_token_7 = x_127[(slice(None, None, None), slice(None, 1, None))] + img_tokens_7 = x_127[(slice(None, None, None), slice(1, None, None))] + x_127 = None + transpose_46 = img_tokens_7.transpose(1, 2) + img_tokens_7 = None + feat_7 = transpose_46.view(1, 320, 7, 7) + transpose_46 = None + conv2d_32 = torch.conv2d( + feat_7, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 320, + ) + l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_ = (None) + x_128 = conv2d_32 + feat_7 + conv2d_32 = feat_7 = None + flatten_11 = x_128.flatten(2) + x_128 = None + x_129 = flatten_11.transpose(1, 2) + flatten_11 = None + x_130 = torch.cat((cls_token_7, x_129), dim=1) + cls_token_7 = x_129 = None + x_131 = torch.nn.functional.layer_norm( + x_130, + (320,), + l_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_, + l_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_131, + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_, + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_, + ) + x_131 = l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_ = l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_ = (None) + reshape_31 = linear_28.reshape(1, 50, 3, 8, 40) + linear_28 = None + qkv_7 = reshape_31.permute(2, 0, 3, 1, 4) + reshape_31 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + k_softmax_7 = k_7.softmax(dim=2) + k_7 = None + transpose_48 = k_softmax_7.transpose(-1, -2) + k_softmax_7 = None + factor_att_14 = transpose_48 @ v_7 + transpose_48 = None + factor_att_15 = q_7 @ factor_att_14 + factor_att_14 = None + q_img_7 = q_7[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + q_7 = None + v_img_14 = v_7[ + ( + slice(None, None, None), + slice(None, None, None), + slice(1, None, None), + slice(None, None, None), + ) + ] + v_7 = None + transpose_49 = v_img_14.transpose(-1, -2) + v_img_14 = None + v_img_15 = transpose_49.reshape(1, 320, 7, 7) + transpose_49 = None + split_7 = torch.functional.split(v_img_15, [80, 120, 120], dim=1) + v_img_15 = None + getitem_88 = split_7[0] + getitem_89 = split_7[1] + getitem_90 = split_7[2] + split_7 = None + conv2d_33 = torch.conv2d( + getitem_88, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 80, + ) + getitem_88 = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_ = (None) + conv2d_34 = torch.conv2d( + getitem_89, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_, + (1, 1), + (2, 2), + (1, 1), + 120, + ) + getitem_89 = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_ = (None) + conv2d_35 = torch.conv2d( + getitem_90, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_, + l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 120, + ) + getitem_90 = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_ = l_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_ = (None) + conv_v_img_14 = torch.cat([conv2d_33, conv2d_34, conv2d_35], dim=1) + conv2d_33 = conv2d_34 = conv2d_35 = None + reshape_33 = conv_v_img_14.reshape(1, 8, 40, 49) + conv_v_img_14 = None + conv_v_img_15 = reshape_33.transpose(-1, -2) + reshape_33 = None + EV_hat_14 = q_img_7 * conv_v_img_15 + q_img_7 = conv_v_img_15 = None + EV_hat_15 = torch._C._nn.pad(EV_hat_14, (0, 0, 1, 0, 0, 0), "constant", None) + EV_hat_14 = None + mul_15 = 0.15811388300841897 * factor_att_15 + factor_att_15 = None + x_132 = mul_15 + EV_hat_15 + mul_15 = EV_hat_15 = None + transpose_51 = x_132.transpose(1, 2) + x_132 = None + x_133 = transpose_51.reshape(1, 50, 320) + transpose_51 = None + x_134 = torch._C._nn.linear( + x_133, + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_, + l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_, + ) + x_133 = l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_ = l_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_ = (None) + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = x_130 + x_135 + x_130 = x_135 = None + x_137 = torch.nn.functional.layer_norm( + x_136, + (320,), + l_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_, + l_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_ + ) = None + x_138 = torch._C._nn.linear( + x_137, + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_137 = l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_139 = torch._C._nn.gelu(x_138, approximate="none") + x_138 = None + x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) + x_139 = None + x_141 = torch._C._nn.linear( + x_140, + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_140 = l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_136 + x_142 + x_136 = x_142 = None + getitem_91 = x_143[ + (slice(None, None, None), slice(1, None, None), slice(None, None, None)) + ] + reshape_35 = getitem_91.reshape(1, 7, 7, -1) + getitem_91 = None + permute_11 = reshape_35.permute(0, 3, 1, 2) + reshape_35 = None + x4_nocls = permute_11.contiguous() + permute_11 = x4_nocls = None + x_144 = torch.nn.functional.layer_norm( + x_143, + (320,), + l_self_modules_norm4_parameters_weight_, + l_self_modules_norm4_parameters_bias_, + 1e-06, + ) + x_143 = ( + l_self_modules_norm4_parameters_weight_ + ) = l_self_modules_norm4_parameters_bias_ = None + x_145 = x_144[(slice(None, None, None), 0)] + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = torch._C._nn.linear( + x_146, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_146 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_147,) diff --git a/samples/timm/coat_lite_tiny.in1k/weight_meta.py b/samples/timm/coat_lite_tiny.in1k/weight_meta.py new file mode 100644 index 0000000000..cd8cf94139 --- /dev/null +++ b/samples/timm/coat_lite_tiny.in1k/weight_meta.py @@ -0,0 +1,1528 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.231 + std = 1.288 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed1_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed1_modules_proj_parameters_weight_" + shape = [64, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed1_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed1_modules_proj_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.005 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed1_modules_norm_parameters_weight_: + name = "L_self_modules_patch_embed1_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed1_modules_norm_parameters_bias_: + name = "L_self_modules_patch_embed1_modules_norm_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token1_: + name = "L_self_parameters_cls_token1_" + shape = [1, 1, 64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_weight_" + shape = [64, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.014 + std = 0.194 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_cpe_modules_proj_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.023 + std = 0.198 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_norm1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [192, 64] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_" + shape = [16, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.188 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_" + shape = [16] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.025 + std = 0.199 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_" + shape = [24, 1, 5, 5] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.115 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.012 + std = 0.103 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_" + shape = [24, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.085 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.049 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [64, 64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_norm2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_1_modules_norm1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [192, 64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [64, 64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_1_modules_norm2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 64] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed2_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed2_modules_proj_parameters_weight_" + shape = [128, 64, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed2_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed2_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.035 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed2_modules_norm_parameters_weight_: + name = "L_self_modules_patch_embed2_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed2_modules_norm_parameters_bias_: + name = "L_self_modules_patch_embed2_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token2_: + name = "L_self_parameters_cls_token2_" + shape = [1, 1, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_weight_" + shape = [128, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.187 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_cpe_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.018 + std = 0.194 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_" + shape = [32, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.011 + std = 0.183 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.030 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_" + shape = [48, 1, 5, 5] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.115 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.024 + std = 0.124 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_" + shape = [48, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.081 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.008 + std = 0.091 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_1_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks2_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed3_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed3_modules_proj_parameters_weight_" + shape = [256, 128, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed3_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed3_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed3_modules_norm_parameters_weight_: + name = "L_self_modules_patch_embed3_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed3_modules_norm_parameters_bias_: + name = "L_self_modules_patch_embed3_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token3_: + name = "L_self_parameters_cls_token3_" + shape = [1, 1, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_weight_" + shape = [256, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.191 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_cpe_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.016 + std = 0.196 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_" + shape = [64, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.193 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.029 + std = 0.201 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_" + shape = [96, 1, 5, 5] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.115 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.008 + std = 0.106 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_1_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_1_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks3_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed4_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed4_modules_proj_parameters_weight_" + shape = [320, 256, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed4_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed4_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed4_modules_norm_parameters_weight_: + name = "L_self_modules_patch_embed4_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed4_modules_norm_parameters_bias_: + name = "L_self_modules_patch_embed4_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token4_: + name = "L_self_parameters_cls_token4_" + shape = [1, 1, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_weight_" + shape = [320, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.194 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_cpe_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.013 + std = 0.192 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_weight_" + shape = [80, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.196 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_0_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.006 + std = 0.183 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_weight_" + shape = [120, 1, 5, 5] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.116 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.018 + std = 0.116 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_weight_" + shape = [120, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_crpe_modules_conv_list_modules_2_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.009 + std = 0.086 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_norm2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_1_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_weight_" + shape = [960, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_qkv_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_weight_" + shape = [320, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_1_modules_factoratt_crpe_modules_proj_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_1_modules_norm2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_serial_blocks4_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm4_parameters_weight_: + name = "L_self_modules_norm4_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm4_parameters_bias_: + name = "L_self_modules_norm4_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None From 75b2eab8fa0e2700f4929df38769cf5eb5138dd8 Mon Sep 17 00:00:00 2001 From: Liyulingyue <852433440@qq.com> Date: Mon, 4 Aug 2025 23:18:29 +0800 Subject: [PATCH 5/7] timm group --- .../graph_hash.txt | 1 - .../model.py | 4246 ----------------- .../weight_meta.py | 4090 ---------------- .../graph_hash.txt | 1 - .../model.py | 4234 ---------------- .../weight_meta.py | 4070 ---------------- .../beit3_large_patch16_224.pt/graph_hash.txt | 1 - .../timm/beit3_large_patch16_224.pt/model.py | 4234 ---------------- .../beit3_large_patch16_224.pt/weight_meta.py | 4070 ---------------- .../graph_hash.txt | 1 + .../graph_net.json | 0 .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../convmixer_1024_20_ks9_p14.in1k/model.py | 1924 ++++++++ .../weight_meta.py | 2498 ++++++++++ .../convmixer_1536_20.in1k/graph_hash.txt | 1 + .../graph_net.json | 0 .../input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/convmixer_1536_20.in1k/model.py | 1924 ++++++++ .../convmixer_1536_20.in1k/weight_meta.py | 2498 ++++++++++ .../timm/convmixer_768_32.in1k/graph_hash.txt | 1 + .../graph_net.json | 0 .../input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/convmixer_768_32.in1k/model.py | 3028 ++++++++++++ .../timm/convmixer_768_32.in1k/weight_meta.py | 3938 +++++++++++++++ .../timm/convnext_atto.d2_in1k/graph_hash.txt | 1 + .../timm/convnext_atto.d2_in1k/graph_net.json | 5 + .../timm/convnext_atto.d2_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/convnext_atto.d2_in1k/model.py | 1123 +++++ .../timm/convnext_atto.d2_in1k/weight_meta.py | 1310 +++++ .../convnext_atto_ols.a2_in1k/graph_hash.txt | 1 + .../convnext_atto_ols.a2_in1k/graph_net.json | 5 + .../convnext_atto_ols.a2_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/convnext_atto_ols.a2_in1k/model.py | 1143 +++++ .../convnext_atto_ols.a2_in1k/weight_meta.py | 1330 ++++++ .../convnext_base.clip_laion2b/graph_hash.txt | 1 + .../convnext_base.clip_laion2b/graph_net.json | 5 + .../convnext_base.clip_laion2b/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/convnext_base.clip_laion2b/model.py | 2707 +++++++++++ .../convnext_base.clip_laion2b/weight_meta.py | 3470 ++++++++++++++ 45 files changed, 26914 insertions(+), 24947 deletions(-) delete mode 100644 samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/graph_hash.txt delete mode 100644 samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/model.py delete mode 100644 samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/weight_meta.py delete mode 100644 samples/timm/beit3_large_patch16_224.indomain_pt/graph_hash.txt delete mode 100644 samples/timm/beit3_large_patch16_224.indomain_pt/model.py delete mode 100644 samples/timm/beit3_large_patch16_224.indomain_pt/weight_meta.py delete mode 100644 samples/timm/beit3_large_patch16_224.pt/graph_hash.txt delete mode 100644 samples/timm/beit3_large_patch16_224.pt/model.py delete mode 100644 samples/timm/beit3_large_patch16_224.pt/weight_meta.py create mode 100644 samples/timm/convmixer_1024_20_ks9_p14.in1k/graph_hash.txt rename samples/timm/{beit3_large_patch16_224.indomain_in22k_ft_in1k => convmixer_1024_20_ks9_p14.in1k}/graph_net.json (100%) rename samples/timm/{beit3_large_patch16_224.indomain_in22k_ft_in1k => convmixer_1024_20_ks9_p14.in1k}/input_meta.py (100%) rename samples/timm/{beit3_large_patch16_224.indomain_in22k_ft_in1k => convmixer_1024_20_ks9_p14.in1k}/input_tensor_constraints.py (100%) create mode 100644 samples/timm/convmixer_1024_20_ks9_p14.in1k/model.py create mode 100644 samples/timm/convmixer_1024_20_ks9_p14.in1k/weight_meta.py create mode 100644 samples/timm/convmixer_1536_20.in1k/graph_hash.txt rename samples/timm/{beit3_large_patch16_224.indomain_pt => convmixer_1536_20.in1k}/graph_net.json (100%) rename samples/timm/{beit3_large_patch16_224.indomain_pt => convmixer_1536_20.in1k}/input_meta.py (100%) rename samples/timm/{beit3_large_patch16_224.indomain_pt => convmixer_1536_20.in1k}/input_tensor_constraints.py (100%) create mode 100644 samples/timm/convmixer_1536_20.in1k/model.py create mode 100644 samples/timm/convmixer_1536_20.in1k/weight_meta.py create mode 100644 samples/timm/convmixer_768_32.in1k/graph_hash.txt rename samples/timm/{beit3_large_patch16_224.pt => convmixer_768_32.in1k}/graph_net.json (100%) rename samples/timm/{beit3_large_patch16_224.pt => convmixer_768_32.in1k}/input_meta.py (100%) rename samples/timm/{beit3_large_patch16_224.pt => convmixer_768_32.in1k}/input_tensor_constraints.py (100%) create mode 100644 samples/timm/convmixer_768_32.in1k/model.py create mode 100644 samples/timm/convmixer_768_32.in1k/weight_meta.py create mode 100644 samples/timm/convnext_atto.d2_in1k/graph_hash.txt create mode 100644 samples/timm/convnext_atto.d2_in1k/graph_net.json create mode 100644 samples/timm/convnext_atto.d2_in1k/input_meta.py create mode 100644 samples/timm/convnext_atto.d2_in1k/input_tensor_constraints.py create mode 100644 samples/timm/convnext_atto.d2_in1k/model.py create mode 100644 samples/timm/convnext_atto.d2_in1k/weight_meta.py create mode 100644 samples/timm/convnext_atto_ols.a2_in1k/graph_hash.txt create mode 100644 samples/timm/convnext_atto_ols.a2_in1k/graph_net.json create mode 100644 samples/timm/convnext_atto_ols.a2_in1k/input_meta.py create mode 100644 samples/timm/convnext_atto_ols.a2_in1k/input_tensor_constraints.py create mode 100644 samples/timm/convnext_atto_ols.a2_in1k/model.py create mode 100644 samples/timm/convnext_atto_ols.a2_in1k/weight_meta.py create mode 100644 samples/timm/convnext_base.clip_laion2b/graph_hash.txt create mode 100644 samples/timm/convnext_base.clip_laion2b/graph_net.json create mode 100644 samples/timm/convnext_base.clip_laion2b/input_meta.py create mode 100644 samples/timm/convnext_base.clip_laion2b/input_tensor_constraints.py create mode 100644 samples/timm/convnext_base.clip_laion2b/model.py create mode 100644 samples/timm/convnext_base.clip_laion2b/weight_meta.py diff --git a/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/graph_hash.txt b/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/graph_hash.txt deleted file mode 100644 index 44b074954d..0000000000 --- a/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/graph_hash.txt +++ /dev/null @@ -1 +0,0 @@ -a79caac738b040f293c01cd92e1770f312f059371fd757fa138f608c37b3cf6e \ No newline at end of file diff --git a/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/model.py b/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/model.py deleted file mode 100644 index c424406a3e..0000000000 --- a/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/model.py +++ /dev/null @@ -1,4246 +0,0 @@ -import torch - - -class GraphModule(torch.nn.Module): - def forward( - self, - L_x_: torch.Tensor, - L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, - L_self_parameters_cls_token_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_num_prefix_tokens: torch.SymInt, - L_self_modules_fc_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_fc_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, - ): - l_x_ = L_x_ - l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( - L_self_modules_patch_embed_modules_proj_parameters_weight_ - ) - l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( - L_self_modules_patch_embed_modules_proj_parameters_bias_ - ) - l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ - l_self_parameters_cls_token_ = L_self_parameters_cls_token_ - l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_num_prefix_tokens = L_self_num_prefix_tokens - l_self_modules_fc_norm_parameters_weight_ = ( - L_self_modules_fc_norm_parameters_weight_ - ) - l_self_modules_fc_norm_parameters_bias_ = ( - L_self_modules_fc_norm_parameters_bias_ - ) - l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ - l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ - x = torch.conv2d( - l_x_, - l_self_modules_patch_embed_modules_proj_parameters_weight_, - l_self_modules_patch_embed_modules_proj_parameters_bias_, - (16, 16), - (0, 0), - (1, 1), - 1, - ) - l_x_ = ( - l_self_modules_patch_embed_modules_proj_parameters_weight_ - ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None - flatten = x.flatten(2) - x = None - x_1 = flatten.transpose(1, 2) - flatten = None - expand = l_self_parameters_cls_token_.expand(1, -1, -1) - l_self_parameters_cls_token_ = None - x_2 = torch.cat([expand, x_1], dim=1) - expand = x_1 = None - x_3 = x_2 + l_self_parameters_pos_embed_ - x_2 = l_self_parameters_pos_embed_ = None - x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) - x_3 = None - x_5 = torch.nn.functional.layer_norm( - x_4, - (1024,), - l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ - ) = None - linear = torch._C._nn.linear( - x_5, - l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, - ) - x_5 = ( - l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape = linear.reshape(1, 197, 3, 16, 64) - linear = None - qkv = reshape.permute(2, 0, 3, 1, 4) - reshape = None - unbind = qkv.unbind(0) - qkv = None - q = unbind[0] - k = unbind[1] - v = unbind[2] - unbind = None - x_6 = torch._C._nn.scaled_dot_product_attention( - q, k, v, attn_mask=None, dropout_p=0.0 - ) - q = k = v = None - transpose_1 = x_6.transpose(1, 2) - x_6 = None - x_7 = transpose_1.reshape(1, 197, 1024) - transpose_1 = None - x_8 = torch.nn.functional.layer_norm( - x_7, - (1024,), - l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_7 = ( - l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_ - ) = None - x_9 = torch._C._nn.linear( - x_8, - l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, - ) - x_8 = ( - l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ - ) = None - x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) - x_9 = None - x_11 = x_4 + x_10 - x_4 = x_10 = None - x_12 = torch.nn.functional.layer_norm( - x_11, - (1024,), - l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ - ) = None - x_13 = torch._C._nn.linear( - x_12, - l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, - ) - x_12 = ( - l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_14 = torch._C._nn.gelu(x_13, approximate="none") - x_13 = None - x_15 = torch.nn.functional.dropout(x_14, 0.0, False, False) - x_14 = None - x_16 = torch.nn.functional.layer_norm( - x_15, - (4096,), - l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_15 = ( - l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_17 = torch._C._nn.linear( - x_16, - l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, - ) - x_16 = ( - l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_18 = torch.nn.functional.dropout(x_17, 0.0, False, False) - x_17 = None - x_19 = x_11 + x_18 - x_11 = x_18 = None - x_20 = torch.nn.functional.layer_norm( - x_19, - (1024,), - l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ - ) = None - linear_4 = torch._C._nn.linear( - x_20, - l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, - ) - x_20 = ( - l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_2 = linear_4.reshape(1, 197, 3, 16, 64) - linear_4 = None - qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) - reshape_2 = None - unbind_1 = qkv_1.unbind(0) - qkv_1 = None - q_1 = unbind_1[0] - k_1 = unbind_1[1] - v_1 = unbind_1[2] - unbind_1 = None - x_21 = torch._C._nn.scaled_dot_product_attention( - q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 - ) - q_1 = k_1 = v_1 = None - transpose_2 = x_21.transpose(1, 2) - x_21 = None - x_22 = transpose_2.reshape(1, 197, 1024) - transpose_2 = None - x_23 = torch.nn.functional.layer_norm( - x_22, - (1024,), - l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_22 = ( - l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_ - ) = None - x_24 = torch._C._nn.linear( - x_23, - l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, - ) - x_23 = ( - l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ - ) = None - x_25 = torch.nn.functional.dropout(x_24, 0.0, False, False) - x_24 = None - x_26 = x_19 + x_25 - x_19 = x_25 = None - x_27 = torch.nn.functional.layer_norm( - x_26, - (1024,), - l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ - ) = None - x_28 = torch._C._nn.linear( - x_27, - l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, - ) - x_27 = ( - l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_29 = torch._C._nn.gelu(x_28, approximate="none") - x_28 = None - x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) - x_29 = None - x_31 = torch.nn.functional.layer_norm( - x_30, - (4096,), - l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_30 = ( - l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_32 = torch._C._nn.linear( - x_31, - l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, - ) - x_31 = ( - l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) - x_32 = None - x_34 = x_26 + x_33 - x_26 = x_33 = None - x_35 = torch.nn.functional.layer_norm( - x_34, - (1024,), - l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ - ) = None - linear_8 = torch._C._nn.linear( - x_35, - l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, - ) - x_35 = ( - l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_4 = linear_8.reshape(1, 197, 3, 16, 64) - linear_8 = None - qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) - reshape_4 = None - unbind_2 = qkv_2.unbind(0) - qkv_2 = None - q_2 = unbind_2[0] - k_2 = unbind_2[1] - v_2 = unbind_2[2] - unbind_2 = None - x_36 = torch._C._nn.scaled_dot_product_attention( - q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 - ) - q_2 = k_2 = v_2 = None - transpose_3 = x_36.transpose(1, 2) - x_36 = None - x_37 = transpose_3.reshape(1, 197, 1024) - transpose_3 = None - x_38 = torch.nn.functional.layer_norm( - x_37, - (1024,), - l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_37 = ( - l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_ - ) = None - x_39 = torch._C._nn.linear( - x_38, - l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, - ) - x_38 = ( - l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ - ) = None - x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) - x_39 = None - x_41 = x_34 + x_40 - x_34 = x_40 = None - x_42 = torch.nn.functional.layer_norm( - x_41, - (1024,), - l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ - ) = None - x_43 = torch._C._nn.linear( - x_42, - l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, - ) - x_42 = ( - l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_44 = torch._C._nn.gelu(x_43, approximate="none") - x_43 = None - x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) - x_44 = None - x_46 = torch.nn.functional.layer_norm( - x_45, - (4096,), - l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_45 = ( - l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_47 = torch._C._nn.linear( - x_46, - l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, - ) - x_46 = ( - l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) - x_47 = None - x_49 = x_41 + x_48 - x_41 = x_48 = None - x_50 = torch.nn.functional.layer_norm( - x_49, - (1024,), - l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ - ) = None - linear_12 = torch._C._nn.linear( - x_50, - l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, - ) - x_50 = ( - l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_6 = linear_12.reshape(1, 197, 3, 16, 64) - linear_12 = None - qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) - reshape_6 = None - unbind_3 = qkv_3.unbind(0) - qkv_3 = None - q_3 = unbind_3[0] - k_3 = unbind_3[1] - v_3 = unbind_3[2] - unbind_3 = None - x_51 = torch._C._nn.scaled_dot_product_attention( - q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 - ) - q_3 = k_3 = v_3 = None - transpose_4 = x_51.transpose(1, 2) - x_51 = None - x_52 = transpose_4.reshape(1, 197, 1024) - transpose_4 = None - x_53 = torch.nn.functional.layer_norm( - x_52, - (1024,), - l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_52 = ( - l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_ - ) = None - x_54 = torch._C._nn.linear( - x_53, - l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, - ) - x_53 = ( - l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ - ) = None - x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) - x_54 = None - x_56 = x_49 + x_55 - x_49 = x_55 = None - x_57 = torch.nn.functional.layer_norm( - x_56, - (1024,), - l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ - ) = None - x_58 = torch._C._nn.linear( - x_57, - l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, - ) - x_57 = ( - l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_59 = torch._C._nn.gelu(x_58, approximate="none") - x_58 = None - x_60 = torch.nn.functional.dropout(x_59, 0.0, False, False) - x_59 = None - x_61 = torch.nn.functional.layer_norm( - x_60, - (4096,), - l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_60 = ( - l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_62 = torch._C._nn.linear( - x_61, - l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, - ) - x_61 = ( - l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_63 = torch.nn.functional.dropout(x_62, 0.0, False, False) - x_62 = None - x_64 = x_56 + x_63 - x_56 = x_63 = None - x_65 = torch.nn.functional.layer_norm( - x_64, - (1024,), - l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ - ) = None - linear_16 = torch._C._nn.linear( - x_65, - l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, - ) - x_65 = ( - l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_8 = linear_16.reshape(1, 197, 3, 16, 64) - linear_16 = None - qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) - reshape_8 = None - unbind_4 = qkv_4.unbind(0) - qkv_4 = None - q_4 = unbind_4[0] - k_4 = unbind_4[1] - v_4 = unbind_4[2] - unbind_4 = None - x_66 = torch._C._nn.scaled_dot_product_attention( - q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 - ) - q_4 = k_4 = v_4 = None - transpose_5 = x_66.transpose(1, 2) - x_66 = None - x_67 = transpose_5.reshape(1, 197, 1024) - transpose_5 = None - x_68 = torch.nn.functional.layer_norm( - x_67, - (1024,), - l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_67 = ( - l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_ - ) = None - x_69 = torch._C._nn.linear( - x_68, - l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, - ) - x_68 = ( - l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ - ) = None - x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) - x_69 = None - x_71 = x_64 + x_70 - x_64 = x_70 = None - x_72 = torch.nn.functional.layer_norm( - x_71, - (1024,), - l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ - ) = None - x_73 = torch._C._nn.linear( - x_72, - l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, - ) - x_72 = ( - l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_74 = torch._C._nn.gelu(x_73, approximate="none") - x_73 = None - x_75 = torch.nn.functional.dropout(x_74, 0.0, False, False) - x_74 = None - x_76 = torch.nn.functional.layer_norm( - x_75, - (4096,), - l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_75 = ( - l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_77 = torch._C._nn.linear( - x_76, - l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, - ) - x_76 = ( - l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_78 = torch.nn.functional.dropout(x_77, 0.0, False, False) - x_77 = None - x_79 = x_71 + x_78 - x_71 = x_78 = None - x_80 = torch.nn.functional.layer_norm( - x_79, - (1024,), - l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ - ) = None - linear_20 = torch._C._nn.linear( - x_80, - l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, - ) - x_80 = ( - l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_10 = linear_20.reshape(1, 197, 3, 16, 64) - linear_20 = None - qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) - reshape_10 = None - unbind_5 = qkv_5.unbind(0) - qkv_5 = None - q_5 = unbind_5[0] - k_5 = unbind_5[1] - v_5 = unbind_5[2] - unbind_5 = None - x_81 = torch._C._nn.scaled_dot_product_attention( - q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 - ) - q_5 = k_5 = v_5 = None - transpose_6 = x_81.transpose(1, 2) - x_81 = None - x_82 = transpose_6.reshape(1, 197, 1024) - transpose_6 = None - x_83 = torch.nn.functional.layer_norm( - x_82, - (1024,), - l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_82 = ( - l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_ - ) = None - x_84 = torch._C._nn.linear( - x_83, - l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, - ) - x_83 = ( - l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ - ) = None - x_85 = torch.nn.functional.dropout(x_84, 0.0, False, False) - x_84 = None - x_86 = x_79 + x_85 - x_79 = x_85 = None - x_87 = torch.nn.functional.layer_norm( - x_86, - (1024,), - l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ - ) = None - x_88 = torch._C._nn.linear( - x_87, - l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, - ) - x_87 = ( - l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_89 = torch._C._nn.gelu(x_88, approximate="none") - x_88 = None - x_90 = torch.nn.functional.dropout(x_89, 0.0, False, False) - x_89 = None - x_91 = torch.nn.functional.layer_norm( - x_90, - (4096,), - l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_90 = ( - l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_92 = torch._C._nn.linear( - x_91, - l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, - ) - x_91 = ( - l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_93 = torch.nn.functional.dropout(x_92, 0.0, False, False) - x_92 = None - x_94 = x_86 + x_93 - x_86 = x_93 = None - x_95 = torch.nn.functional.layer_norm( - x_94, - (1024,), - l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ - ) = None - linear_24 = torch._C._nn.linear( - x_95, - l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, - ) - x_95 = ( - l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_12 = linear_24.reshape(1, 197, 3, 16, 64) - linear_24 = None - qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) - reshape_12 = None - unbind_6 = qkv_6.unbind(0) - qkv_6 = None - q_6 = unbind_6[0] - k_6 = unbind_6[1] - v_6 = unbind_6[2] - unbind_6 = None - x_96 = torch._C._nn.scaled_dot_product_attention( - q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 - ) - q_6 = k_6 = v_6 = None - transpose_7 = x_96.transpose(1, 2) - x_96 = None - x_97 = transpose_7.reshape(1, 197, 1024) - transpose_7 = None - x_98 = torch.nn.functional.layer_norm( - x_97, - (1024,), - l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_97 = ( - l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_ - ) = None - x_99 = torch._C._nn.linear( - x_98, - l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, - ) - x_98 = ( - l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ - ) = None - x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) - x_99 = None - x_101 = x_94 + x_100 - x_94 = x_100 = None - x_102 = torch.nn.functional.layer_norm( - x_101, - (1024,), - l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ - ) = None - x_103 = torch._C._nn.linear( - x_102, - l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, - ) - x_102 = ( - l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_104 = torch._C._nn.gelu(x_103, approximate="none") - x_103 = None - x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) - x_104 = None - x_106 = torch.nn.functional.layer_norm( - x_105, - (4096,), - l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_105 = ( - l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_107 = torch._C._nn.linear( - x_106, - l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, - ) - x_106 = ( - l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) - x_107 = None - x_109 = x_101 + x_108 - x_101 = x_108 = None - x_110 = torch.nn.functional.layer_norm( - x_109, - (1024,), - l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ - ) = None - linear_28 = torch._C._nn.linear( - x_110, - l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, - ) - x_110 = ( - l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_14 = linear_28.reshape(1, 197, 3, 16, 64) - linear_28 = None - qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) - reshape_14 = None - unbind_7 = qkv_7.unbind(0) - qkv_7 = None - q_7 = unbind_7[0] - k_7 = unbind_7[1] - v_7 = unbind_7[2] - unbind_7 = None - x_111 = torch._C._nn.scaled_dot_product_attention( - q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 - ) - q_7 = k_7 = v_7 = None - transpose_8 = x_111.transpose(1, 2) - x_111 = None - x_112 = transpose_8.reshape(1, 197, 1024) - transpose_8 = None - x_113 = torch.nn.functional.layer_norm( - x_112, - (1024,), - l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_112 = ( - l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_ - ) = None - x_114 = torch._C._nn.linear( - x_113, - l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, - ) - x_113 = ( - l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ - ) = None - x_115 = torch.nn.functional.dropout(x_114, 0.0, False, False) - x_114 = None - x_116 = x_109 + x_115 - x_109 = x_115 = None - x_117 = torch.nn.functional.layer_norm( - x_116, - (1024,), - l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ - ) = None - x_118 = torch._C._nn.linear( - x_117, - l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, - ) - x_117 = ( - l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_119 = torch._C._nn.gelu(x_118, approximate="none") - x_118 = None - x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) - x_119 = None - x_121 = torch.nn.functional.layer_norm( - x_120, - (4096,), - l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_120 = ( - l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_122 = torch._C._nn.linear( - x_121, - l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, - ) - x_121 = ( - l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) - x_122 = None - x_124 = x_116 + x_123 - x_116 = x_123 = None - x_125 = torch.nn.functional.layer_norm( - x_124, - (1024,), - l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ - ) = None - linear_32 = torch._C._nn.linear( - x_125, - l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, - ) - x_125 = ( - l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_16 = linear_32.reshape(1, 197, 3, 16, 64) - linear_32 = None - qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) - reshape_16 = None - unbind_8 = qkv_8.unbind(0) - qkv_8 = None - q_8 = unbind_8[0] - k_8 = unbind_8[1] - v_8 = unbind_8[2] - unbind_8 = None - x_126 = torch._C._nn.scaled_dot_product_attention( - q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 - ) - q_8 = k_8 = v_8 = None - transpose_9 = x_126.transpose(1, 2) - x_126 = None - x_127 = transpose_9.reshape(1, 197, 1024) - transpose_9 = None - x_128 = torch.nn.functional.layer_norm( - x_127, - (1024,), - l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_127 = ( - l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_ - ) = None - x_129 = torch._C._nn.linear( - x_128, - l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, - ) - x_128 = ( - l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ - ) = None - x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) - x_129 = None - x_131 = x_124 + x_130 - x_124 = x_130 = None - x_132 = torch.nn.functional.layer_norm( - x_131, - (1024,), - l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ - ) = None - x_133 = torch._C._nn.linear( - x_132, - l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, - ) - x_132 = ( - l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_134 = torch._C._nn.gelu(x_133, approximate="none") - x_133 = None - x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) - x_134 = None - x_136 = torch.nn.functional.layer_norm( - x_135, - (4096,), - l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_135 = ( - l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_137 = torch._C._nn.linear( - x_136, - l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, - ) - x_136 = ( - l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) - x_137 = None - x_139 = x_131 + x_138 - x_131 = x_138 = None - x_140 = torch.nn.functional.layer_norm( - x_139, - (1024,), - l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ - ) = None - linear_36 = torch._C._nn.linear( - x_140, - l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, - ) - x_140 = ( - l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_18 = linear_36.reshape(1, 197, 3, 16, 64) - linear_36 = None - qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) - reshape_18 = None - unbind_9 = qkv_9.unbind(0) - qkv_9 = None - q_9 = unbind_9[0] - k_9 = unbind_9[1] - v_9 = unbind_9[2] - unbind_9 = None - x_141 = torch._C._nn.scaled_dot_product_attention( - q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 - ) - q_9 = k_9 = v_9 = None - transpose_10 = x_141.transpose(1, 2) - x_141 = None - x_142 = transpose_10.reshape(1, 197, 1024) - transpose_10 = None - x_143 = torch.nn.functional.layer_norm( - x_142, - (1024,), - l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_142 = ( - l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_ - ) = None - x_144 = torch._C._nn.linear( - x_143, - l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, - ) - x_143 = ( - l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ - ) = None - x_145 = torch.nn.functional.dropout(x_144, 0.0, False, False) - x_144 = None - x_146 = x_139 + x_145 - x_139 = x_145 = None - x_147 = torch.nn.functional.layer_norm( - x_146, - (1024,), - l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ - ) = None - x_148 = torch._C._nn.linear( - x_147, - l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, - ) - x_147 = ( - l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_149 = torch._C._nn.gelu(x_148, approximate="none") - x_148 = None - x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) - x_149 = None - x_151 = torch.nn.functional.layer_norm( - x_150, - (4096,), - l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_150 = ( - l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_152 = torch._C._nn.linear( - x_151, - l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, - ) - x_151 = ( - l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_153 = torch.nn.functional.dropout(x_152, 0.0, False, False) - x_152 = None - x_154 = x_146 + x_153 - x_146 = x_153 = None - x_155 = torch.nn.functional.layer_norm( - x_154, - (1024,), - l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ - ) = None - linear_40 = torch._C._nn.linear( - x_155, - l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, - ) - x_155 = ( - l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_20 = linear_40.reshape(1, 197, 3, 16, 64) - linear_40 = None - qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) - reshape_20 = None - unbind_10 = qkv_10.unbind(0) - qkv_10 = None - q_10 = unbind_10[0] - k_10 = unbind_10[1] - v_10 = unbind_10[2] - unbind_10 = None - x_156 = torch._C._nn.scaled_dot_product_attention( - q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 - ) - q_10 = k_10 = v_10 = None - transpose_11 = x_156.transpose(1, 2) - x_156 = None - x_157 = transpose_11.reshape(1, 197, 1024) - transpose_11 = None - x_158 = torch.nn.functional.layer_norm( - x_157, - (1024,), - l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_157 = l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_ - ) = None - x_159 = torch._C._nn.linear( - x_158, - l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, - ) - x_158 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ - ) = None - x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) - x_159 = None - x_161 = x_154 + x_160 - x_154 = x_160 = None - x_162 = torch.nn.functional.layer_norm( - x_161, - (1024,), - l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ - ) = None - x_163 = torch._C._nn.linear( - x_162, - l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, - ) - x_162 = ( - l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_164 = torch._C._nn.gelu(x_163, approximate="none") - x_163 = None - x_165 = torch.nn.functional.dropout(x_164, 0.0, False, False) - x_164 = None - x_166 = torch.nn.functional.layer_norm( - x_165, - (4096,), - l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_165 = ( - l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_167 = torch._C._nn.linear( - x_166, - l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, - ) - x_166 = ( - l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) - x_167 = None - x_169 = x_161 + x_168 - x_161 = x_168 = None - x_170 = torch.nn.functional.layer_norm( - x_169, - (1024,), - l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ - ) = None - linear_44 = torch._C._nn.linear( - x_170, - l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, - ) - x_170 = ( - l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_22 = linear_44.reshape(1, 197, 3, 16, 64) - linear_44 = None - qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) - reshape_22 = None - unbind_11 = qkv_11.unbind(0) - qkv_11 = None - q_11 = unbind_11[0] - k_11 = unbind_11[1] - v_11 = unbind_11[2] - unbind_11 = None - x_171 = torch._C._nn.scaled_dot_product_attention( - q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 - ) - q_11 = k_11 = v_11 = None - transpose_12 = x_171.transpose(1, 2) - x_171 = None - x_172 = transpose_12.reshape(1, 197, 1024) - transpose_12 = None - x_173 = torch.nn.functional.layer_norm( - x_172, - (1024,), - l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_172 = l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_ - ) = None - x_174 = torch._C._nn.linear( - x_173, - l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, - ) - x_173 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ - ) = None - x_175 = torch.nn.functional.dropout(x_174, 0.0, False, False) - x_174 = None - x_176 = x_169 + x_175 - x_169 = x_175 = None - x_177 = torch.nn.functional.layer_norm( - x_176, - (1024,), - l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ - ) = None - x_178 = torch._C._nn.linear( - x_177, - l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, - ) - x_177 = ( - l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_179 = torch._C._nn.gelu(x_178, approximate="none") - x_178 = None - x_180 = torch.nn.functional.dropout(x_179, 0.0, False, False) - x_179 = None - x_181 = torch.nn.functional.layer_norm( - x_180, - (4096,), - l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_180 = ( - l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_182 = torch._C._nn.linear( - x_181, - l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, - ) - x_181 = ( - l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_183 = torch.nn.functional.dropout(x_182, 0.0, False, False) - x_182 = None - x_184 = x_176 + x_183 - x_176 = x_183 = None - x_185 = torch.nn.functional.layer_norm( - x_184, - (1024,), - l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ - ) = None - linear_48 = torch._C._nn.linear( - x_185, - l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_, - ) - x_185 = ( - l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_24 = linear_48.reshape(1, 197, 3, 16, 64) - linear_48 = None - qkv_12 = reshape_24.permute(2, 0, 3, 1, 4) - reshape_24 = None - unbind_12 = qkv_12.unbind(0) - qkv_12 = None - q_12 = unbind_12[0] - k_12 = unbind_12[1] - v_12 = unbind_12[2] - unbind_12 = None - x_186 = torch._C._nn.scaled_dot_product_attention( - q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 - ) - q_12 = k_12 = v_12 = None - transpose_13 = x_186.transpose(1, 2) - x_186 = None - x_187 = transpose_13.reshape(1, 197, 1024) - transpose_13 = None - x_188 = torch.nn.functional.layer_norm( - x_187, - (1024,), - l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_187 = l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_ - ) = None - x_189 = torch._C._nn.linear( - x_188, - l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_, - ) - x_188 = l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ - ) = None - x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) - x_189 = None - x_191 = x_184 + x_190 - x_184 = x_190 = None - x_192 = torch.nn.functional.layer_norm( - x_191, - (1024,), - l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ - ) = None - x_193 = torch._C._nn.linear( - x_192, - l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, - ) - x_192 = ( - l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_194 = torch._C._nn.gelu(x_193, approximate="none") - x_193 = None - x_195 = torch.nn.functional.dropout(x_194, 0.0, False, False) - x_194 = None - x_196 = torch.nn.functional.layer_norm( - x_195, - (4096,), - l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_195 = ( - l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_197 = torch._C._nn.linear( - x_196, - l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, - ) - x_196 = ( - l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_198 = torch.nn.functional.dropout(x_197, 0.0, False, False) - x_197 = None - x_199 = x_191 + x_198 - x_191 = x_198 = None - x_200 = torch.nn.functional.layer_norm( - x_199, - (1024,), - l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ - ) = None - linear_52 = torch._C._nn.linear( - x_200, - l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_, - ) - x_200 = ( - l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_26 = linear_52.reshape(1, 197, 3, 16, 64) - linear_52 = None - qkv_13 = reshape_26.permute(2, 0, 3, 1, 4) - reshape_26 = None - unbind_13 = qkv_13.unbind(0) - qkv_13 = None - q_13 = unbind_13[0] - k_13 = unbind_13[1] - v_13 = unbind_13[2] - unbind_13 = None - x_201 = torch._C._nn.scaled_dot_product_attention( - q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 - ) - q_13 = k_13 = v_13 = None - transpose_14 = x_201.transpose(1, 2) - x_201 = None - x_202 = transpose_14.reshape(1, 197, 1024) - transpose_14 = None - x_203 = torch.nn.functional.layer_norm( - x_202, - (1024,), - l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_202 = l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_ - ) = None - x_204 = torch._C._nn.linear( - x_203, - l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_, - ) - x_203 = l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ - ) = None - x_205 = torch.nn.functional.dropout(x_204, 0.0, False, False) - x_204 = None - x_206 = x_199 + x_205 - x_199 = x_205 = None - x_207 = torch.nn.functional.layer_norm( - x_206, - (1024,), - l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ - ) = None - x_208 = torch._C._nn.linear( - x_207, - l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, - ) - x_207 = ( - l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_209 = torch._C._nn.gelu(x_208, approximate="none") - x_208 = None - x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) - x_209 = None - x_211 = torch.nn.functional.layer_norm( - x_210, - (4096,), - l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_210 = ( - l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_212 = torch._C._nn.linear( - x_211, - l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, - ) - x_211 = ( - l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_213 = torch.nn.functional.dropout(x_212, 0.0, False, False) - x_212 = None - x_214 = x_206 + x_213 - x_206 = x_213 = None - x_215 = torch.nn.functional.layer_norm( - x_214, - (1024,), - l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ - ) = None - linear_56 = torch._C._nn.linear( - x_215, - l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_, - ) - x_215 = ( - l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_28 = linear_56.reshape(1, 197, 3, 16, 64) - linear_56 = None - qkv_14 = reshape_28.permute(2, 0, 3, 1, 4) - reshape_28 = None - unbind_14 = qkv_14.unbind(0) - qkv_14 = None - q_14 = unbind_14[0] - k_14 = unbind_14[1] - v_14 = unbind_14[2] - unbind_14 = None - x_216 = torch._C._nn.scaled_dot_product_attention( - q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 - ) - q_14 = k_14 = v_14 = None - transpose_15 = x_216.transpose(1, 2) - x_216 = None - x_217 = transpose_15.reshape(1, 197, 1024) - transpose_15 = None - x_218 = torch.nn.functional.layer_norm( - x_217, - (1024,), - l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_217 = l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_ - ) = None - x_219 = torch._C._nn.linear( - x_218, - l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_, - ) - x_218 = l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ - ) = None - x_220 = torch.nn.functional.dropout(x_219, 0.0, False, False) - x_219 = None - x_221 = x_214 + x_220 - x_214 = x_220 = None - x_222 = torch.nn.functional.layer_norm( - x_221, - (1024,), - l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ - ) = None - x_223 = torch._C._nn.linear( - x_222, - l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, - ) - x_222 = ( - l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_224 = torch._C._nn.gelu(x_223, approximate="none") - x_223 = None - x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) - x_224 = None - x_226 = torch.nn.functional.layer_norm( - x_225, - (4096,), - l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_225 = ( - l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_227 = torch._C._nn.linear( - x_226, - l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, - ) - x_226 = ( - l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_228 = torch.nn.functional.dropout(x_227, 0.0, False, False) - x_227 = None - x_229 = x_221 + x_228 - x_221 = x_228 = None - x_230 = torch.nn.functional.layer_norm( - x_229, - (1024,), - l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ - ) = None - linear_60 = torch._C._nn.linear( - x_230, - l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_, - ) - x_230 = ( - l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_30 = linear_60.reshape(1, 197, 3, 16, 64) - linear_60 = None - qkv_15 = reshape_30.permute(2, 0, 3, 1, 4) - reshape_30 = None - unbind_15 = qkv_15.unbind(0) - qkv_15 = None - q_15 = unbind_15[0] - k_15 = unbind_15[1] - v_15 = unbind_15[2] - unbind_15 = None - x_231 = torch._C._nn.scaled_dot_product_attention( - q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 - ) - q_15 = k_15 = v_15 = None - transpose_16 = x_231.transpose(1, 2) - x_231 = None - x_232 = transpose_16.reshape(1, 197, 1024) - transpose_16 = None - x_233 = torch.nn.functional.layer_norm( - x_232, - (1024,), - l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_232 = l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_ - ) = None - x_234 = torch._C._nn.linear( - x_233, - l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_, - ) - x_233 = l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ - ) = None - x_235 = torch.nn.functional.dropout(x_234, 0.0, False, False) - x_234 = None - x_236 = x_229 + x_235 - x_229 = x_235 = None - x_237 = torch.nn.functional.layer_norm( - x_236, - (1024,), - l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ - ) = None - x_238 = torch._C._nn.linear( - x_237, - l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, - ) - x_237 = ( - l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_239 = torch._C._nn.gelu(x_238, approximate="none") - x_238 = None - x_240 = torch.nn.functional.dropout(x_239, 0.0, False, False) - x_239 = None - x_241 = torch.nn.functional.layer_norm( - x_240, - (4096,), - l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_240 = ( - l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_242 = torch._C._nn.linear( - x_241, - l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, - ) - x_241 = ( - l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_243 = torch.nn.functional.dropout(x_242, 0.0, False, False) - x_242 = None - x_244 = x_236 + x_243 - x_236 = x_243 = None - x_245 = torch.nn.functional.layer_norm( - x_244, - (1024,), - l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ - ) = None - linear_64 = torch._C._nn.linear( - x_245, - l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_, - ) - x_245 = ( - l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_32 = linear_64.reshape(1, 197, 3, 16, 64) - linear_64 = None - qkv_16 = reshape_32.permute(2, 0, 3, 1, 4) - reshape_32 = None - unbind_16 = qkv_16.unbind(0) - qkv_16 = None - q_16 = unbind_16[0] - k_16 = unbind_16[1] - v_16 = unbind_16[2] - unbind_16 = None - x_246 = torch._C._nn.scaled_dot_product_attention( - q_16, k_16, v_16, attn_mask=None, dropout_p=0.0 - ) - q_16 = k_16 = v_16 = None - transpose_17 = x_246.transpose(1, 2) - x_246 = None - x_247 = transpose_17.reshape(1, 197, 1024) - transpose_17 = None - x_248 = torch.nn.functional.layer_norm( - x_247, - (1024,), - l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_247 = l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_ - ) = None - x_249 = torch._C._nn.linear( - x_248, - l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_, - ) - x_248 = l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ - ) = None - x_250 = torch.nn.functional.dropout(x_249, 0.0, False, False) - x_249 = None - x_251 = x_244 + x_250 - x_244 = x_250 = None - x_252 = torch.nn.functional.layer_norm( - x_251, - (1024,), - l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ - ) = None - x_253 = torch._C._nn.linear( - x_252, - l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, - ) - x_252 = ( - l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_254 = torch._C._nn.gelu(x_253, approximate="none") - x_253 = None - x_255 = torch.nn.functional.dropout(x_254, 0.0, False, False) - x_254 = None - x_256 = torch.nn.functional.layer_norm( - x_255, - (4096,), - l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_255 = ( - l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_257 = torch._C._nn.linear( - x_256, - l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, - ) - x_256 = ( - l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_258 = torch.nn.functional.dropout(x_257, 0.0, False, False) - x_257 = None - x_259 = x_251 + x_258 - x_251 = x_258 = None - x_260 = torch.nn.functional.layer_norm( - x_259, - (1024,), - l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ - ) = None - linear_68 = torch._C._nn.linear( - x_260, - l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_, - ) - x_260 = ( - l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_34 = linear_68.reshape(1, 197, 3, 16, 64) - linear_68 = None - qkv_17 = reshape_34.permute(2, 0, 3, 1, 4) - reshape_34 = None - unbind_17 = qkv_17.unbind(0) - qkv_17 = None - q_17 = unbind_17[0] - k_17 = unbind_17[1] - v_17 = unbind_17[2] - unbind_17 = None - x_261 = torch._C._nn.scaled_dot_product_attention( - q_17, k_17, v_17, attn_mask=None, dropout_p=0.0 - ) - q_17 = k_17 = v_17 = None - transpose_18 = x_261.transpose(1, 2) - x_261 = None - x_262 = transpose_18.reshape(1, 197, 1024) - transpose_18 = None - x_263 = torch.nn.functional.layer_norm( - x_262, - (1024,), - l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_262 = l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_ - ) = None - x_264 = torch._C._nn.linear( - x_263, - l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_, - ) - x_263 = l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ - ) = None - x_265 = torch.nn.functional.dropout(x_264, 0.0, False, False) - x_264 = None - x_266 = x_259 + x_265 - x_259 = x_265 = None - x_267 = torch.nn.functional.layer_norm( - x_266, - (1024,), - l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ - ) = None - x_268 = torch._C._nn.linear( - x_267, - l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, - ) - x_267 = ( - l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_269 = torch._C._nn.gelu(x_268, approximate="none") - x_268 = None - x_270 = torch.nn.functional.dropout(x_269, 0.0, False, False) - x_269 = None - x_271 = torch.nn.functional.layer_norm( - x_270, - (4096,), - l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_270 = ( - l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_272 = torch._C._nn.linear( - x_271, - l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, - ) - x_271 = ( - l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_273 = torch.nn.functional.dropout(x_272, 0.0, False, False) - x_272 = None - x_274 = x_266 + x_273 - x_266 = x_273 = None - x_275 = torch.nn.functional.layer_norm( - x_274, - (1024,), - l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ - ) = None - linear_72 = torch._C._nn.linear( - x_275, - l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_, - ) - x_275 = ( - l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_36 = linear_72.reshape(1, 197, 3, 16, 64) - linear_72 = None - qkv_18 = reshape_36.permute(2, 0, 3, 1, 4) - reshape_36 = None - unbind_18 = qkv_18.unbind(0) - qkv_18 = None - q_18 = unbind_18[0] - k_18 = unbind_18[1] - v_18 = unbind_18[2] - unbind_18 = None - x_276 = torch._C._nn.scaled_dot_product_attention( - q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 - ) - q_18 = k_18 = v_18 = None - transpose_19 = x_276.transpose(1, 2) - x_276 = None - x_277 = transpose_19.reshape(1, 197, 1024) - transpose_19 = None - x_278 = torch.nn.functional.layer_norm( - x_277, - (1024,), - l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_277 = l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_ - ) = None - x_279 = torch._C._nn.linear( - x_278, - l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_, - ) - x_278 = l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ - ) = None - x_280 = torch.nn.functional.dropout(x_279, 0.0, False, False) - x_279 = None - x_281 = x_274 + x_280 - x_274 = x_280 = None - x_282 = torch.nn.functional.layer_norm( - x_281, - (1024,), - l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ - ) = None - x_283 = torch._C._nn.linear( - x_282, - l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, - ) - x_282 = ( - l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_284 = torch._C._nn.gelu(x_283, approximate="none") - x_283 = None - x_285 = torch.nn.functional.dropout(x_284, 0.0, False, False) - x_284 = None - x_286 = torch.nn.functional.layer_norm( - x_285, - (4096,), - l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_285 = ( - l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_287 = torch._C._nn.linear( - x_286, - l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, - ) - x_286 = ( - l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) - x_287 = None - x_289 = x_281 + x_288 - x_281 = x_288 = None - x_290 = torch.nn.functional.layer_norm( - x_289, - (1024,), - l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ - ) = None - linear_76 = torch._C._nn.linear( - x_290, - l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_, - ) - x_290 = ( - l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_38 = linear_76.reshape(1, 197, 3, 16, 64) - linear_76 = None - qkv_19 = reshape_38.permute(2, 0, 3, 1, 4) - reshape_38 = None - unbind_19 = qkv_19.unbind(0) - qkv_19 = None - q_19 = unbind_19[0] - k_19 = unbind_19[1] - v_19 = unbind_19[2] - unbind_19 = None - x_291 = torch._C._nn.scaled_dot_product_attention( - q_19, k_19, v_19, attn_mask=None, dropout_p=0.0 - ) - q_19 = k_19 = v_19 = None - transpose_20 = x_291.transpose(1, 2) - x_291 = None - x_292 = transpose_20.reshape(1, 197, 1024) - transpose_20 = None - x_293 = torch.nn.functional.layer_norm( - x_292, - (1024,), - l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_292 = l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_ - ) = None - x_294 = torch._C._nn.linear( - x_293, - l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_, - ) - x_293 = l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ - ) = None - x_295 = torch.nn.functional.dropout(x_294, 0.0, False, False) - x_294 = None - x_296 = x_289 + x_295 - x_289 = x_295 = None - x_297 = torch.nn.functional.layer_norm( - x_296, - (1024,), - l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ - ) = None - x_298 = torch._C._nn.linear( - x_297, - l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, - ) - x_297 = ( - l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_299 = torch._C._nn.gelu(x_298, approximate="none") - x_298 = None - x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) - x_299 = None - x_301 = torch.nn.functional.layer_norm( - x_300, - (4096,), - l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_300 = ( - l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_302 = torch._C._nn.linear( - x_301, - l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, - ) - x_301 = ( - l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_303 = torch.nn.functional.dropout(x_302, 0.0, False, False) - x_302 = None - x_304 = x_296 + x_303 - x_296 = x_303 = None - x_305 = torch.nn.functional.layer_norm( - x_304, - (1024,), - l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ - ) = None - linear_80 = torch._C._nn.linear( - x_305, - l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_, - ) - x_305 = ( - l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_40 = linear_80.reshape(1, 197, 3, 16, 64) - linear_80 = None - qkv_20 = reshape_40.permute(2, 0, 3, 1, 4) - reshape_40 = None - unbind_20 = qkv_20.unbind(0) - qkv_20 = None - q_20 = unbind_20[0] - k_20 = unbind_20[1] - v_20 = unbind_20[2] - unbind_20 = None - x_306 = torch._C._nn.scaled_dot_product_attention( - q_20, k_20, v_20, attn_mask=None, dropout_p=0.0 - ) - q_20 = k_20 = v_20 = None - transpose_21 = x_306.transpose(1, 2) - x_306 = None - x_307 = transpose_21.reshape(1, 197, 1024) - transpose_21 = None - x_308 = torch.nn.functional.layer_norm( - x_307, - (1024,), - l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_307 = l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_ - ) = None - x_309 = torch._C._nn.linear( - x_308, - l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_, - ) - x_308 = l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ - ) = None - x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) - x_309 = None - x_311 = x_304 + x_310 - x_304 = x_310 = None - x_312 = torch.nn.functional.layer_norm( - x_311, - (1024,), - l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ - ) = None - x_313 = torch._C._nn.linear( - x_312, - l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, - ) - x_312 = ( - l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_314 = torch._C._nn.gelu(x_313, approximate="none") - x_313 = None - x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) - x_314 = None - x_316 = torch.nn.functional.layer_norm( - x_315, - (4096,), - l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_315 = ( - l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_317 = torch._C._nn.linear( - x_316, - l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, - ) - x_316 = ( - l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_318 = torch.nn.functional.dropout(x_317, 0.0, False, False) - x_317 = None - x_319 = x_311 + x_318 - x_311 = x_318 = None - x_320 = torch.nn.functional.layer_norm( - x_319, - (1024,), - l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ - ) = None - linear_84 = torch._C._nn.linear( - x_320, - l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_, - ) - x_320 = ( - l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_42 = linear_84.reshape(1, 197, 3, 16, 64) - linear_84 = None - qkv_21 = reshape_42.permute(2, 0, 3, 1, 4) - reshape_42 = None - unbind_21 = qkv_21.unbind(0) - qkv_21 = None - q_21 = unbind_21[0] - k_21 = unbind_21[1] - v_21 = unbind_21[2] - unbind_21 = None - x_321 = torch._C._nn.scaled_dot_product_attention( - q_21, k_21, v_21, attn_mask=None, dropout_p=0.0 - ) - q_21 = k_21 = v_21 = None - transpose_22 = x_321.transpose(1, 2) - x_321 = None - x_322 = transpose_22.reshape(1, 197, 1024) - transpose_22 = None - x_323 = torch.nn.functional.layer_norm( - x_322, - (1024,), - l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_322 = l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_ - ) = None - x_324 = torch._C._nn.linear( - x_323, - l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_, - ) - x_323 = l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ - ) = None - x_325 = torch.nn.functional.dropout(x_324, 0.0, False, False) - x_324 = None - x_326 = x_319 + x_325 - x_319 = x_325 = None - x_327 = torch.nn.functional.layer_norm( - x_326, - (1024,), - l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ - ) = None - x_328 = torch._C._nn.linear( - x_327, - l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, - ) - x_327 = ( - l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_329 = torch._C._nn.gelu(x_328, approximate="none") - x_328 = None - x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) - x_329 = None - x_331 = torch.nn.functional.layer_norm( - x_330, - (4096,), - l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_330 = ( - l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_332 = torch._C._nn.linear( - x_331, - l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, - ) - x_331 = ( - l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) - x_332 = None - x_334 = x_326 + x_333 - x_326 = x_333 = None - x_335 = torch.nn.functional.layer_norm( - x_334, - (1024,), - l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ - ) = None - linear_88 = torch._C._nn.linear( - x_335, - l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_, - ) - x_335 = ( - l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_44 = linear_88.reshape(1, 197, 3, 16, 64) - linear_88 = None - qkv_22 = reshape_44.permute(2, 0, 3, 1, 4) - reshape_44 = None - unbind_22 = qkv_22.unbind(0) - qkv_22 = None - q_22 = unbind_22[0] - k_22 = unbind_22[1] - v_22 = unbind_22[2] - unbind_22 = None - x_336 = torch._C._nn.scaled_dot_product_attention( - q_22, k_22, v_22, attn_mask=None, dropout_p=0.0 - ) - q_22 = k_22 = v_22 = None - transpose_23 = x_336.transpose(1, 2) - x_336 = None - x_337 = transpose_23.reshape(1, 197, 1024) - transpose_23 = None - x_338 = torch.nn.functional.layer_norm( - x_337, - (1024,), - l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_337 = l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_ - ) = None - x_339 = torch._C._nn.linear( - x_338, - l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_, - ) - x_338 = l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ - ) = None - x_340 = torch.nn.functional.dropout(x_339, 0.0, False, False) - x_339 = None - x_341 = x_334 + x_340 - x_334 = x_340 = None - x_342 = torch.nn.functional.layer_norm( - x_341, - (1024,), - l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ - ) = None - x_343 = torch._C._nn.linear( - x_342, - l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, - ) - x_342 = ( - l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_344 = torch._C._nn.gelu(x_343, approximate="none") - x_343 = None - x_345 = torch.nn.functional.dropout(x_344, 0.0, False, False) - x_344 = None - x_346 = torch.nn.functional.layer_norm( - x_345, - (4096,), - l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_345 = ( - l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_347 = torch._C._nn.linear( - x_346, - l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, - ) - x_346 = ( - l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) - x_347 = None - x_349 = x_341 + x_348 - x_341 = x_348 = None - x_350 = torch.nn.functional.layer_norm( - x_349, - (1024,), - l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ - ) = None - linear_92 = torch._C._nn.linear( - x_350, - l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_, - ) - x_350 = ( - l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_46 = linear_92.reshape(1, 197, 3, 16, 64) - linear_92 = None - qkv_23 = reshape_46.permute(2, 0, 3, 1, 4) - reshape_46 = None - unbind_23 = qkv_23.unbind(0) - qkv_23 = None - q_23 = unbind_23[0] - k_23 = unbind_23[1] - v_23 = unbind_23[2] - unbind_23 = None - x_351 = torch._C._nn.scaled_dot_product_attention( - q_23, k_23, v_23, attn_mask=None, dropout_p=0.0 - ) - q_23 = k_23 = v_23 = None - transpose_24 = x_351.transpose(1, 2) - x_351 = None - x_352 = transpose_24.reshape(1, 197, 1024) - transpose_24 = None - x_353 = torch.nn.functional.layer_norm( - x_352, - (1024,), - l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_352 = l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_ - ) = None - x_354 = torch._C._nn.linear( - x_353, - l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_, - ) - x_353 = l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ - ) = None - x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) - x_354 = None - x_356 = x_349 + x_355 - x_349 = x_355 = None - x_357 = torch.nn.functional.layer_norm( - x_356, - (1024,), - l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ - ) = None - x_358 = torch._C._nn.linear( - x_357, - l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, - ) - x_357 = ( - l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_359 = torch._C._nn.gelu(x_358, approximate="none") - x_358 = None - x_360 = torch.nn.functional.dropout(x_359, 0.0, False, False) - x_359 = None - x_361 = torch.nn.functional.layer_norm( - x_360, - (4096,), - l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_360 = ( - l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_362 = torch._C._nn.linear( - x_361, - l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, - ) - x_361 = ( - l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) - x_362 = None - x_364 = x_356 + x_363 - x_356 = x_363 = None - x_365 = x_364[ - (slice(None, None, None), slice(l_self_num_prefix_tokens, None, None)) - ] - x_364 = l_self_num_prefix_tokens = None - x_366 = x_365.mean(dim=1) - x_365 = None - x_367 = torch.nn.functional.layer_norm( - x_366, - (1024,), - l_self_modules_fc_norm_parameters_weight_, - l_self_modules_fc_norm_parameters_bias_, - 1e-05, - ) - x_366 = ( - l_self_modules_fc_norm_parameters_weight_ - ) = l_self_modules_fc_norm_parameters_bias_ = None - x_368 = torch.nn.functional.dropout(x_367, 0.0, False, False) - x_367 = None - x_369 = torch._C._nn.linear( - x_368, - l_self_modules_head_parameters_weight_, - l_self_modules_head_parameters_bias_, - ) - x_368 = ( - l_self_modules_head_parameters_weight_ - ) = l_self_modules_head_parameters_bias_ = None - return (x_369,) diff --git a/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/weight_meta.py b/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/weight_meta.py deleted file mode 100644 index 2f8b8e8681..0000000000 --- a/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/weight_meta.py +++ /dev/null @@ -1,4090 +0,0 @@ -class Program_weight_tensor_meta_L_x_: - name = "L_x_" - shape = [1, 3, 224, 224] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.222 - std = 1.286 - data = None - - -class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: - name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" - shape = [1024, 3, 16, 16] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.021 - data = None - - -class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: - name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.021 - data = None - - -class Program_weight_tensor_meta_L_self_parameters_pos_embed_: - name = "L_self_parameters_pos_embed_" - shape = [1, 197, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_parameters_cls_token_: - name = "L_self_parameters_cls_token_" - shape = [1, 1, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_num_prefix_tokens: - name = "L_self_num_prefix_tokens" - shape = [] - dtype = "torch.int64" - device = "cpu" - mean = None - std = None - data = [4] - - -class Program_weight_tensor_meta_L_self_modules_fc_norm_parameters_weight_: - name = "L_self_modules_fc_norm_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_fc_norm_parameters_bias_: - name = "L_self_modules_fc_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: - name = "L_self_modules_head_parameters_weight_" - shape = [1000, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: - name = "L_self_modules_head_parameters_bias_" - shape = [1000] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None diff --git a/samples/timm/beit3_large_patch16_224.indomain_pt/graph_hash.txt b/samples/timm/beit3_large_patch16_224.indomain_pt/graph_hash.txt deleted file mode 100644 index 44b074954d..0000000000 --- a/samples/timm/beit3_large_patch16_224.indomain_pt/graph_hash.txt +++ /dev/null @@ -1 +0,0 @@ -a79caac738b040f293c01cd92e1770f312f059371fd757fa138f608c37b3cf6e \ No newline at end of file diff --git a/samples/timm/beit3_large_patch16_224.indomain_pt/model.py b/samples/timm/beit3_large_patch16_224.indomain_pt/model.py deleted file mode 100644 index 59f9368528..0000000000 --- a/samples/timm/beit3_large_patch16_224.indomain_pt/model.py +++ /dev/null @@ -1,4234 +0,0 @@ -import torch - - -class GraphModule(torch.nn.Module): - def forward( - self, - L_x_: torch.Tensor, - L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, - L_self_parameters_cls_token_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_num_prefix_tokens: torch.SymInt, - L_self_modules_fc_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_fc_norm_parameters_bias_: torch.nn.parameter.Parameter, - ): - l_x_ = L_x_ - l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( - L_self_modules_patch_embed_modules_proj_parameters_weight_ - ) - l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( - L_self_modules_patch_embed_modules_proj_parameters_bias_ - ) - l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ - l_self_parameters_cls_token_ = L_self_parameters_cls_token_ - l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_num_prefix_tokens = L_self_num_prefix_tokens - l_self_modules_fc_norm_parameters_weight_ = ( - L_self_modules_fc_norm_parameters_weight_ - ) - l_self_modules_fc_norm_parameters_bias_ = ( - L_self_modules_fc_norm_parameters_bias_ - ) - x = torch.conv2d( - l_x_, - l_self_modules_patch_embed_modules_proj_parameters_weight_, - l_self_modules_patch_embed_modules_proj_parameters_bias_, - (16, 16), - (0, 0), - (1, 1), - 1, - ) - l_x_ = ( - l_self_modules_patch_embed_modules_proj_parameters_weight_ - ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None - flatten = x.flatten(2) - x = None - x_1 = flatten.transpose(1, 2) - flatten = None - expand = l_self_parameters_cls_token_.expand(1, -1, -1) - l_self_parameters_cls_token_ = None - x_2 = torch.cat([expand, x_1], dim=1) - expand = x_1 = None - x_3 = x_2 + l_self_parameters_pos_embed_ - x_2 = l_self_parameters_pos_embed_ = None - x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) - x_3 = None - x_5 = torch.nn.functional.layer_norm( - x_4, - (1024,), - l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ - ) = None - linear = torch._C._nn.linear( - x_5, - l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, - ) - x_5 = ( - l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape = linear.reshape(1, 197, 3, 16, 64) - linear = None - qkv = reshape.permute(2, 0, 3, 1, 4) - reshape = None - unbind = qkv.unbind(0) - qkv = None - q = unbind[0] - k = unbind[1] - v = unbind[2] - unbind = None - x_6 = torch._C._nn.scaled_dot_product_attention( - q, k, v, attn_mask=None, dropout_p=0.0 - ) - q = k = v = None - transpose_1 = x_6.transpose(1, 2) - x_6 = None - x_7 = transpose_1.reshape(1, 197, 1024) - transpose_1 = None - x_8 = torch.nn.functional.layer_norm( - x_7, - (1024,), - l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_7 = ( - l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_ - ) = None - x_9 = torch._C._nn.linear( - x_8, - l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, - ) - x_8 = ( - l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ - ) = None - x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) - x_9 = None - x_11 = x_4 + x_10 - x_4 = x_10 = None - x_12 = torch.nn.functional.layer_norm( - x_11, - (1024,), - l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ - ) = None - x_13 = torch._C._nn.linear( - x_12, - l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, - ) - x_12 = ( - l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_14 = torch._C._nn.gelu(x_13, approximate="none") - x_13 = None - x_15 = torch.nn.functional.dropout(x_14, 0.0, False, False) - x_14 = None - x_16 = torch.nn.functional.layer_norm( - x_15, - (4096,), - l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_15 = ( - l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_17 = torch._C._nn.linear( - x_16, - l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, - ) - x_16 = ( - l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_18 = torch.nn.functional.dropout(x_17, 0.0, False, False) - x_17 = None - x_19 = x_11 + x_18 - x_11 = x_18 = None - x_20 = torch.nn.functional.layer_norm( - x_19, - (1024,), - l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ - ) = None - linear_4 = torch._C._nn.linear( - x_20, - l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, - ) - x_20 = ( - l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_2 = linear_4.reshape(1, 197, 3, 16, 64) - linear_4 = None - qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) - reshape_2 = None - unbind_1 = qkv_1.unbind(0) - qkv_1 = None - q_1 = unbind_1[0] - k_1 = unbind_1[1] - v_1 = unbind_1[2] - unbind_1 = None - x_21 = torch._C._nn.scaled_dot_product_attention( - q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 - ) - q_1 = k_1 = v_1 = None - transpose_2 = x_21.transpose(1, 2) - x_21 = None - x_22 = transpose_2.reshape(1, 197, 1024) - transpose_2 = None - x_23 = torch.nn.functional.layer_norm( - x_22, - (1024,), - l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_22 = ( - l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_ - ) = None - x_24 = torch._C._nn.linear( - x_23, - l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, - ) - x_23 = ( - l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ - ) = None - x_25 = torch.nn.functional.dropout(x_24, 0.0, False, False) - x_24 = None - x_26 = x_19 + x_25 - x_19 = x_25 = None - x_27 = torch.nn.functional.layer_norm( - x_26, - (1024,), - l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ - ) = None - x_28 = torch._C._nn.linear( - x_27, - l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, - ) - x_27 = ( - l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_29 = torch._C._nn.gelu(x_28, approximate="none") - x_28 = None - x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) - x_29 = None - x_31 = torch.nn.functional.layer_norm( - x_30, - (4096,), - l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_30 = ( - l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_32 = torch._C._nn.linear( - x_31, - l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, - ) - x_31 = ( - l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) - x_32 = None - x_34 = x_26 + x_33 - x_26 = x_33 = None - x_35 = torch.nn.functional.layer_norm( - x_34, - (1024,), - l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ - ) = None - linear_8 = torch._C._nn.linear( - x_35, - l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, - ) - x_35 = ( - l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_4 = linear_8.reshape(1, 197, 3, 16, 64) - linear_8 = None - qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) - reshape_4 = None - unbind_2 = qkv_2.unbind(0) - qkv_2 = None - q_2 = unbind_2[0] - k_2 = unbind_2[1] - v_2 = unbind_2[2] - unbind_2 = None - x_36 = torch._C._nn.scaled_dot_product_attention( - q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 - ) - q_2 = k_2 = v_2 = None - transpose_3 = x_36.transpose(1, 2) - x_36 = None - x_37 = transpose_3.reshape(1, 197, 1024) - transpose_3 = None - x_38 = torch.nn.functional.layer_norm( - x_37, - (1024,), - l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_37 = ( - l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_ - ) = None - x_39 = torch._C._nn.linear( - x_38, - l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, - ) - x_38 = ( - l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ - ) = None - x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) - x_39 = None - x_41 = x_34 + x_40 - x_34 = x_40 = None - x_42 = torch.nn.functional.layer_norm( - x_41, - (1024,), - l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ - ) = None - x_43 = torch._C._nn.linear( - x_42, - l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, - ) - x_42 = ( - l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_44 = torch._C._nn.gelu(x_43, approximate="none") - x_43 = None - x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) - x_44 = None - x_46 = torch.nn.functional.layer_norm( - x_45, - (4096,), - l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_45 = ( - l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_47 = torch._C._nn.linear( - x_46, - l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, - ) - x_46 = ( - l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) - x_47 = None - x_49 = x_41 + x_48 - x_41 = x_48 = None - x_50 = torch.nn.functional.layer_norm( - x_49, - (1024,), - l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ - ) = None - linear_12 = torch._C._nn.linear( - x_50, - l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, - ) - x_50 = ( - l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_6 = linear_12.reshape(1, 197, 3, 16, 64) - linear_12 = None - qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) - reshape_6 = None - unbind_3 = qkv_3.unbind(0) - qkv_3 = None - q_3 = unbind_3[0] - k_3 = unbind_3[1] - v_3 = unbind_3[2] - unbind_3 = None - x_51 = torch._C._nn.scaled_dot_product_attention( - q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 - ) - q_3 = k_3 = v_3 = None - transpose_4 = x_51.transpose(1, 2) - x_51 = None - x_52 = transpose_4.reshape(1, 197, 1024) - transpose_4 = None - x_53 = torch.nn.functional.layer_norm( - x_52, - (1024,), - l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_52 = ( - l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_ - ) = None - x_54 = torch._C._nn.linear( - x_53, - l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, - ) - x_53 = ( - l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ - ) = None - x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) - x_54 = None - x_56 = x_49 + x_55 - x_49 = x_55 = None - x_57 = torch.nn.functional.layer_norm( - x_56, - (1024,), - l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ - ) = None - x_58 = torch._C._nn.linear( - x_57, - l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, - ) - x_57 = ( - l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_59 = torch._C._nn.gelu(x_58, approximate="none") - x_58 = None - x_60 = torch.nn.functional.dropout(x_59, 0.0, False, False) - x_59 = None - x_61 = torch.nn.functional.layer_norm( - x_60, - (4096,), - l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_60 = ( - l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_62 = torch._C._nn.linear( - x_61, - l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, - ) - x_61 = ( - l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_63 = torch.nn.functional.dropout(x_62, 0.0, False, False) - x_62 = None - x_64 = x_56 + x_63 - x_56 = x_63 = None - x_65 = torch.nn.functional.layer_norm( - x_64, - (1024,), - l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ - ) = None - linear_16 = torch._C._nn.linear( - x_65, - l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, - ) - x_65 = ( - l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_8 = linear_16.reshape(1, 197, 3, 16, 64) - linear_16 = None - qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) - reshape_8 = None - unbind_4 = qkv_4.unbind(0) - qkv_4 = None - q_4 = unbind_4[0] - k_4 = unbind_4[1] - v_4 = unbind_4[2] - unbind_4 = None - x_66 = torch._C._nn.scaled_dot_product_attention( - q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 - ) - q_4 = k_4 = v_4 = None - transpose_5 = x_66.transpose(1, 2) - x_66 = None - x_67 = transpose_5.reshape(1, 197, 1024) - transpose_5 = None - x_68 = torch.nn.functional.layer_norm( - x_67, - (1024,), - l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_67 = ( - l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_ - ) = None - x_69 = torch._C._nn.linear( - x_68, - l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, - ) - x_68 = ( - l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ - ) = None - x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) - x_69 = None - x_71 = x_64 + x_70 - x_64 = x_70 = None - x_72 = torch.nn.functional.layer_norm( - x_71, - (1024,), - l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ - ) = None - x_73 = torch._C._nn.linear( - x_72, - l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, - ) - x_72 = ( - l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_74 = torch._C._nn.gelu(x_73, approximate="none") - x_73 = None - x_75 = torch.nn.functional.dropout(x_74, 0.0, False, False) - x_74 = None - x_76 = torch.nn.functional.layer_norm( - x_75, - (4096,), - l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_75 = ( - l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_77 = torch._C._nn.linear( - x_76, - l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, - ) - x_76 = ( - l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_78 = torch.nn.functional.dropout(x_77, 0.0, False, False) - x_77 = None - x_79 = x_71 + x_78 - x_71 = x_78 = None - x_80 = torch.nn.functional.layer_norm( - x_79, - (1024,), - l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ - ) = None - linear_20 = torch._C._nn.linear( - x_80, - l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, - ) - x_80 = ( - l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_10 = linear_20.reshape(1, 197, 3, 16, 64) - linear_20 = None - qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) - reshape_10 = None - unbind_5 = qkv_5.unbind(0) - qkv_5 = None - q_5 = unbind_5[0] - k_5 = unbind_5[1] - v_5 = unbind_5[2] - unbind_5 = None - x_81 = torch._C._nn.scaled_dot_product_attention( - q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 - ) - q_5 = k_5 = v_5 = None - transpose_6 = x_81.transpose(1, 2) - x_81 = None - x_82 = transpose_6.reshape(1, 197, 1024) - transpose_6 = None - x_83 = torch.nn.functional.layer_norm( - x_82, - (1024,), - l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_82 = ( - l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_ - ) = None - x_84 = torch._C._nn.linear( - x_83, - l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, - ) - x_83 = ( - l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ - ) = None - x_85 = torch.nn.functional.dropout(x_84, 0.0, False, False) - x_84 = None - x_86 = x_79 + x_85 - x_79 = x_85 = None - x_87 = torch.nn.functional.layer_norm( - x_86, - (1024,), - l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ - ) = None - x_88 = torch._C._nn.linear( - x_87, - l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, - ) - x_87 = ( - l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_89 = torch._C._nn.gelu(x_88, approximate="none") - x_88 = None - x_90 = torch.nn.functional.dropout(x_89, 0.0, False, False) - x_89 = None - x_91 = torch.nn.functional.layer_norm( - x_90, - (4096,), - l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_90 = ( - l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_92 = torch._C._nn.linear( - x_91, - l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, - ) - x_91 = ( - l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_93 = torch.nn.functional.dropout(x_92, 0.0, False, False) - x_92 = None - x_94 = x_86 + x_93 - x_86 = x_93 = None - x_95 = torch.nn.functional.layer_norm( - x_94, - (1024,), - l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ - ) = None - linear_24 = torch._C._nn.linear( - x_95, - l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, - ) - x_95 = ( - l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_12 = linear_24.reshape(1, 197, 3, 16, 64) - linear_24 = None - qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) - reshape_12 = None - unbind_6 = qkv_6.unbind(0) - qkv_6 = None - q_6 = unbind_6[0] - k_6 = unbind_6[1] - v_6 = unbind_6[2] - unbind_6 = None - x_96 = torch._C._nn.scaled_dot_product_attention( - q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 - ) - q_6 = k_6 = v_6 = None - transpose_7 = x_96.transpose(1, 2) - x_96 = None - x_97 = transpose_7.reshape(1, 197, 1024) - transpose_7 = None - x_98 = torch.nn.functional.layer_norm( - x_97, - (1024,), - l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_97 = ( - l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_ - ) = None - x_99 = torch._C._nn.linear( - x_98, - l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, - ) - x_98 = ( - l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ - ) = None - x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) - x_99 = None - x_101 = x_94 + x_100 - x_94 = x_100 = None - x_102 = torch.nn.functional.layer_norm( - x_101, - (1024,), - l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ - ) = None - x_103 = torch._C._nn.linear( - x_102, - l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, - ) - x_102 = ( - l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_104 = torch._C._nn.gelu(x_103, approximate="none") - x_103 = None - x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) - x_104 = None - x_106 = torch.nn.functional.layer_norm( - x_105, - (4096,), - l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_105 = ( - l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_107 = torch._C._nn.linear( - x_106, - l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, - ) - x_106 = ( - l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) - x_107 = None - x_109 = x_101 + x_108 - x_101 = x_108 = None - x_110 = torch.nn.functional.layer_norm( - x_109, - (1024,), - l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ - ) = None - linear_28 = torch._C._nn.linear( - x_110, - l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, - ) - x_110 = ( - l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_14 = linear_28.reshape(1, 197, 3, 16, 64) - linear_28 = None - qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) - reshape_14 = None - unbind_7 = qkv_7.unbind(0) - qkv_7 = None - q_7 = unbind_7[0] - k_7 = unbind_7[1] - v_7 = unbind_7[2] - unbind_7 = None - x_111 = torch._C._nn.scaled_dot_product_attention( - q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 - ) - q_7 = k_7 = v_7 = None - transpose_8 = x_111.transpose(1, 2) - x_111 = None - x_112 = transpose_8.reshape(1, 197, 1024) - transpose_8 = None - x_113 = torch.nn.functional.layer_norm( - x_112, - (1024,), - l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_112 = ( - l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_ - ) = None - x_114 = torch._C._nn.linear( - x_113, - l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, - ) - x_113 = ( - l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ - ) = None - x_115 = torch.nn.functional.dropout(x_114, 0.0, False, False) - x_114 = None - x_116 = x_109 + x_115 - x_109 = x_115 = None - x_117 = torch.nn.functional.layer_norm( - x_116, - (1024,), - l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ - ) = None - x_118 = torch._C._nn.linear( - x_117, - l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, - ) - x_117 = ( - l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_119 = torch._C._nn.gelu(x_118, approximate="none") - x_118 = None - x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) - x_119 = None - x_121 = torch.nn.functional.layer_norm( - x_120, - (4096,), - l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_120 = ( - l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_122 = torch._C._nn.linear( - x_121, - l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, - ) - x_121 = ( - l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) - x_122 = None - x_124 = x_116 + x_123 - x_116 = x_123 = None - x_125 = torch.nn.functional.layer_norm( - x_124, - (1024,), - l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ - ) = None - linear_32 = torch._C._nn.linear( - x_125, - l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, - ) - x_125 = ( - l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_16 = linear_32.reshape(1, 197, 3, 16, 64) - linear_32 = None - qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) - reshape_16 = None - unbind_8 = qkv_8.unbind(0) - qkv_8 = None - q_8 = unbind_8[0] - k_8 = unbind_8[1] - v_8 = unbind_8[2] - unbind_8 = None - x_126 = torch._C._nn.scaled_dot_product_attention( - q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 - ) - q_8 = k_8 = v_8 = None - transpose_9 = x_126.transpose(1, 2) - x_126 = None - x_127 = transpose_9.reshape(1, 197, 1024) - transpose_9 = None - x_128 = torch.nn.functional.layer_norm( - x_127, - (1024,), - l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_127 = ( - l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_ - ) = None - x_129 = torch._C._nn.linear( - x_128, - l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, - ) - x_128 = ( - l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ - ) = None - x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) - x_129 = None - x_131 = x_124 + x_130 - x_124 = x_130 = None - x_132 = torch.nn.functional.layer_norm( - x_131, - (1024,), - l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ - ) = None - x_133 = torch._C._nn.linear( - x_132, - l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, - ) - x_132 = ( - l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_134 = torch._C._nn.gelu(x_133, approximate="none") - x_133 = None - x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) - x_134 = None - x_136 = torch.nn.functional.layer_norm( - x_135, - (4096,), - l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_135 = ( - l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_137 = torch._C._nn.linear( - x_136, - l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, - ) - x_136 = ( - l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) - x_137 = None - x_139 = x_131 + x_138 - x_131 = x_138 = None - x_140 = torch.nn.functional.layer_norm( - x_139, - (1024,), - l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ - ) = None - linear_36 = torch._C._nn.linear( - x_140, - l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, - ) - x_140 = ( - l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_18 = linear_36.reshape(1, 197, 3, 16, 64) - linear_36 = None - qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) - reshape_18 = None - unbind_9 = qkv_9.unbind(0) - qkv_9 = None - q_9 = unbind_9[0] - k_9 = unbind_9[1] - v_9 = unbind_9[2] - unbind_9 = None - x_141 = torch._C._nn.scaled_dot_product_attention( - q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 - ) - q_9 = k_9 = v_9 = None - transpose_10 = x_141.transpose(1, 2) - x_141 = None - x_142 = transpose_10.reshape(1, 197, 1024) - transpose_10 = None - x_143 = torch.nn.functional.layer_norm( - x_142, - (1024,), - l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_142 = ( - l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_ - ) = None - x_144 = torch._C._nn.linear( - x_143, - l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, - ) - x_143 = ( - l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ - ) = None - x_145 = torch.nn.functional.dropout(x_144, 0.0, False, False) - x_144 = None - x_146 = x_139 + x_145 - x_139 = x_145 = None - x_147 = torch.nn.functional.layer_norm( - x_146, - (1024,), - l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ - ) = None - x_148 = torch._C._nn.linear( - x_147, - l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, - ) - x_147 = ( - l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_149 = torch._C._nn.gelu(x_148, approximate="none") - x_148 = None - x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) - x_149 = None - x_151 = torch.nn.functional.layer_norm( - x_150, - (4096,), - l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_150 = ( - l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_152 = torch._C._nn.linear( - x_151, - l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, - ) - x_151 = ( - l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_153 = torch.nn.functional.dropout(x_152, 0.0, False, False) - x_152 = None - x_154 = x_146 + x_153 - x_146 = x_153 = None - x_155 = torch.nn.functional.layer_norm( - x_154, - (1024,), - l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ - ) = None - linear_40 = torch._C._nn.linear( - x_155, - l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, - ) - x_155 = ( - l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_20 = linear_40.reshape(1, 197, 3, 16, 64) - linear_40 = None - qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) - reshape_20 = None - unbind_10 = qkv_10.unbind(0) - qkv_10 = None - q_10 = unbind_10[0] - k_10 = unbind_10[1] - v_10 = unbind_10[2] - unbind_10 = None - x_156 = torch._C._nn.scaled_dot_product_attention( - q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 - ) - q_10 = k_10 = v_10 = None - transpose_11 = x_156.transpose(1, 2) - x_156 = None - x_157 = transpose_11.reshape(1, 197, 1024) - transpose_11 = None - x_158 = torch.nn.functional.layer_norm( - x_157, - (1024,), - l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_157 = l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_ - ) = None - x_159 = torch._C._nn.linear( - x_158, - l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, - ) - x_158 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ - ) = None - x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) - x_159 = None - x_161 = x_154 + x_160 - x_154 = x_160 = None - x_162 = torch.nn.functional.layer_norm( - x_161, - (1024,), - l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ - ) = None - x_163 = torch._C._nn.linear( - x_162, - l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, - ) - x_162 = ( - l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_164 = torch._C._nn.gelu(x_163, approximate="none") - x_163 = None - x_165 = torch.nn.functional.dropout(x_164, 0.0, False, False) - x_164 = None - x_166 = torch.nn.functional.layer_norm( - x_165, - (4096,), - l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_165 = ( - l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_167 = torch._C._nn.linear( - x_166, - l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, - ) - x_166 = ( - l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) - x_167 = None - x_169 = x_161 + x_168 - x_161 = x_168 = None - x_170 = torch.nn.functional.layer_norm( - x_169, - (1024,), - l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ - ) = None - linear_44 = torch._C._nn.linear( - x_170, - l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, - ) - x_170 = ( - l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_22 = linear_44.reshape(1, 197, 3, 16, 64) - linear_44 = None - qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) - reshape_22 = None - unbind_11 = qkv_11.unbind(0) - qkv_11 = None - q_11 = unbind_11[0] - k_11 = unbind_11[1] - v_11 = unbind_11[2] - unbind_11 = None - x_171 = torch._C._nn.scaled_dot_product_attention( - q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 - ) - q_11 = k_11 = v_11 = None - transpose_12 = x_171.transpose(1, 2) - x_171 = None - x_172 = transpose_12.reshape(1, 197, 1024) - transpose_12 = None - x_173 = torch.nn.functional.layer_norm( - x_172, - (1024,), - l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_172 = l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_ - ) = None - x_174 = torch._C._nn.linear( - x_173, - l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, - ) - x_173 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ - ) = None - x_175 = torch.nn.functional.dropout(x_174, 0.0, False, False) - x_174 = None - x_176 = x_169 + x_175 - x_169 = x_175 = None - x_177 = torch.nn.functional.layer_norm( - x_176, - (1024,), - l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ - ) = None - x_178 = torch._C._nn.linear( - x_177, - l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, - ) - x_177 = ( - l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_179 = torch._C._nn.gelu(x_178, approximate="none") - x_178 = None - x_180 = torch.nn.functional.dropout(x_179, 0.0, False, False) - x_179 = None - x_181 = torch.nn.functional.layer_norm( - x_180, - (4096,), - l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_180 = ( - l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_182 = torch._C._nn.linear( - x_181, - l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, - ) - x_181 = ( - l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_183 = torch.nn.functional.dropout(x_182, 0.0, False, False) - x_182 = None - x_184 = x_176 + x_183 - x_176 = x_183 = None - x_185 = torch.nn.functional.layer_norm( - x_184, - (1024,), - l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ - ) = None - linear_48 = torch._C._nn.linear( - x_185, - l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_, - ) - x_185 = ( - l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_24 = linear_48.reshape(1, 197, 3, 16, 64) - linear_48 = None - qkv_12 = reshape_24.permute(2, 0, 3, 1, 4) - reshape_24 = None - unbind_12 = qkv_12.unbind(0) - qkv_12 = None - q_12 = unbind_12[0] - k_12 = unbind_12[1] - v_12 = unbind_12[2] - unbind_12 = None - x_186 = torch._C._nn.scaled_dot_product_attention( - q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 - ) - q_12 = k_12 = v_12 = None - transpose_13 = x_186.transpose(1, 2) - x_186 = None - x_187 = transpose_13.reshape(1, 197, 1024) - transpose_13 = None - x_188 = torch.nn.functional.layer_norm( - x_187, - (1024,), - l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_187 = l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_ - ) = None - x_189 = torch._C._nn.linear( - x_188, - l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_, - ) - x_188 = l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ - ) = None - x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) - x_189 = None - x_191 = x_184 + x_190 - x_184 = x_190 = None - x_192 = torch.nn.functional.layer_norm( - x_191, - (1024,), - l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ - ) = None - x_193 = torch._C._nn.linear( - x_192, - l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, - ) - x_192 = ( - l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_194 = torch._C._nn.gelu(x_193, approximate="none") - x_193 = None - x_195 = torch.nn.functional.dropout(x_194, 0.0, False, False) - x_194 = None - x_196 = torch.nn.functional.layer_norm( - x_195, - (4096,), - l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_195 = ( - l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_197 = torch._C._nn.linear( - x_196, - l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, - ) - x_196 = ( - l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_198 = torch.nn.functional.dropout(x_197, 0.0, False, False) - x_197 = None - x_199 = x_191 + x_198 - x_191 = x_198 = None - x_200 = torch.nn.functional.layer_norm( - x_199, - (1024,), - l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ - ) = None - linear_52 = torch._C._nn.linear( - x_200, - l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_, - ) - x_200 = ( - l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_26 = linear_52.reshape(1, 197, 3, 16, 64) - linear_52 = None - qkv_13 = reshape_26.permute(2, 0, 3, 1, 4) - reshape_26 = None - unbind_13 = qkv_13.unbind(0) - qkv_13 = None - q_13 = unbind_13[0] - k_13 = unbind_13[1] - v_13 = unbind_13[2] - unbind_13 = None - x_201 = torch._C._nn.scaled_dot_product_attention( - q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 - ) - q_13 = k_13 = v_13 = None - transpose_14 = x_201.transpose(1, 2) - x_201 = None - x_202 = transpose_14.reshape(1, 197, 1024) - transpose_14 = None - x_203 = torch.nn.functional.layer_norm( - x_202, - (1024,), - l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_202 = l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_ - ) = None - x_204 = torch._C._nn.linear( - x_203, - l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_, - ) - x_203 = l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ - ) = None - x_205 = torch.nn.functional.dropout(x_204, 0.0, False, False) - x_204 = None - x_206 = x_199 + x_205 - x_199 = x_205 = None - x_207 = torch.nn.functional.layer_norm( - x_206, - (1024,), - l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ - ) = None - x_208 = torch._C._nn.linear( - x_207, - l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, - ) - x_207 = ( - l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_209 = torch._C._nn.gelu(x_208, approximate="none") - x_208 = None - x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) - x_209 = None - x_211 = torch.nn.functional.layer_norm( - x_210, - (4096,), - l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_210 = ( - l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_212 = torch._C._nn.linear( - x_211, - l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, - ) - x_211 = ( - l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_213 = torch.nn.functional.dropout(x_212, 0.0, False, False) - x_212 = None - x_214 = x_206 + x_213 - x_206 = x_213 = None - x_215 = torch.nn.functional.layer_norm( - x_214, - (1024,), - l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ - ) = None - linear_56 = torch._C._nn.linear( - x_215, - l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_, - ) - x_215 = ( - l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_28 = linear_56.reshape(1, 197, 3, 16, 64) - linear_56 = None - qkv_14 = reshape_28.permute(2, 0, 3, 1, 4) - reshape_28 = None - unbind_14 = qkv_14.unbind(0) - qkv_14 = None - q_14 = unbind_14[0] - k_14 = unbind_14[1] - v_14 = unbind_14[2] - unbind_14 = None - x_216 = torch._C._nn.scaled_dot_product_attention( - q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 - ) - q_14 = k_14 = v_14 = None - transpose_15 = x_216.transpose(1, 2) - x_216 = None - x_217 = transpose_15.reshape(1, 197, 1024) - transpose_15 = None - x_218 = torch.nn.functional.layer_norm( - x_217, - (1024,), - l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_217 = l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_ - ) = None - x_219 = torch._C._nn.linear( - x_218, - l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_, - ) - x_218 = l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ - ) = None - x_220 = torch.nn.functional.dropout(x_219, 0.0, False, False) - x_219 = None - x_221 = x_214 + x_220 - x_214 = x_220 = None - x_222 = torch.nn.functional.layer_norm( - x_221, - (1024,), - l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ - ) = None - x_223 = torch._C._nn.linear( - x_222, - l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, - ) - x_222 = ( - l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_224 = torch._C._nn.gelu(x_223, approximate="none") - x_223 = None - x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) - x_224 = None - x_226 = torch.nn.functional.layer_norm( - x_225, - (4096,), - l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_225 = ( - l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_227 = torch._C._nn.linear( - x_226, - l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, - ) - x_226 = ( - l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_228 = torch.nn.functional.dropout(x_227, 0.0, False, False) - x_227 = None - x_229 = x_221 + x_228 - x_221 = x_228 = None - x_230 = torch.nn.functional.layer_norm( - x_229, - (1024,), - l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ - ) = None - linear_60 = torch._C._nn.linear( - x_230, - l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_, - ) - x_230 = ( - l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_30 = linear_60.reshape(1, 197, 3, 16, 64) - linear_60 = None - qkv_15 = reshape_30.permute(2, 0, 3, 1, 4) - reshape_30 = None - unbind_15 = qkv_15.unbind(0) - qkv_15 = None - q_15 = unbind_15[0] - k_15 = unbind_15[1] - v_15 = unbind_15[2] - unbind_15 = None - x_231 = torch._C._nn.scaled_dot_product_attention( - q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 - ) - q_15 = k_15 = v_15 = None - transpose_16 = x_231.transpose(1, 2) - x_231 = None - x_232 = transpose_16.reshape(1, 197, 1024) - transpose_16 = None - x_233 = torch.nn.functional.layer_norm( - x_232, - (1024,), - l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_232 = l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_ - ) = None - x_234 = torch._C._nn.linear( - x_233, - l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_, - ) - x_233 = l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ - ) = None - x_235 = torch.nn.functional.dropout(x_234, 0.0, False, False) - x_234 = None - x_236 = x_229 + x_235 - x_229 = x_235 = None - x_237 = torch.nn.functional.layer_norm( - x_236, - (1024,), - l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ - ) = None - x_238 = torch._C._nn.linear( - x_237, - l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, - ) - x_237 = ( - l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_239 = torch._C._nn.gelu(x_238, approximate="none") - x_238 = None - x_240 = torch.nn.functional.dropout(x_239, 0.0, False, False) - x_239 = None - x_241 = torch.nn.functional.layer_norm( - x_240, - (4096,), - l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_240 = ( - l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_242 = torch._C._nn.linear( - x_241, - l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, - ) - x_241 = ( - l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_243 = torch.nn.functional.dropout(x_242, 0.0, False, False) - x_242 = None - x_244 = x_236 + x_243 - x_236 = x_243 = None - x_245 = torch.nn.functional.layer_norm( - x_244, - (1024,), - l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ - ) = None - linear_64 = torch._C._nn.linear( - x_245, - l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_, - ) - x_245 = ( - l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_32 = linear_64.reshape(1, 197, 3, 16, 64) - linear_64 = None - qkv_16 = reshape_32.permute(2, 0, 3, 1, 4) - reshape_32 = None - unbind_16 = qkv_16.unbind(0) - qkv_16 = None - q_16 = unbind_16[0] - k_16 = unbind_16[1] - v_16 = unbind_16[2] - unbind_16 = None - x_246 = torch._C._nn.scaled_dot_product_attention( - q_16, k_16, v_16, attn_mask=None, dropout_p=0.0 - ) - q_16 = k_16 = v_16 = None - transpose_17 = x_246.transpose(1, 2) - x_246 = None - x_247 = transpose_17.reshape(1, 197, 1024) - transpose_17 = None - x_248 = torch.nn.functional.layer_norm( - x_247, - (1024,), - l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_247 = l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_ - ) = None - x_249 = torch._C._nn.linear( - x_248, - l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_, - ) - x_248 = l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ - ) = None - x_250 = torch.nn.functional.dropout(x_249, 0.0, False, False) - x_249 = None - x_251 = x_244 + x_250 - x_244 = x_250 = None - x_252 = torch.nn.functional.layer_norm( - x_251, - (1024,), - l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ - ) = None - x_253 = torch._C._nn.linear( - x_252, - l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, - ) - x_252 = ( - l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_254 = torch._C._nn.gelu(x_253, approximate="none") - x_253 = None - x_255 = torch.nn.functional.dropout(x_254, 0.0, False, False) - x_254 = None - x_256 = torch.nn.functional.layer_norm( - x_255, - (4096,), - l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_255 = ( - l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_257 = torch._C._nn.linear( - x_256, - l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, - ) - x_256 = ( - l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_258 = torch.nn.functional.dropout(x_257, 0.0, False, False) - x_257 = None - x_259 = x_251 + x_258 - x_251 = x_258 = None - x_260 = torch.nn.functional.layer_norm( - x_259, - (1024,), - l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ - ) = None - linear_68 = torch._C._nn.linear( - x_260, - l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_, - ) - x_260 = ( - l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_34 = linear_68.reshape(1, 197, 3, 16, 64) - linear_68 = None - qkv_17 = reshape_34.permute(2, 0, 3, 1, 4) - reshape_34 = None - unbind_17 = qkv_17.unbind(0) - qkv_17 = None - q_17 = unbind_17[0] - k_17 = unbind_17[1] - v_17 = unbind_17[2] - unbind_17 = None - x_261 = torch._C._nn.scaled_dot_product_attention( - q_17, k_17, v_17, attn_mask=None, dropout_p=0.0 - ) - q_17 = k_17 = v_17 = None - transpose_18 = x_261.transpose(1, 2) - x_261 = None - x_262 = transpose_18.reshape(1, 197, 1024) - transpose_18 = None - x_263 = torch.nn.functional.layer_norm( - x_262, - (1024,), - l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_262 = l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_ - ) = None - x_264 = torch._C._nn.linear( - x_263, - l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_, - ) - x_263 = l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ - ) = None - x_265 = torch.nn.functional.dropout(x_264, 0.0, False, False) - x_264 = None - x_266 = x_259 + x_265 - x_259 = x_265 = None - x_267 = torch.nn.functional.layer_norm( - x_266, - (1024,), - l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ - ) = None - x_268 = torch._C._nn.linear( - x_267, - l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, - ) - x_267 = ( - l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_269 = torch._C._nn.gelu(x_268, approximate="none") - x_268 = None - x_270 = torch.nn.functional.dropout(x_269, 0.0, False, False) - x_269 = None - x_271 = torch.nn.functional.layer_norm( - x_270, - (4096,), - l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_270 = ( - l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_272 = torch._C._nn.linear( - x_271, - l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, - ) - x_271 = ( - l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_273 = torch.nn.functional.dropout(x_272, 0.0, False, False) - x_272 = None - x_274 = x_266 + x_273 - x_266 = x_273 = None - x_275 = torch.nn.functional.layer_norm( - x_274, - (1024,), - l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ - ) = None - linear_72 = torch._C._nn.linear( - x_275, - l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_, - ) - x_275 = ( - l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_36 = linear_72.reshape(1, 197, 3, 16, 64) - linear_72 = None - qkv_18 = reshape_36.permute(2, 0, 3, 1, 4) - reshape_36 = None - unbind_18 = qkv_18.unbind(0) - qkv_18 = None - q_18 = unbind_18[0] - k_18 = unbind_18[1] - v_18 = unbind_18[2] - unbind_18 = None - x_276 = torch._C._nn.scaled_dot_product_attention( - q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 - ) - q_18 = k_18 = v_18 = None - transpose_19 = x_276.transpose(1, 2) - x_276 = None - x_277 = transpose_19.reshape(1, 197, 1024) - transpose_19 = None - x_278 = torch.nn.functional.layer_norm( - x_277, - (1024,), - l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_277 = l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_ - ) = None - x_279 = torch._C._nn.linear( - x_278, - l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_, - ) - x_278 = l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ - ) = None - x_280 = torch.nn.functional.dropout(x_279, 0.0, False, False) - x_279 = None - x_281 = x_274 + x_280 - x_274 = x_280 = None - x_282 = torch.nn.functional.layer_norm( - x_281, - (1024,), - l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ - ) = None - x_283 = torch._C._nn.linear( - x_282, - l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, - ) - x_282 = ( - l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_284 = torch._C._nn.gelu(x_283, approximate="none") - x_283 = None - x_285 = torch.nn.functional.dropout(x_284, 0.0, False, False) - x_284 = None - x_286 = torch.nn.functional.layer_norm( - x_285, - (4096,), - l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_285 = ( - l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_287 = torch._C._nn.linear( - x_286, - l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, - ) - x_286 = ( - l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) - x_287 = None - x_289 = x_281 + x_288 - x_281 = x_288 = None - x_290 = torch.nn.functional.layer_norm( - x_289, - (1024,), - l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ - ) = None - linear_76 = torch._C._nn.linear( - x_290, - l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_, - ) - x_290 = ( - l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_38 = linear_76.reshape(1, 197, 3, 16, 64) - linear_76 = None - qkv_19 = reshape_38.permute(2, 0, 3, 1, 4) - reshape_38 = None - unbind_19 = qkv_19.unbind(0) - qkv_19 = None - q_19 = unbind_19[0] - k_19 = unbind_19[1] - v_19 = unbind_19[2] - unbind_19 = None - x_291 = torch._C._nn.scaled_dot_product_attention( - q_19, k_19, v_19, attn_mask=None, dropout_p=0.0 - ) - q_19 = k_19 = v_19 = None - transpose_20 = x_291.transpose(1, 2) - x_291 = None - x_292 = transpose_20.reshape(1, 197, 1024) - transpose_20 = None - x_293 = torch.nn.functional.layer_norm( - x_292, - (1024,), - l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_292 = l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_ - ) = None - x_294 = torch._C._nn.linear( - x_293, - l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_, - ) - x_293 = l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ - ) = None - x_295 = torch.nn.functional.dropout(x_294, 0.0, False, False) - x_294 = None - x_296 = x_289 + x_295 - x_289 = x_295 = None - x_297 = torch.nn.functional.layer_norm( - x_296, - (1024,), - l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ - ) = None - x_298 = torch._C._nn.linear( - x_297, - l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, - ) - x_297 = ( - l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_299 = torch._C._nn.gelu(x_298, approximate="none") - x_298 = None - x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) - x_299 = None - x_301 = torch.nn.functional.layer_norm( - x_300, - (4096,), - l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_300 = ( - l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_302 = torch._C._nn.linear( - x_301, - l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, - ) - x_301 = ( - l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_303 = torch.nn.functional.dropout(x_302, 0.0, False, False) - x_302 = None - x_304 = x_296 + x_303 - x_296 = x_303 = None - x_305 = torch.nn.functional.layer_norm( - x_304, - (1024,), - l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ - ) = None - linear_80 = torch._C._nn.linear( - x_305, - l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_, - ) - x_305 = ( - l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_40 = linear_80.reshape(1, 197, 3, 16, 64) - linear_80 = None - qkv_20 = reshape_40.permute(2, 0, 3, 1, 4) - reshape_40 = None - unbind_20 = qkv_20.unbind(0) - qkv_20 = None - q_20 = unbind_20[0] - k_20 = unbind_20[1] - v_20 = unbind_20[2] - unbind_20 = None - x_306 = torch._C._nn.scaled_dot_product_attention( - q_20, k_20, v_20, attn_mask=None, dropout_p=0.0 - ) - q_20 = k_20 = v_20 = None - transpose_21 = x_306.transpose(1, 2) - x_306 = None - x_307 = transpose_21.reshape(1, 197, 1024) - transpose_21 = None - x_308 = torch.nn.functional.layer_norm( - x_307, - (1024,), - l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_307 = l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_ - ) = None - x_309 = torch._C._nn.linear( - x_308, - l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_, - ) - x_308 = l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ - ) = None - x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) - x_309 = None - x_311 = x_304 + x_310 - x_304 = x_310 = None - x_312 = torch.nn.functional.layer_norm( - x_311, - (1024,), - l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ - ) = None - x_313 = torch._C._nn.linear( - x_312, - l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, - ) - x_312 = ( - l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_314 = torch._C._nn.gelu(x_313, approximate="none") - x_313 = None - x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) - x_314 = None - x_316 = torch.nn.functional.layer_norm( - x_315, - (4096,), - l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_315 = ( - l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_317 = torch._C._nn.linear( - x_316, - l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, - ) - x_316 = ( - l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_318 = torch.nn.functional.dropout(x_317, 0.0, False, False) - x_317 = None - x_319 = x_311 + x_318 - x_311 = x_318 = None - x_320 = torch.nn.functional.layer_norm( - x_319, - (1024,), - l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ - ) = None - linear_84 = torch._C._nn.linear( - x_320, - l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_, - ) - x_320 = ( - l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_42 = linear_84.reshape(1, 197, 3, 16, 64) - linear_84 = None - qkv_21 = reshape_42.permute(2, 0, 3, 1, 4) - reshape_42 = None - unbind_21 = qkv_21.unbind(0) - qkv_21 = None - q_21 = unbind_21[0] - k_21 = unbind_21[1] - v_21 = unbind_21[2] - unbind_21 = None - x_321 = torch._C._nn.scaled_dot_product_attention( - q_21, k_21, v_21, attn_mask=None, dropout_p=0.0 - ) - q_21 = k_21 = v_21 = None - transpose_22 = x_321.transpose(1, 2) - x_321 = None - x_322 = transpose_22.reshape(1, 197, 1024) - transpose_22 = None - x_323 = torch.nn.functional.layer_norm( - x_322, - (1024,), - l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_322 = l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_ - ) = None - x_324 = torch._C._nn.linear( - x_323, - l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_, - ) - x_323 = l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ - ) = None - x_325 = torch.nn.functional.dropout(x_324, 0.0, False, False) - x_324 = None - x_326 = x_319 + x_325 - x_319 = x_325 = None - x_327 = torch.nn.functional.layer_norm( - x_326, - (1024,), - l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ - ) = None - x_328 = torch._C._nn.linear( - x_327, - l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, - ) - x_327 = ( - l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_329 = torch._C._nn.gelu(x_328, approximate="none") - x_328 = None - x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) - x_329 = None - x_331 = torch.nn.functional.layer_norm( - x_330, - (4096,), - l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_330 = ( - l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_332 = torch._C._nn.linear( - x_331, - l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, - ) - x_331 = ( - l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) - x_332 = None - x_334 = x_326 + x_333 - x_326 = x_333 = None - x_335 = torch.nn.functional.layer_norm( - x_334, - (1024,), - l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ - ) = None - linear_88 = torch._C._nn.linear( - x_335, - l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_, - ) - x_335 = ( - l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_44 = linear_88.reshape(1, 197, 3, 16, 64) - linear_88 = None - qkv_22 = reshape_44.permute(2, 0, 3, 1, 4) - reshape_44 = None - unbind_22 = qkv_22.unbind(0) - qkv_22 = None - q_22 = unbind_22[0] - k_22 = unbind_22[1] - v_22 = unbind_22[2] - unbind_22 = None - x_336 = torch._C._nn.scaled_dot_product_attention( - q_22, k_22, v_22, attn_mask=None, dropout_p=0.0 - ) - q_22 = k_22 = v_22 = None - transpose_23 = x_336.transpose(1, 2) - x_336 = None - x_337 = transpose_23.reshape(1, 197, 1024) - transpose_23 = None - x_338 = torch.nn.functional.layer_norm( - x_337, - (1024,), - l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_337 = l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_ - ) = None - x_339 = torch._C._nn.linear( - x_338, - l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_, - ) - x_338 = l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ - ) = None - x_340 = torch.nn.functional.dropout(x_339, 0.0, False, False) - x_339 = None - x_341 = x_334 + x_340 - x_334 = x_340 = None - x_342 = torch.nn.functional.layer_norm( - x_341, - (1024,), - l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ - ) = None - x_343 = torch._C._nn.linear( - x_342, - l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, - ) - x_342 = ( - l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_344 = torch._C._nn.gelu(x_343, approximate="none") - x_343 = None - x_345 = torch.nn.functional.dropout(x_344, 0.0, False, False) - x_344 = None - x_346 = torch.nn.functional.layer_norm( - x_345, - (4096,), - l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_345 = ( - l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_347 = torch._C._nn.linear( - x_346, - l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, - ) - x_346 = ( - l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) - x_347 = None - x_349 = x_341 + x_348 - x_341 = x_348 = None - x_350 = torch.nn.functional.layer_norm( - x_349, - (1024,), - l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ - ) = None - linear_92 = torch._C._nn.linear( - x_350, - l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_, - ) - x_350 = ( - l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_46 = linear_92.reshape(1, 197, 3, 16, 64) - linear_92 = None - qkv_23 = reshape_46.permute(2, 0, 3, 1, 4) - reshape_46 = None - unbind_23 = qkv_23.unbind(0) - qkv_23 = None - q_23 = unbind_23[0] - k_23 = unbind_23[1] - v_23 = unbind_23[2] - unbind_23 = None - x_351 = torch._C._nn.scaled_dot_product_attention( - q_23, k_23, v_23, attn_mask=None, dropout_p=0.0 - ) - q_23 = k_23 = v_23 = None - transpose_24 = x_351.transpose(1, 2) - x_351 = None - x_352 = transpose_24.reshape(1, 197, 1024) - transpose_24 = None - x_353 = torch.nn.functional.layer_norm( - x_352, - (1024,), - l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_352 = l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_ - ) = None - x_354 = torch._C._nn.linear( - x_353, - l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_, - ) - x_353 = l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ - ) = None - x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) - x_354 = None - x_356 = x_349 + x_355 - x_349 = x_355 = None - x_357 = torch.nn.functional.layer_norm( - x_356, - (1024,), - l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ - ) = None - x_358 = torch._C._nn.linear( - x_357, - l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, - ) - x_357 = ( - l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_359 = torch._C._nn.gelu(x_358, approximate="none") - x_358 = None - x_360 = torch.nn.functional.dropout(x_359, 0.0, False, False) - x_359 = None - x_361 = torch.nn.functional.layer_norm( - x_360, - (4096,), - l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_360 = ( - l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_362 = torch._C._nn.linear( - x_361, - l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, - ) - x_361 = ( - l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) - x_362 = None - x_364 = x_356 + x_363 - x_356 = x_363 = None - x_365 = x_364[ - (slice(None, None, None), slice(l_self_num_prefix_tokens, None, None)) - ] - x_364 = l_self_num_prefix_tokens = None - x_366 = x_365.mean(dim=1) - x_365 = None - x_367 = torch.nn.functional.layer_norm( - x_366, - (1024,), - l_self_modules_fc_norm_parameters_weight_, - l_self_modules_fc_norm_parameters_bias_, - 1e-05, - ) - x_366 = ( - l_self_modules_fc_norm_parameters_weight_ - ) = l_self_modules_fc_norm_parameters_bias_ = None - x_368 = torch.nn.functional.dropout(x_367, 0.0, False, False) - x_367 = None - return (x_368,) diff --git a/samples/timm/beit3_large_patch16_224.indomain_pt/weight_meta.py b/samples/timm/beit3_large_patch16_224.indomain_pt/weight_meta.py deleted file mode 100644 index 5bda823d60..0000000000 --- a/samples/timm/beit3_large_patch16_224.indomain_pt/weight_meta.py +++ /dev/null @@ -1,4070 +0,0 @@ -class Program_weight_tensor_meta_L_x_: - name = "L_x_" - shape = [1, 3, 224, 224] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.224 - std = 1.286 - data = None - - -class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: - name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" - shape = [1024, 3, 16, 16] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.021 - data = None - - -class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: - name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.021 - data = None - - -class Program_weight_tensor_meta_L_self_parameters_pos_embed_: - name = "L_self_parameters_pos_embed_" - shape = [1, 197, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_parameters_cls_token_: - name = "L_self_parameters_cls_token_" - shape = [1, 1, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_num_prefix_tokens: - name = "L_self_num_prefix_tokens" - shape = [] - dtype = "torch.int64" - device = "cpu" - mean = None - std = None - data = [4] - - -class Program_weight_tensor_meta_L_self_modules_fc_norm_parameters_weight_: - name = "L_self_modules_fc_norm_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_fc_norm_parameters_bias_: - name = "L_self_modules_fc_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None diff --git a/samples/timm/beit3_large_patch16_224.pt/graph_hash.txt b/samples/timm/beit3_large_patch16_224.pt/graph_hash.txt deleted file mode 100644 index 44b074954d..0000000000 --- a/samples/timm/beit3_large_patch16_224.pt/graph_hash.txt +++ /dev/null @@ -1 +0,0 @@ -a79caac738b040f293c01cd92e1770f312f059371fd757fa138f608c37b3cf6e \ No newline at end of file diff --git a/samples/timm/beit3_large_patch16_224.pt/model.py b/samples/timm/beit3_large_patch16_224.pt/model.py deleted file mode 100644 index 59f9368528..0000000000 --- a/samples/timm/beit3_large_patch16_224.pt/model.py +++ /dev/null @@ -1,4234 +0,0 @@ -import torch - - -class GraphModule(torch.nn.Module): - def forward( - self, - L_x_: torch.Tensor, - L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, - L_self_parameters_cls_token_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, - L_self_num_prefix_tokens: torch.SymInt, - L_self_modules_fc_norm_parameters_weight_: torch.nn.parameter.Parameter, - L_self_modules_fc_norm_parameters_bias_: torch.nn.parameter.Parameter, - ): - l_x_ = L_x_ - l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( - L_self_modules_patch_embed_modules_proj_parameters_weight_ - ) - l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( - L_self_modules_patch_embed_modules_proj_parameters_bias_ - ) - l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ - l_self_parameters_cls_token_ = L_self_parameters_cls_token_ - l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( - L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ - ) - l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( - L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ - ) - l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ = ( - L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ - ) - l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ = ( - L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ - ) - l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ = ( - L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ - ) - l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_ - l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ - l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ = ( - L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ - ) - l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( - L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ - ) - l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ = ( - L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ - ) - l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = ( - L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ - ) - l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = ( - L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ - ) - l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_ = ( - L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_ - ) - l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_ = ( - L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_ - ) - l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = ( - L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ - ) - l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = ( - L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ - ) - l_self_num_prefix_tokens = L_self_num_prefix_tokens - l_self_modules_fc_norm_parameters_weight_ = ( - L_self_modules_fc_norm_parameters_weight_ - ) - l_self_modules_fc_norm_parameters_bias_ = ( - L_self_modules_fc_norm_parameters_bias_ - ) - x = torch.conv2d( - l_x_, - l_self_modules_patch_embed_modules_proj_parameters_weight_, - l_self_modules_patch_embed_modules_proj_parameters_bias_, - (16, 16), - (0, 0), - (1, 1), - 1, - ) - l_x_ = ( - l_self_modules_patch_embed_modules_proj_parameters_weight_ - ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None - flatten = x.flatten(2) - x = None - x_1 = flatten.transpose(1, 2) - flatten = None - expand = l_self_parameters_cls_token_.expand(1, -1, -1) - l_self_parameters_cls_token_ = None - x_2 = torch.cat([expand, x_1], dim=1) - expand = x_1 = None - x_3 = x_2 + l_self_parameters_pos_embed_ - x_2 = l_self_parameters_pos_embed_ = None - x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) - x_3 = None - x_5 = torch.nn.functional.layer_norm( - x_4, - (1024,), - l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ - ) = None - linear = torch._C._nn.linear( - x_5, - l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, - ) - x_5 = ( - l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape = linear.reshape(1, 197, 3, 16, 64) - linear = None - qkv = reshape.permute(2, 0, 3, 1, 4) - reshape = None - unbind = qkv.unbind(0) - qkv = None - q = unbind[0] - k = unbind[1] - v = unbind[2] - unbind = None - x_6 = torch._C._nn.scaled_dot_product_attention( - q, k, v, attn_mask=None, dropout_p=0.0 - ) - q = k = v = None - transpose_1 = x_6.transpose(1, 2) - x_6 = None - x_7 = transpose_1.reshape(1, 197, 1024) - transpose_1 = None - x_8 = torch.nn.functional.layer_norm( - x_7, - (1024,), - l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_7 = ( - l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_ - ) = None - x_9 = torch._C._nn.linear( - x_8, - l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, - ) - x_8 = ( - l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ - ) = None - x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) - x_9 = None - x_11 = x_4 + x_10 - x_4 = x_10 = None - x_12 = torch.nn.functional.layer_norm( - x_11, - (1024,), - l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ - ) = None - x_13 = torch._C._nn.linear( - x_12, - l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, - ) - x_12 = ( - l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_14 = torch._C._nn.gelu(x_13, approximate="none") - x_13 = None - x_15 = torch.nn.functional.dropout(x_14, 0.0, False, False) - x_14 = None - x_16 = torch.nn.functional.layer_norm( - x_15, - (4096,), - l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_15 = ( - l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_17 = torch._C._nn.linear( - x_16, - l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, - ) - x_16 = ( - l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_18 = torch.nn.functional.dropout(x_17, 0.0, False, False) - x_17 = None - x_19 = x_11 + x_18 - x_11 = x_18 = None - x_20 = torch.nn.functional.layer_norm( - x_19, - (1024,), - l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ - ) = None - linear_4 = torch._C._nn.linear( - x_20, - l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, - ) - x_20 = ( - l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_2 = linear_4.reshape(1, 197, 3, 16, 64) - linear_4 = None - qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) - reshape_2 = None - unbind_1 = qkv_1.unbind(0) - qkv_1 = None - q_1 = unbind_1[0] - k_1 = unbind_1[1] - v_1 = unbind_1[2] - unbind_1 = None - x_21 = torch._C._nn.scaled_dot_product_attention( - q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 - ) - q_1 = k_1 = v_1 = None - transpose_2 = x_21.transpose(1, 2) - x_21 = None - x_22 = transpose_2.reshape(1, 197, 1024) - transpose_2 = None - x_23 = torch.nn.functional.layer_norm( - x_22, - (1024,), - l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_22 = ( - l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_ - ) = None - x_24 = torch._C._nn.linear( - x_23, - l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, - ) - x_23 = ( - l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ - ) = None - x_25 = torch.nn.functional.dropout(x_24, 0.0, False, False) - x_24 = None - x_26 = x_19 + x_25 - x_19 = x_25 = None - x_27 = torch.nn.functional.layer_norm( - x_26, - (1024,), - l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ - ) = None - x_28 = torch._C._nn.linear( - x_27, - l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, - ) - x_27 = ( - l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_29 = torch._C._nn.gelu(x_28, approximate="none") - x_28 = None - x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) - x_29 = None - x_31 = torch.nn.functional.layer_norm( - x_30, - (4096,), - l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_30 = ( - l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_32 = torch._C._nn.linear( - x_31, - l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, - ) - x_31 = ( - l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) - x_32 = None - x_34 = x_26 + x_33 - x_26 = x_33 = None - x_35 = torch.nn.functional.layer_norm( - x_34, - (1024,), - l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ - ) = None - linear_8 = torch._C._nn.linear( - x_35, - l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, - ) - x_35 = ( - l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_4 = linear_8.reshape(1, 197, 3, 16, 64) - linear_8 = None - qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) - reshape_4 = None - unbind_2 = qkv_2.unbind(0) - qkv_2 = None - q_2 = unbind_2[0] - k_2 = unbind_2[1] - v_2 = unbind_2[2] - unbind_2 = None - x_36 = torch._C._nn.scaled_dot_product_attention( - q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 - ) - q_2 = k_2 = v_2 = None - transpose_3 = x_36.transpose(1, 2) - x_36 = None - x_37 = transpose_3.reshape(1, 197, 1024) - transpose_3 = None - x_38 = torch.nn.functional.layer_norm( - x_37, - (1024,), - l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_37 = ( - l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_ - ) = None - x_39 = torch._C._nn.linear( - x_38, - l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, - ) - x_38 = ( - l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ - ) = None - x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) - x_39 = None - x_41 = x_34 + x_40 - x_34 = x_40 = None - x_42 = torch.nn.functional.layer_norm( - x_41, - (1024,), - l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ - ) = None - x_43 = torch._C._nn.linear( - x_42, - l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, - ) - x_42 = ( - l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_44 = torch._C._nn.gelu(x_43, approximate="none") - x_43 = None - x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) - x_44 = None - x_46 = torch.nn.functional.layer_norm( - x_45, - (4096,), - l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_45 = ( - l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_47 = torch._C._nn.linear( - x_46, - l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, - ) - x_46 = ( - l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) - x_47 = None - x_49 = x_41 + x_48 - x_41 = x_48 = None - x_50 = torch.nn.functional.layer_norm( - x_49, - (1024,), - l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ - ) = None - linear_12 = torch._C._nn.linear( - x_50, - l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, - ) - x_50 = ( - l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_6 = linear_12.reshape(1, 197, 3, 16, 64) - linear_12 = None - qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) - reshape_6 = None - unbind_3 = qkv_3.unbind(0) - qkv_3 = None - q_3 = unbind_3[0] - k_3 = unbind_3[1] - v_3 = unbind_3[2] - unbind_3 = None - x_51 = torch._C._nn.scaled_dot_product_attention( - q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 - ) - q_3 = k_3 = v_3 = None - transpose_4 = x_51.transpose(1, 2) - x_51 = None - x_52 = transpose_4.reshape(1, 197, 1024) - transpose_4 = None - x_53 = torch.nn.functional.layer_norm( - x_52, - (1024,), - l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_52 = ( - l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_ - ) = None - x_54 = torch._C._nn.linear( - x_53, - l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, - ) - x_53 = ( - l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ - ) = None - x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) - x_54 = None - x_56 = x_49 + x_55 - x_49 = x_55 = None - x_57 = torch.nn.functional.layer_norm( - x_56, - (1024,), - l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ - ) = None - x_58 = torch._C._nn.linear( - x_57, - l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, - ) - x_57 = ( - l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_59 = torch._C._nn.gelu(x_58, approximate="none") - x_58 = None - x_60 = torch.nn.functional.dropout(x_59, 0.0, False, False) - x_59 = None - x_61 = torch.nn.functional.layer_norm( - x_60, - (4096,), - l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_60 = ( - l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_62 = torch._C._nn.linear( - x_61, - l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, - ) - x_61 = ( - l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_63 = torch.nn.functional.dropout(x_62, 0.0, False, False) - x_62 = None - x_64 = x_56 + x_63 - x_56 = x_63 = None - x_65 = torch.nn.functional.layer_norm( - x_64, - (1024,), - l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ - ) = None - linear_16 = torch._C._nn.linear( - x_65, - l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, - ) - x_65 = ( - l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_8 = linear_16.reshape(1, 197, 3, 16, 64) - linear_16 = None - qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) - reshape_8 = None - unbind_4 = qkv_4.unbind(0) - qkv_4 = None - q_4 = unbind_4[0] - k_4 = unbind_4[1] - v_4 = unbind_4[2] - unbind_4 = None - x_66 = torch._C._nn.scaled_dot_product_attention( - q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 - ) - q_4 = k_4 = v_4 = None - transpose_5 = x_66.transpose(1, 2) - x_66 = None - x_67 = transpose_5.reshape(1, 197, 1024) - transpose_5 = None - x_68 = torch.nn.functional.layer_norm( - x_67, - (1024,), - l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_67 = ( - l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_ - ) = None - x_69 = torch._C._nn.linear( - x_68, - l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, - ) - x_68 = ( - l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ - ) = None - x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) - x_69 = None - x_71 = x_64 + x_70 - x_64 = x_70 = None - x_72 = torch.nn.functional.layer_norm( - x_71, - (1024,), - l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ - ) = None - x_73 = torch._C._nn.linear( - x_72, - l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, - ) - x_72 = ( - l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_74 = torch._C._nn.gelu(x_73, approximate="none") - x_73 = None - x_75 = torch.nn.functional.dropout(x_74, 0.0, False, False) - x_74 = None - x_76 = torch.nn.functional.layer_norm( - x_75, - (4096,), - l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_75 = ( - l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_77 = torch._C._nn.linear( - x_76, - l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, - ) - x_76 = ( - l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_78 = torch.nn.functional.dropout(x_77, 0.0, False, False) - x_77 = None - x_79 = x_71 + x_78 - x_71 = x_78 = None - x_80 = torch.nn.functional.layer_norm( - x_79, - (1024,), - l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ - ) = None - linear_20 = torch._C._nn.linear( - x_80, - l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, - ) - x_80 = ( - l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_10 = linear_20.reshape(1, 197, 3, 16, 64) - linear_20 = None - qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) - reshape_10 = None - unbind_5 = qkv_5.unbind(0) - qkv_5 = None - q_5 = unbind_5[0] - k_5 = unbind_5[1] - v_5 = unbind_5[2] - unbind_5 = None - x_81 = torch._C._nn.scaled_dot_product_attention( - q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 - ) - q_5 = k_5 = v_5 = None - transpose_6 = x_81.transpose(1, 2) - x_81 = None - x_82 = transpose_6.reshape(1, 197, 1024) - transpose_6 = None - x_83 = torch.nn.functional.layer_norm( - x_82, - (1024,), - l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_82 = ( - l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_ - ) = None - x_84 = torch._C._nn.linear( - x_83, - l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, - ) - x_83 = ( - l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ - ) = None - x_85 = torch.nn.functional.dropout(x_84, 0.0, False, False) - x_84 = None - x_86 = x_79 + x_85 - x_79 = x_85 = None - x_87 = torch.nn.functional.layer_norm( - x_86, - (1024,), - l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ - ) = None - x_88 = torch._C._nn.linear( - x_87, - l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, - ) - x_87 = ( - l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_89 = torch._C._nn.gelu(x_88, approximate="none") - x_88 = None - x_90 = torch.nn.functional.dropout(x_89, 0.0, False, False) - x_89 = None - x_91 = torch.nn.functional.layer_norm( - x_90, - (4096,), - l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_90 = ( - l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_92 = torch._C._nn.linear( - x_91, - l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, - ) - x_91 = ( - l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_93 = torch.nn.functional.dropout(x_92, 0.0, False, False) - x_92 = None - x_94 = x_86 + x_93 - x_86 = x_93 = None - x_95 = torch.nn.functional.layer_norm( - x_94, - (1024,), - l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ - ) = None - linear_24 = torch._C._nn.linear( - x_95, - l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, - ) - x_95 = ( - l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_12 = linear_24.reshape(1, 197, 3, 16, 64) - linear_24 = None - qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) - reshape_12 = None - unbind_6 = qkv_6.unbind(0) - qkv_6 = None - q_6 = unbind_6[0] - k_6 = unbind_6[1] - v_6 = unbind_6[2] - unbind_6 = None - x_96 = torch._C._nn.scaled_dot_product_attention( - q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 - ) - q_6 = k_6 = v_6 = None - transpose_7 = x_96.transpose(1, 2) - x_96 = None - x_97 = transpose_7.reshape(1, 197, 1024) - transpose_7 = None - x_98 = torch.nn.functional.layer_norm( - x_97, - (1024,), - l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_97 = ( - l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_ - ) = None - x_99 = torch._C._nn.linear( - x_98, - l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, - ) - x_98 = ( - l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ - ) = None - x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) - x_99 = None - x_101 = x_94 + x_100 - x_94 = x_100 = None - x_102 = torch.nn.functional.layer_norm( - x_101, - (1024,), - l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ - ) = None - x_103 = torch._C._nn.linear( - x_102, - l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, - ) - x_102 = ( - l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_104 = torch._C._nn.gelu(x_103, approximate="none") - x_103 = None - x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) - x_104 = None - x_106 = torch.nn.functional.layer_norm( - x_105, - (4096,), - l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_105 = ( - l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_107 = torch._C._nn.linear( - x_106, - l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, - ) - x_106 = ( - l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) - x_107 = None - x_109 = x_101 + x_108 - x_101 = x_108 = None - x_110 = torch.nn.functional.layer_norm( - x_109, - (1024,), - l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ - ) = None - linear_28 = torch._C._nn.linear( - x_110, - l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, - ) - x_110 = ( - l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_14 = linear_28.reshape(1, 197, 3, 16, 64) - linear_28 = None - qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) - reshape_14 = None - unbind_7 = qkv_7.unbind(0) - qkv_7 = None - q_7 = unbind_7[0] - k_7 = unbind_7[1] - v_7 = unbind_7[2] - unbind_7 = None - x_111 = torch._C._nn.scaled_dot_product_attention( - q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 - ) - q_7 = k_7 = v_7 = None - transpose_8 = x_111.transpose(1, 2) - x_111 = None - x_112 = transpose_8.reshape(1, 197, 1024) - transpose_8 = None - x_113 = torch.nn.functional.layer_norm( - x_112, - (1024,), - l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_112 = ( - l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_ - ) = None - x_114 = torch._C._nn.linear( - x_113, - l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, - ) - x_113 = ( - l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ - ) = None - x_115 = torch.nn.functional.dropout(x_114, 0.0, False, False) - x_114 = None - x_116 = x_109 + x_115 - x_109 = x_115 = None - x_117 = torch.nn.functional.layer_norm( - x_116, - (1024,), - l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ - ) = None - x_118 = torch._C._nn.linear( - x_117, - l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, - ) - x_117 = ( - l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_119 = torch._C._nn.gelu(x_118, approximate="none") - x_118 = None - x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) - x_119 = None - x_121 = torch.nn.functional.layer_norm( - x_120, - (4096,), - l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_120 = ( - l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_122 = torch._C._nn.linear( - x_121, - l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, - ) - x_121 = ( - l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) - x_122 = None - x_124 = x_116 + x_123 - x_116 = x_123 = None - x_125 = torch.nn.functional.layer_norm( - x_124, - (1024,), - l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ - ) = None - linear_32 = torch._C._nn.linear( - x_125, - l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, - ) - x_125 = ( - l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_16 = linear_32.reshape(1, 197, 3, 16, 64) - linear_32 = None - qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) - reshape_16 = None - unbind_8 = qkv_8.unbind(0) - qkv_8 = None - q_8 = unbind_8[0] - k_8 = unbind_8[1] - v_8 = unbind_8[2] - unbind_8 = None - x_126 = torch._C._nn.scaled_dot_product_attention( - q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 - ) - q_8 = k_8 = v_8 = None - transpose_9 = x_126.transpose(1, 2) - x_126 = None - x_127 = transpose_9.reshape(1, 197, 1024) - transpose_9 = None - x_128 = torch.nn.functional.layer_norm( - x_127, - (1024,), - l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_127 = ( - l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_ - ) = None - x_129 = torch._C._nn.linear( - x_128, - l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, - ) - x_128 = ( - l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ - ) = None - x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) - x_129 = None - x_131 = x_124 + x_130 - x_124 = x_130 = None - x_132 = torch.nn.functional.layer_norm( - x_131, - (1024,), - l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ - ) = None - x_133 = torch._C._nn.linear( - x_132, - l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, - ) - x_132 = ( - l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_134 = torch._C._nn.gelu(x_133, approximate="none") - x_133 = None - x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) - x_134 = None - x_136 = torch.nn.functional.layer_norm( - x_135, - (4096,), - l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_135 = ( - l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_137 = torch._C._nn.linear( - x_136, - l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, - ) - x_136 = ( - l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) - x_137 = None - x_139 = x_131 + x_138 - x_131 = x_138 = None - x_140 = torch.nn.functional.layer_norm( - x_139, - (1024,), - l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ - ) = None - linear_36 = torch._C._nn.linear( - x_140, - l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, - ) - x_140 = ( - l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_18 = linear_36.reshape(1, 197, 3, 16, 64) - linear_36 = None - qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) - reshape_18 = None - unbind_9 = qkv_9.unbind(0) - qkv_9 = None - q_9 = unbind_9[0] - k_9 = unbind_9[1] - v_9 = unbind_9[2] - unbind_9 = None - x_141 = torch._C._nn.scaled_dot_product_attention( - q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 - ) - q_9 = k_9 = v_9 = None - transpose_10 = x_141.transpose(1, 2) - x_141 = None - x_142 = transpose_10.reshape(1, 197, 1024) - transpose_10 = None - x_143 = torch.nn.functional.layer_norm( - x_142, - (1024,), - l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_142 = ( - l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_ - ) = None - x_144 = torch._C._nn.linear( - x_143, - l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, - ) - x_143 = ( - l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ - ) = None - x_145 = torch.nn.functional.dropout(x_144, 0.0, False, False) - x_144 = None - x_146 = x_139 + x_145 - x_139 = x_145 = None - x_147 = torch.nn.functional.layer_norm( - x_146, - (1024,), - l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ - ) = None - x_148 = torch._C._nn.linear( - x_147, - l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, - ) - x_147 = ( - l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_149 = torch._C._nn.gelu(x_148, approximate="none") - x_148 = None - x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) - x_149 = None - x_151 = torch.nn.functional.layer_norm( - x_150, - (4096,), - l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_150 = ( - l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_152 = torch._C._nn.linear( - x_151, - l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, - ) - x_151 = ( - l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_153 = torch.nn.functional.dropout(x_152, 0.0, False, False) - x_152 = None - x_154 = x_146 + x_153 - x_146 = x_153 = None - x_155 = torch.nn.functional.layer_norm( - x_154, - (1024,), - l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ - ) = None - linear_40 = torch._C._nn.linear( - x_155, - l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, - ) - x_155 = ( - l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_20 = linear_40.reshape(1, 197, 3, 16, 64) - linear_40 = None - qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) - reshape_20 = None - unbind_10 = qkv_10.unbind(0) - qkv_10 = None - q_10 = unbind_10[0] - k_10 = unbind_10[1] - v_10 = unbind_10[2] - unbind_10 = None - x_156 = torch._C._nn.scaled_dot_product_attention( - q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 - ) - q_10 = k_10 = v_10 = None - transpose_11 = x_156.transpose(1, 2) - x_156 = None - x_157 = transpose_11.reshape(1, 197, 1024) - transpose_11 = None - x_158 = torch.nn.functional.layer_norm( - x_157, - (1024,), - l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_157 = l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_ - ) = None - x_159 = torch._C._nn.linear( - x_158, - l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, - ) - x_158 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ - ) = None - x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) - x_159 = None - x_161 = x_154 + x_160 - x_154 = x_160 = None - x_162 = torch.nn.functional.layer_norm( - x_161, - (1024,), - l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ - ) = None - x_163 = torch._C._nn.linear( - x_162, - l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, - ) - x_162 = ( - l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_164 = torch._C._nn.gelu(x_163, approximate="none") - x_163 = None - x_165 = torch.nn.functional.dropout(x_164, 0.0, False, False) - x_164 = None - x_166 = torch.nn.functional.layer_norm( - x_165, - (4096,), - l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_165 = ( - l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_167 = torch._C._nn.linear( - x_166, - l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, - ) - x_166 = ( - l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) - x_167 = None - x_169 = x_161 + x_168 - x_161 = x_168 = None - x_170 = torch.nn.functional.layer_norm( - x_169, - (1024,), - l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ - ) = None - linear_44 = torch._C._nn.linear( - x_170, - l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, - ) - x_170 = ( - l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_22 = linear_44.reshape(1, 197, 3, 16, 64) - linear_44 = None - qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) - reshape_22 = None - unbind_11 = qkv_11.unbind(0) - qkv_11 = None - q_11 = unbind_11[0] - k_11 = unbind_11[1] - v_11 = unbind_11[2] - unbind_11 = None - x_171 = torch._C._nn.scaled_dot_product_attention( - q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 - ) - q_11 = k_11 = v_11 = None - transpose_12 = x_171.transpose(1, 2) - x_171 = None - x_172 = transpose_12.reshape(1, 197, 1024) - transpose_12 = None - x_173 = torch.nn.functional.layer_norm( - x_172, - (1024,), - l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_172 = l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_ - ) = None - x_174 = torch._C._nn.linear( - x_173, - l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, - ) - x_173 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ - ) = None - x_175 = torch.nn.functional.dropout(x_174, 0.0, False, False) - x_174 = None - x_176 = x_169 + x_175 - x_169 = x_175 = None - x_177 = torch.nn.functional.layer_norm( - x_176, - (1024,), - l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ - ) = None - x_178 = torch._C._nn.linear( - x_177, - l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, - ) - x_177 = ( - l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_179 = torch._C._nn.gelu(x_178, approximate="none") - x_178 = None - x_180 = torch.nn.functional.dropout(x_179, 0.0, False, False) - x_179 = None - x_181 = torch.nn.functional.layer_norm( - x_180, - (4096,), - l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_180 = ( - l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_182 = torch._C._nn.linear( - x_181, - l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, - ) - x_181 = ( - l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_183 = torch.nn.functional.dropout(x_182, 0.0, False, False) - x_182 = None - x_184 = x_176 + x_183 - x_176 = x_183 = None - x_185 = torch.nn.functional.layer_norm( - x_184, - (1024,), - l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ - ) = None - linear_48 = torch._C._nn.linear( - x_185, - l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_, - ) - x_185 = ( - l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_24 = linear_48.reshape(1, 197, 3, 16, 64) - linear_48 = None - qkv_12 = reshape_24.permute(2, 0, 3, 1, 4) - reshape_24 = None - unbind_12 = qkv_12.unbind(0) - qkv_12 = None - q_12 = unbind_12[0] - k_12 = unbind_12[1] - v_12 = unbind_12[2] - unbind_12 = None - x_186 = torch._C._nn.scaled_dot_product_attention( - q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 - ) - q_12 = k_12 = v_12 = None - transpose_13 = x_186.transpose(1, 2) - x_186 = None - x_187 = transpose_13.reshape(1, 197, 1024) - transpose_13 = None - x_188 = torch.nn.functional.layer_norm( - x_187, - (1024,), - l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_187 = l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_ - ) = None - x_189 = torch._C._nn.linear( - x_188, - l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_, - ) - x_188 = l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ - ) = None - x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) - x_189 = None - x_191 = x_184 + x_190 - x_184 = x_190 = None - x_192 = torch.nn.functional.layer_norm( - x_191, - (1024,), - l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ - ) = None - x_193 = torch._C._nn.linear( - x_192, - l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, - ) - x_192 = ( - l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_194 = torch._C._nn.gelu(x_193, approximate="none") - x_193 = None - x_195 = torch.nn.functional.dropout(x_194, 0.0, False, False) - x_194 = None - x_196 = torch.nn.functional.layer_norm( - x_195, - (4096,), - l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_195 = ( - l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_197 = torch._C._nn.linear( - x_196, - l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, - ) - x_196 = ( - l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_198 = torch.nn.functional.dropout(x_197, 0.0, False, False) - x_197 = None - x_199 = x_191 + x_198 - x_191 = x_198 = None - x_200 = torch.nn.functional.layer_norm( - x_199, - (1024,), - l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ - ) = None - linear_52 = torch._C._nn.linear( - x_200, - l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_, - ) - x_200 = ( - l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_26 = linear_52.reshape(1, 197, 3, 16, 64) - linear_52 = None - qkv_13 = reshape_26.permute(2, 0, 3, 1, 4) - reshape_26 = None - unbind_13 = qkv_13.unbind(0) - qkv_13 = None - q_13 = unbind_13[0] - k_13 = unbind_13[1] - v_13 = unbind_13[2] - unbind_13 = None - x_201 = torch._C._nn.scaled_dot_product_attention( - q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 - ) - q_13 = k_13 = v_13 = None - transpose_14 = x_201.transpose(1, 2) - x_201 = None - x_202 = transpose_14.reshape(1, 197, 1024) - transpose_14 = None - x_203 = torch.nn.functional.layer_norm( - x_202, - (1024,), - l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_202 = l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_ - ) = None - x_204 = torch._C._nn.linear( - x_203, - l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_, - ) - x_203 = l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ - ) = None - x_205 = torch.nn.functional.dropout(x_204, 0.0, False, False) - x_204 = None - x_206 = x_199 + x_205 - x_199 = x_205 = None - x_207 = torch.nn.functional.layer_norm( - x_206, - (1024,), - l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ - ) = None - x_208 = torch._C._nn.linear( - x_207, - l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, - ) - x_207 = ( - l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_209 = torch._C._nn.gelu(x_208, approximate="none") - x_208 = None - x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) - x_209 = None - x_211 = torch.nn.functional.layer_norm( - x_210, - (4096,), - l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_210 = ( - l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_212 = torch._C._nn.linear( - x_211, - l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, - ) - x_211 = ( - l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_213 = torch.nn.functional.dropout(x_212, 0.0, False, False) - x_212 = None - x_214 = x_206 + x_213 - x_206 = x_213 = None - x_215 = torch.nn.functional.layer_norm( - x_214, - (1024,), - l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ - ) = None - linear_56 = torch._C._nn.linear( - x_215, - l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_, - ) - x_215 = ( - l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_28 = linear_56.reshape(1, 197, 3, 16, 64) - linear_56 = None - qkv_14 = reshape_28.permute(2, 0, 3, 1, 4) - reshape_28 = None - unbind_14 = qkv_14.unbind(0) - qkv_14 = None - q_14 = unbind_14[0] - k_14 = unbind_14[1] - v_14 = unbind_14[2] - unbind_14 = None - x_216 = torch._C._nn.scaled_dot_product_attention( - q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 - ) - q_14 = k_14 = v_14 = None - transpose_15 = x_216.transpose(1, 2) - x_216 = None - x_217 = transpose_15.reshape(1, 197, 1024) - transpose_15 = None - x_218 = torch.nn.functional.layer_norm( - x_217, - (1024,), - l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_217 = l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_ - ) = None - x_219 = torch._C._nn.linear( - x_218, - l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_, - ) - x_218 = l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ - ) = None - x_220 = torch.nn.functional.dropout(x_219, 0.0, False, False) - x_219 = None - x_221 = x_214 + x_220 - x_214 = x_220 = None - x_222 = torch.nn.functional.layer_norm( - x_221, - (1024,), - l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ - ) = None - x_223 = torch._C._nn.linear( - x_222, - l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, - ) - x_222 = ( - l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_224 = torch._C._nn.gelu(x_223, approximate="none") - x_223 = None - x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) - x_224 = None - x_226 = torch.nn.functional.layer_norm( - x_225, - (4096,), - l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_225 = ( - l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_227 = torch._C._nn.linear( - x_226, - l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, - ) - x_226 = ( - l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_228 = torch.nn.functional.dropout(x_227, 0.0, False, False) - x_227 = None - x_229 = x_221 + x_228 - x_221 = x_228 = None - x_230 = torch.nn.functional.layer_norm( - x_229, - (1024,), - l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ - ) = None - linear_60 = torch._C._nn.linear( - x_230, - l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_, - ) - x_230 = ( - l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_30 = linear_60.reshape(1, 197, 3, 16, 64) - linear_60 = None - qkv_15 = reshape_30.permute(2, 0, 3, 1, 4) - reshape_30 = None - unbind_15 = qkv_15.unbind(0) - qkv_15 = None - q_15 = unbind_15[0] - k_15 = unbind_15[1] - v_15 = unbind_15[2] - unbind_15 = None - x_231 = torch._C._nn.scaled_dot_product_attention( - q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 - ) - q_15 = k_15 = v_15 = None - transpose_16 = x_231.transpose(1, 2) - x_231 = None - x_232 = transpose_16.reshape(1, 197, 1024) - transpose_16 = None - x_233 = torch.nn.functional.layer_norm( - x_232, - (1024,), - l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_232 = l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_ - ) = None - x_234 = torch._C._nn.linear( - x_233, - l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_, - ) - x_233 = l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ - ) = None - x_235 = torch.nn.functional.dropout(x_234, 0.0, False, False) - x_234 = None - x_236 = x_229 + x_235 - x_229 = x_235 = None - x_237 = torch.nn.functional.layer_norm( - x_236, - (1024,), - l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ - ) = None - x_238 = torch._C._nn.linear( - x_237, - l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, - ) - x_237 = ( - l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_239 = torch._C._nn.gelu(x_238, approximate="none") - x_238 = None - x_240 = torch.nn.functional.dropout(x_239, 0.0, False, False) - x_239 = None - x_241 = torch.nn.functional.layer_norm( - x_240, - (4096,), - l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_240 = ( - l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_242 = torch._C._nn.linear( - x_241, - l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, - ) - x_241 = ( - l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_243 = torch.nn.functional.dropout(x_242, 0.0, False, False) - x_242 = None - x_244 = x_236 + x_243 - x_236 = x_243 = None - x_245 = torch.nn.functional.layer_norm( - x_244, - (1024,), - l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ - ) = None - linear_64 = torch._C._nn.linear( - x_245, - l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_, - ) - x_245 = ( - l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_32 = linear_64.reshape(1, 197, 3, 16, 64) - linear_64 = None - qkv_16 = reshape_32.permute(2, 0, 3, 1, 4) - reshape_32 = None - unbind_16 = qkv_16.unbind(0) - qkv_16 = None - q_16 = unbind_16[0] - k_16 = unbind_16[1] - v_16 = unbind_16[2] - unbind_16 = None - x_246 = torch._C._nn.scaled_dot_product_attention( - q_16, k_16, v_16, attn_mask=None, dropout_p=0.0 - ) - q_16 = k_16 = v_16 = None - transpose_17 = x_246.transpose(1, 2) - x_246 = None - x_247 = transpose_17.reshape(1, 197, 1024) - transpose_17 = None - x_248 = torch.nn.functional.layer_norm( - x_247, - (1024,), - l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_247 = l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_ - ) = None - x_249 = torch._C._nn.linear( - x_248, - l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_, - ) - x_248 = l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ - ) = None - x_250 = torch.nn.functional.dropout(x_249, 0.0, False, False) - x_249 = None - x_251 = x_244 + x_250 - x_244 = x_250 = None - x_252 = torch.nn.functional.layer_norm( - x_251, - (1024,), - l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ - ) = None - x_253 = torch._C._nn.linear( - x_252, - l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, - ) - x_252 = ( - l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_254 = torch._C._nn.gelu(x_253, approximate="none") - x_253 = None - x_255 = torch.nn.functional.dropout(x_254, 0.0, False, False) - x_254 = None - x_256 = torch.nn.functional.layer_norm( - x_255, - (4096,), - l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_255 = ( - l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_257 = torch._C._nn.linear( - x_256, - l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, - ) - x_256 = ( - l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_258 = torch.nn.functional.dropout(x_257, 0.0, False, False) - x_257 = None - x_259 = x_251 + x_258 - x_251 = x_258 = None - x_260 = torch.nn.functional.layer_norm( - x_259, - (1024,), - l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ - ) = None - linear_68 = torch._C._nn.linear( - x_260, - l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_, - ) - x_260 = ( - l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_34 = linear_68.reshape(1, 197, 3, 16, 64) - linear_68 = None - qkv_17 = reshape_34.permute(2, 0, 3, 1, 4) - reshape_34 = None - unbind_17 = qkv_17.unbind(0) - qkv_17 = None - q_17 = unbind_17[0] - k_17 = unbind_17[1] - v_17 = unbind_17[2] - unbind_17 = None - x_261 = torch._C._nn.scaled_dot_product_attention( - q_17, k_17, v_17, attn_mask=None, dropout_p=0.0 - ) - q_17 = k_17 = v_17 = None - transpose_18 = x_261.transpose(1, 2) - x_261 = None - x_262 = transpose_18.reshape(1, 197, 1024) - transpose_18 = None - x_263 = torch.nn.functional.layer_norm( - x_262, - (1024,), - l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_262 = l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_ - ) = None - x_264 = torch._C._nn.linear( - x_263, - l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_, - ) - x_263 = l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ - ) = None - x_265 = torch.nn.functional.dropout(x_264, 0.0, False, False) - x_264 = None - x_266 = x_259 + x_265 - x_259 = x_265 = None - x_267 = torch.nn.functional.layer_norm( - x_266, - (1024,), - l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ - ) = None - x_268 = torch._C._nn.linear( - x_267, - l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, - ) - x_267 = ( - l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_269 = torch._C._nn.gelu(x_268, approximate="none") - x_268 = None - x_270 = torch.nn.functional.dropout(x_269, 0.0, False, False) - x_269 = None - x_271 = torch.nn.functional.layer_norm( - x_270, - (4096,), - l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_270 = ( - l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_272 = torch._C._nn.linear( - x_271, - l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, - ) - x_271 = ( - l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_273 = torch.nn.functional.dropout(x_272, 0.0, False, False) - x_272 = None - x_274 = x_266 + x_273 - x_266 = x_273 = None - x_275 = torch.nn.functional.layer_norm( - x_274, - (1024,), - l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ - ) = None - linear_72 = torch._C._nn.linear( - x_275, - l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_, - ) - x_275 = ( - l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_36 = linear_72.reshape(1, 197, 3, 16, 64) - linear_72 = None - qkv_18 = reshape_36.permute(2, 0, 3, 1, 4) - reshape_36 = None - unbind_18 = qkv_18.unbind(0) - qkv_18 = None - q_18 = unbind_18[0] - k_18 = unbind_18[1] - v_18 = unbind_18[2] - unbind_18 = None - x_276 = torch._C._nn.scaled_dot_product_attention( - q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 - ) - q_18 = k_18 = v_18 = None - transpose_19 = x_276.transpose(1, 2) - x_276 = None - x_277 = transpose_19.reshape(1, 197, 1024) - transpose_19 = None - x_278 = torch.nn.functional.layer_norm( - x_277, - (1024,), - l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_277 = l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_ - ) = None - x_279 = torch._C._nn.linear( - x_278, - l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_, - ) - x_278 = l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ - ) = None - x_280 = torch.nn.functional.dropout(x_279, 0.0, False, False) - x_279 = None - x_281 = x_274 + x_280 - x_274 = x_280 = None - x_282 = torch.nn.functional.layer_norm( - x_281, - (1024,), - l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ - ) = None - x_283 = torch._C._nn.linear( - x_282, - l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, - ) - x_282 = ( - l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_284 = torch._C._nn.gelu(x_283, approximate="none") - x_283 = None - x_285 = torch.nn.functional.dropout(x_284, 0.0, False, False) - x_284 = None - x_286 = torch.nn.functional.layer_norm( - x_285, - (4096,), - l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_285 = ( - l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_287 = torch._C._nn.linear( - x_286, - l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, - ) - x_286 = ( - l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_288 = torch.nn.functional.dropout(x_287, 0.0, False, False) - x_287 = None - x_289 = x_281 + x_288 - x_281 = x_288 = None - x_290 = torch.nn.functional.layer_norm( - x_289, - (1024,), - l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ - ) = None - linear_76 = torch._C._nn.linear( - x_290, - l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_, - ) - x_290 = ( - l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_38 = linear_76.reshape(1, 197, 3, 16, 64) - linear_76 = None - qkv_19 = reshape_38.permute(2, 0, 3, 1, 4) - reshape_38 = None - unbind_19 = qkv_19.unbind(0) - qkv_19 = None - q_19 = unbind_19[0] - k_19 = unbind_19[1] - v_19 = unbind_19[2] - unbind_19 = None - x_291 = torch._C._nn.scaled_dot_product_attention( - q_19, k_19, v_19, attn_mask=None, dropout_p=0.0 - ) - q_19 = k_19 = v_19 = None - transpose_20 = x_291.transpose(1, 2) - x_291 = None - x_292 = transpose_20.reshape(1, 197, 1024) - transpose_20 = None - x_293 = torch.nn.functional.layer_norm( - x_292, - (1024,), - l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_292 = l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_ - ) = None - x_294 = torch._C._nn.linear( - x_293, - l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_, - ) - x_293 = l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ - ) = None - x_295 = torch.nn.functional.dropout(x_294, 0.0, False, False) - x_294 = None - x_296 = x_289 + x_295 - x_289 = x_295 = None - x_297 = torch.nn.functional.layer_norm( - x_296, - (1024,), - l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ - ) = None - x_298 = torch._C._nn.linear( - x_297, - l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, - ) - x_297 = ( - l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_299 = torch._C._nn.gelu(x_298, approximate="none") - x_298 = None - x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) - x_299 = None - x_301 = torch.nn.functional.layer_norm( - x_300, - (4096,), - l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_300 = ( - l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_302 = torch._C._nn.linear( - x_301, - l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, - ) - x_301 = ( - l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_303 = torch.nn.functional.dropout(x_302, 0.0, False, False) - x_302 = None - x_304 = x_296 + x_303 - x_296 = x_303 = None - x_305 = torch.nn.functional.layer_norm( - x_304, - (1024,), - l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ - ) = None - linear_80 = torch._C._nn.linear( - x_305, - l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_, - ) - x_305 = ( - l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_40 = linear_80.reshape(1, 197, 3, 16, 64) - linear_80 = None - qkv_20 = reshape_40.permute(2, 0, 3, 1, 4) - reshape_40 = None - unbind_20 = qkv_20.unbind(0) - qkv_20 = None - q_20 = unbind_20[0] - k_20 = unbind_20[1] - v_20 = unbind_20[2] - unbind_20 = None - x_306 = torch._C._nn.scaled_dot_product_attention( - q_20, k_20, v_20, attn_mask=None, dropout_p=0.0 - ) - q_20 = k_20 = v_20 = None - transpose_21 = x_306.transpose(1, 2) - x_306 = None - x_307 = transpose_21.reshape(1, 197, 1024) - transpose_21 = None - x_308 = torch.nn.functional.layer_norm( - x_307, - (1024,), - l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_307 = l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_ - ) = None - x_309 = torch._C._nn.linear( - x_308, - l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_, - ) - x_308 = l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ - ) = None - x_310 = torch.nn.functional.dropout(x_309, 0.0, False, False) - x_309 = None - x_311 = x_304 + x_310 - x_304 = x_310 = None - x_312 = torch.nn.functional.layer_norm( - x_311, - (1024,), - l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ - ) = None - x_313 = torch._C._nn.linear( - x_312, - l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, - ) - x_312 = ( - l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_314 = torch._C._nn.gelu(x_313, approximate="none") - x_313 = None - x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) - x_314 = None - x_316 = torch.nn.functional.layer_norm( - x_315, - (4096,), - l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_315 = ( - l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_317 = torch._C._nn.linear( - x_316, - l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, - ) - x_316 = ( - l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_318 = torch.nn.functional.dropout(x_317, 0.0, False, False) - x_317 = None - x_319 = x_311 + x_318 - x_311 = x_318 = None - x_320 = torch.nn.functional.layer_norm( - x_319, - (1024,), - l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ - ) = None - linear_84 = torch._C._nn.linear( - x_320, - l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_, - ) - x_320 = ( - l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_42 = linear_84.reshape(1, 197, 3, 16, 64) - linear_84 = None - qkv_21 = reshape_42.permute(2, 0, 3, 1, 4) - reshape_42 = None - unbind_21 = qkv_21.unbind(0) - qkv_21 = None - q_21 = unbind_21[0] - k_21 = unbind_21[1] - v_21 = unbind_21[2] - unbind_21 = None - x_321 = torch._C._nn.scaled_dot_product_attention( - q_21, k_21, v_21, attn_mask=None, dropout_p=0.0 - ) - q_21 = k_21 = v_21 = None - transpose_22 = x_321.transpose(1, 2) - x_321 = None - x_322 = transpose_22.reshape(1, 197, 1024) - transpose_22 = None - x_323 = torch.nn.functional.layer_norm( - x_322, - (1024,), - l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_322 = l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_ - ) = None - x_324 = torch._C._nn.linear( - x_323, - l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_, - ) - x_323 = l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ - ) = None - x_325 = torch.nn.functional.dropout(x_324, 0.0, False, False) - x_324 = None - x_326 = x_319 + x_325 - x_319 = x_325 = None - x_327 = torch.nn.functional.layer_norm( - x_326, - (1024,), - l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ - ) = None - x_328 = torch._C._nn.linear( - x_327, - l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, - ) - x_327 = ( - l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_329 = torch._C._nn.gelu(x_328, approximate="none") - x_328 = None - x_330 = torch.nn.functional.dropout(x_329, 0.0, False, False) - x_329 = None - x_331 = torch.nn.functional.layer_norm( - x_330, - (4096,), - l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_330 = ( - l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_332 = torch._C._nn.linear( - x_331, - l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, - ) - x_331 = ( - l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) - x_332 = None - x_334 = x_326 + x_333 - x_326 = x_333 = None - x_335 = torch.nn.functional.layer_norm( - x_334, - (1024,), - l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ - ) = None - linear_88 = torch._C._nn.linear( - x_335, - l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_, - ) - x_335 = ( - l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_44 = linear_88.reshape(1, 197, 3, 16, 64) - linear_88 = None - qkv_22 = reshape_44.permute(2, 0, 3, 1, 4) - reshape_44 = None - unbind_22 = qkv_22.unbind(0) - qkv_22 = None - q_22 = unbind_22[0] - k_22 = unbind_22[1] - v_22 = unbind_22[2] - unbind_22 = None - x_336 = torch._C._nn.scaled_dot_product_attention( - q_22, k_22, v_22, attn_mask=None, dropout_p=0.0 - ) - q_22 = k_22 = v_22 = None - transpose_23 = x_336.transpose(1, 2) - x_336 = None - x_337 = transpose_23.reshape(1, 197, 1024) - transpose_23 = None - x_338 = torch.nn.functional.layer_norm( - x_337, - (1024,), - l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_337 = l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_ - ) = None - x_339 = torch._C._nn.linear( - x_338, - l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_, - ) - x_338 = l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ - ) = None - x_340 = torch.nn.functional.dropout(x_339, 0.0, False, False) - x_339 = None - x_341 = x_334 + x_340 - x_334 = x_340 = None - x_342 = torch.nn.functional.layer_norm( - x_341, - (1024,), - l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ - ) = None - x_343 = torch._C._nn.linear( - x_342, - l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, - ) - x_342 = ( - l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_344 = torch._C._nn.gelu(x_343, approximate="none") - x_343 = None - x_345 = torch.nn.functional.dropout(x_344, 0.0, False, False) - x_344 = None - x_346 = torch.nn.functional.layer_norm( - x_345, - (4096,), - l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_345 = ( - l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_347 = torch._C._nn.linear( - x_346, - l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, - ) - x_346 = ( - l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_348 = torch.nn.functional.dropout(x_347, 0.0, False, False) - x_347 = None - x_349 = x_341 + x_348 - x_341 = x_348 = None - x_350 = torch.nn.functional.layer_norm( - x_349, - (1024,), - l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_, - l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( - l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ - ) = None - linear_92 = torch._C._nn.linear( - x_350, - l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_, - l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_, - ) - x_350 = ( - l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ - ) = None - reshape_46 = linear_92.reshape(1, 197, 3, 16, 64) - linear_92 = None - qkv_23 = reshape_46.permute(2, 0, 3, 1, 4) - reshape_46 = None - unbind_23 = qkv_23.unbind(0) - qkv_23 = None - q_23 = unbind_23[0] - k_23 = unbind_23[1] - v_23 = unbind_23[2] - unbind_23 = None - x_351 = torch._C._nn.scaled_dot_product_attention( - q_23, k_23, v_23, attn_mask=None, dropout_p=0.0 - ) - q_23 = k_23 = v_23 = None - transpose_24 = x_351.transpose(1, 2) - x_351 = None - x_352 = transpose_24.reshape(1, 197, 1024) - transpose_24 = None - x_353 = torch.nn.functional.layer_norm( - x_352, - (1024,), - l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_, - 1e-05, - ) - x_352 = l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_ = ( - l_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_ - ) = None - x_354 = torch._C._nn.linear( - x_353, - l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_, - l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_, - ) - x_353 = l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = ( - l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ - ) = None - x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) - x_354 = None - x_356 = x_349 + x_355 - x_349 = x_355 = None - x_357 = torch.nn.functional.layer_norm( - x_356, - (1024,), - l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_, - l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_, - 1e-05, - ) - l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( - l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ - ) = None - x_358 = torch._C._nn.linear( - x_357, - l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, - l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, - ) - x_357 = ( - l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ - ) = None - x_359 = torch._C._nn.gelu(x_358, approximate="none") - x_358 = None - x_360 = torch.nn.functional.dropout(x_359, 0.0, False, False) - x_359 = None - x_361 = torch.nn.functional.layer_norm( - x_360, - (4096,), - l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_, - l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_, - 1e-05, - ) - x_360 = ( - l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_ - ) = None - x_362 = torch._C._nn.linear( - x_361, - l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, - l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, - ) - x_361 = ( - l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ - ) = ( - l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ - ) = None - x_363 = torch.nn.functional.dropout(x_362, 0.0, False, False) - x_362 = None - x_364 = x_356 + x_363 - x_356 = x_363 = None - x_365 = x_364[ - (slice(None, None, None), slice(l_self_num_prefix_tokens, None, None)) - ] - x_364 = l_self_num_prefix_tokens = None - x_366 = x_365.mean(dim=1) - x_365 = None - x_367 = torch.nn.functional.layer_norm( - x_366, - (1024,), - l_self_modules_fc_norm_parameters_weight_, - l_self_modules_fc_norm_parameters_bias_, - 1e-05, - ) - x_366 = ( - l_self_modules_fc_norm_parameters_weight_ - ) = l_self_modules_fc_norm_parameters_bias_ = None - x_368 = torch.nn.functional.dropout(x_367, 0.0, False, False) - x_367 = None - return (x_368,) diff --git a/samples/timm/beit3_large_patch16_224.pt/weight_meta.py b/samples/timm/beit3_large_patch16_224.pt/weight_meta.py deleted file mode 100644 index 6f3c28f89d..0000000000 --- a/samples/timm/beit3_large_patch16_224.pt/weight_meta.py +++ /dev/null @@ -1,4070 +0,0 @@ -class Program_weight_tensor_meta_L_x_: - name = "L_x_" - shape = [1, 3, 224, 224] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.228 - std = 1.284 - data = None - - -class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: - name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" - shape = [1024, 3, 16, 16] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.021 - data = None - - -class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: - name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.001 - std = 0.021 - data = None - - -class Program_weight_tensor_meta_L_self_parameters_pos_embed_: - name = "L_self_parameters_pos_embed_" - shape = [1, 197, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_parameters_cls_token_: - name = "L_self_parameters_cls_token_" - shape = [1, 1, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_0_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_0_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_1_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_1_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_2_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_2_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_3_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_3_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_4_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_4_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_5_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_5_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_6_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_6_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_7_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_7_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_8_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_8_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: - name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_9_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_: - name = "L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_weight_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_9_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_10_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_10_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_11_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_11_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_12_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_12_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_13_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_13_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_14_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_14_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_15_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_15_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_16_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_16_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_17_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_17_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_18_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_18_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_19_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_19_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_20_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_20_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_21_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_21_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_22_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_22_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: - name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: - name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_" - ) - shape = [3072, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: - name = "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_" - shape = [3072] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_weight_" - ) - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_23_modules_attn_modules_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_" - ) - shape = [1024, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: - name = "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: - name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: - name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: - name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" - shape = [4096, 1024] - dtype = "torch.float32" - device = "cuda:0" - mean = -0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: - name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_: - name = ( - "L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_weight_" - ) - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_: - name = "L_self_modules_blocks_modules_23_modules_mlp_modules_norm_parameters_bias_" - shape = [4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: - name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" - shape = [1024, 4096] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.020 - data = None - - -class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: - name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_num_prefix_tokens: - name = "L_self_num_prefix_tokens" - shape = [] - dtype = "torch.int64" - device = "cpu" - mean = None - std = None - data = [4] - - -class Program_weight_tensor_meta_L_self_modules_fc_norm_parameters_weight_: - name = "L_self_modules_fc_norm_parameters_weight_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 1.000 - std = 0.000 - data = None - - -class Program_weight_tensor_meta_L_self_modules_fc_norm_parameters_bias_: - name = "L_self_modules_fc_norm_parameters_bias_" - shape = [1024] - dtype = "torch.float32" - device = "cuda:0" - mean = 0.000 - std = 0.000 - data = None diff --git a/samples/timm/convmixer_1024_20_ks9_p14.in1k/graph_hash.txt b/samples/timm/convmixer_1024_20_ks9_p14.in1k/graph_hash.txt new file mode 100644 index 0000000000..8c7797f1b4 --- /dev/null +++ b/samples/timm/convmixer_1024_20_ks9_p14.in1k/graph_hash.txt @@ -0,0 +1 @@ +278547c85ed73465aeb4a413847ac5ebf20507ba3921a16f80994470a9d0c746 \ No newline at end of file diff --git a/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/graph_net.json b/samples/timm/convmixer_1024_20_ks9_p14.in1k/graph_net.json similarity index 100% rename from samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/graph_net.json rename to samples/timm/convmixer_1024_20_ks9_p14.in1k/graph_net.json diff --git a/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/input_meta.py b/samples/timm/convmixer_1024_20_ks9_p14.in1k/input_meta.py similarity index 100% rename from samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/input_meta.py rename to samples/timm/convmixer_1024_20_ks9_p14.in1k/input_meta.py diff --git a/samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/input_tensor_constraints.py b/samples/timm/convmixer_1024_20_ks9_p14.in1k/input_tensor_constraints.py similarity index 100% rename from samples/timm/beit3_large_patch16_224.indomain_in22k_ft_in1k/input_tensor_constraints.py rename to samples/timm/convmixer_1024_20_ks9_p14.in1k/input_tensor_constraints.py diff --git a/samples/timm/convmixer_1024_20_ks9_p14.in1k/model.py b/samples/timm/convmixer_1024_20_ks9_p14.in1k/model.py new file mode 100644 index 0000000000..370b48c0e5 --- /dev/null +++ b/samples/timm/convmixer_1024_20_ks9_p14.in1k/model.py @@ -0,0 +1,1924 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_0_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_0_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_1_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_1_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_2_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_2_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_3_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_3_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_4_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_4_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_5_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_5_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_6_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_6_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_7_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_7_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_8_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_8_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_9_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_9_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_10_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_10_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_11_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_11_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_12_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_12_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_13_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_13_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_14_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_14_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_15_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_15_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_16_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_16_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_17_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_17_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_18_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_18_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_19_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_19_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_2_buffers_running_mean_ = ( + L_self_modules_stem_modules_2_buffers_running_mean_ + ) + l_self_modules_stem_modules_2_buffers_running_var_ = ( + L_self_modules_stem_modules_2_buffers_running_var_ + ) + l_self_modules_stem_modules_2_parameters_weight_ = ( + L_self_modules_stem_modules_2_parameters_weight_ + ) + l_self_modules_stem_modules_2_parameters_bias_ = ( + L_self_modules_stem_modules_2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_0_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_0_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_0_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_0_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_1_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_1_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_1_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_1_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_2_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_2_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_2_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_2_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_3_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_3_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_3_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_3_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_3_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_4_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_4_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_4_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_4_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_4_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_5_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_5_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_5_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_5_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_5_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_6_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_6_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_6_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_6_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_6_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_7_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_7_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_7_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_7_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_7_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_8_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_8_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_8_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_8_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_8_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_9_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_9_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_9_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_9_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_9_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_10_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_10_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_10_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_10_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_10_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_11_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_11_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_11_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_11_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_11_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_12_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_12_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_12_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_12_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_12_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_13_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_13_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_13_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_13_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_13_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_14_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_14_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_14_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_14_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_14_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_15_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_15_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_15_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_15_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_15_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_16_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_16_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_16_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_16_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_16_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_17_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_17_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_17_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_17_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_17_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_18_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_18_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_18_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_18_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_18_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_19_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_19_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_19_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_19_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_19_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_3_parameters_bias_ + ) + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (14, 14), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + input_2 = torch._C._nn.gelu(input_1, approximate="none") + input_1 = None + input_3 = torch.nn.functional.batch_norm( + input_2, + l_self_modules_stem_modules_2_buffers_running_mean_, + l_self_modules_stem_modules_2_buffers_running_var_, + l_self_modules_stem_modules_2_parameters_weight_, + l_self_modules_stem_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_2 = ( + l_self_modules_stem_modules_2_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_2_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_2_parameters_weight_ + ) = l_self_modules_stem_modules_2_parameters_bias_ = None + input_4 = torch.conv2d( + input_3, + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1024, + ) + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_5 = torch._C._nn.gelu(input_4, approximate="none") + input_4 = None + input_6 = torch.nn.functional.batch_norm( + input_5, + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_5 = l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_7 = input_6 + input_3 + input_6 = input_3 = None + input_8 = torch.conv2d( + input_7, + l_self_modules_blocks_modules_0_modules_1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_7 = ( + l_self_modules_blocks_modules_0_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_0_modules_1_parameters_bias_ = None + input_9 = torch._C._nn.gelu(input_8, approximate="none") + input_8 = None + input_10 = torch.nn.functional.batch_norm( + input_9, + l_self_modules_blocks_modules_0_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_0_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_0_modules_3_parameters_weight_, + l_self_modules_blocks_modules_0_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_9 = ( + l_self_modules_blocks_modules_0_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_0_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_0_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_0_modules_3_parameters_bias_ = None + input_11 = torch.conv2d( + input_10, + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1024, + ) + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_12 = torch._C._nn.gelu(input_11, approximate="none") + input_11 = None + input_13 = torch.nn.functional.batch_norm( + input_12, + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_12 = l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_14 = input_13 + input_10 + input_13 = input_10 = None + input_15 = torch.conv2d( + input_14, + l_self_modules_blocks_modules_1_modules_1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_14 = ( + l_self_modules_blocks_modules_1_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_1_modules_1_parameters_bias_ = None + input_16 = torch._C._nn.gelu(input_15, approximate="none") + input_15 = None + input_17 = torch.nn.functional.batch_norm( + input_16, + l_self_modules_blocks_modules_1_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_1_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_1_modules_3_parameters_weight_, + l_self_modules_blocks_modules_1_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_16 = ( + l_self_modules_blocks_modules_1_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_1_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_1_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_1_modules_3_parameters_bias_ = None + input_18 = torch.conv2d( + input_17, + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1024, + ) + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_19 = torch._C._nn.gelu(input_18, approximate="none") + input_18 = None + input_20 = torch.nn.functional.batch_norm( + input_19, + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_19 = l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_21 = input_20 + input_17 + input_20 = input_17 = None + input_22 = torch.conv2d( + input_21, + l_self_modules_blocks_modules_2_modules_1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_21 = ( + l_self_modules_blocks_modules_2_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_2_modules_1_parameters_bias_ = None + input_23 = torch._C._nn.gelu(input_22, approximate="none") + input_22 = None + input_24 = torch.nn.functional.batch_norm( + input_23, + l_self_modules_blocks_modules_2_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_2_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_2_modules_3_parameters_weight_, + l_self_modules_blocks_modules_2_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_23 = ( + l_self_modules_blocks_modules_2_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_2_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_2_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_2_modules_3_parameters_bias_ = None + input_25 = torch.conv2d( + input_24, + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1024, + ) + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_26 = torch._C._nn.gelu(input_25, approximate="none") + input_25 = None + input_27 = torch.nn.functional.batch_norm( + input_26, + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_26 = l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_28 = input_27 + input_24 + input_27 = input_24 = None + input_29 = torch.conv2d( + input_28, + l_self_modules_blocks_modules_3_modules_1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_28 = ( + l_self_modules_blocks_modules_3_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_3_modules_1_parameters_bias_ = None + input_30 = torch._C._nn.gelu(input_29, approximate="none") + input_29 = None + input_31 = torch.nn.functional.batch_norm( + input_30, + l_self_modules_blocks_modules_3_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_3_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_3_modules_3_parameters_weight_, + l_self_modules_blocks_modules_3_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_30 = ( + l_self_modules_blocks_modules_3_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_3_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_3_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_3_modules_3_parameters_bias_ = None + input_32 = torch.conv2d( + input_31, + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1024, + ) + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_33 = torch._C._nn.gelu(input_32, approximate="none") + input_32 = None + input_34 = torch.nn.functional.batch_norm( + input_33, + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_33 = l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_35 = input_34 + input_31 + input_34 = input_31 = None + input_36 = torch.conv2d( + input_35, + l_self_modules_blocks_modules_4_modules_1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_35 = ( + l_self_modules_blocks_modules_4_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_4_modules_1_parameters_bias_ = None + input_37 = torch._C._nn.gelu(input_36, approximate="none") + input_36 = None + input_38 = torch.nn.functional.batch_norm( + input_37, + l_self_modules_blocks_modules_4_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_4_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_4_modules_3_parameters_weight_, + l_self_modules_blocks_modules_4_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_37 = ( + l_self_modules_blocks_modules_4_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_4_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_4_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_4_modules_3_parameters_bias_ = None + input_39 = torch.conv2d( + input_38, + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1024, + ) + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_40 = torch._C._nn.gelu(input_39, approximate="none") + input_39 = None + input_41 = torch.nn.functional.batch_norm( + input_40, + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_40 = l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_42 = input_41 + input_38 + input_41 = input_38 = None + input_43 = torch.conv2d( + input_42, + l_self_modules_blocks_modules_5_modules_1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_42 = ( + l_self_modules_blocks_modules_5_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_5_modules_1_parameters_bias_ = None + input_44 = torch._C._nn.gelu(input_43, approximate="none") + input_43 = None + input_45 = torch.nn.functional.batch_norm( + input_44, + l_self_modules_blocks_modules_5_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_5_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_5_modules_3_parameters_weight_, + l_self_modules_blocks_modules_5_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_44 = ( + l_self_modules_blocks_modules_5_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_5_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_5_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_5_modules_3_parameters_bias_ = None + input_46 = torch.conv2d( + input_45, + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1024, + ) + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_47 = torch._C._nn.gelu(input_46, approximate="none") + input_46 = None + input_48 = torch.nn.functional.batch_norm( + input_47, + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_47 = l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_49 = input_48 + input_45 + input_48 = input_45 = None + input_50 = torch.conv2d( + input_49, + l_self_modules_blocks_modules_6_modules_1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_49 = ( + l_self_modules_blocks_modules_6_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_6_modules_1_parameters_bias_ = None + input_51 = torch._C._nn.gelu(input_50, approximate="none") + input_50 = None + input_52 = torch.nn.functional.batch_norm( + input_51, + l_self_modules_blocks_modules_6_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_6_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_6_modules_3_parameters_weight_, + l_self_modules_blocks_modules_6_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_51 = ( + l_self_modules_blocks_modules_6_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_6_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_6_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_6_modules_3_parameters_bias_ = None + input_53 = torch.conv2d( + input_52, + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1024, + ) + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_54 = torch._C._nn.gelu(input_53, approximate="none") + input_53 = None + input_55 = torch.nn.functional.batch_norm( + input_54, + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_54 = l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_56 = input_55 + input_52 + input_55 = input_52 = None + input_57 = torch.conv2d( + input_56, + l_self_modules_blocks_modules_7_modules_1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_56 = ( + l_self_modules_blocks_modules_7_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_7_modules_1_parameters_bias_ = None + input_58 = torch._C._nn.gelu(input_57, approximate="none") + input_57 = None + input_59 = torch.nn.functional.batch_norm( + input_58, + l_self_modules_blocks_modules_7_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_7_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_7_modules_3_parameters_weight_, + l_self_modules_blocks_modules_7_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_58 = ( + l_self_modules_blocks_modules_7_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_7_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_7_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_7_modules_3_parameters_bias_ = None + input_60 = torch.conv2d( + input_59, + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1024, + ) + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_61 = torch._C._nn.gelu(input_60, approximate="none") + input_60 = None + input_62 = torch.nn.functional.batch_norm( + input_61, + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_61 = l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_63 = input_62 + input_59 + input_62 = input_59 = None + input_64 = torch.conv2d( + input_63, + l_self_modules_blocks_modules_8_modules_1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_63 = ( + l_self_modules_blocks_modules_8_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_8_modules_1_parameters_bias_ = None + input_65 = torch._C._nn.gelu(input_64, approximate="none") + input_64 = None + input_66 = torch.nn.functional.batch_norm( + input_65, + l_self_modules_blocks_modules_8_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_8_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_8_modules_3_parameters_weight_, + l_self_modules_blocks_modules_8_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_65 = ( + l_self_modules_blocks_modules_8_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_8_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_8_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_8_modules_3_parameters_bias_ = None + input_67 = torch.conv2d( + input_66, + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1024, + ) + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_68 = torch._C._nn.gelu(input_67, approximate="none") + input_67 = None + input_69 = torch.nn.functional.batch_norm( + input_68, + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_68 = l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_70 = input_69 + input_66 + input_69 = input_66 = None + input_71 = torch.conv2d( + input_70, + l_self_modules_blocks_modules_9_modules_1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_70 = ( + l_self_modules_blocks_modules_9_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_9_modules_1_parameters_bias_ = None + input_72 = torch._C._nn.gelu(input_71, approximate="none") + input_71 = None + input_73 = torch.nn.functional.batch_norm( + input_72, + l_self_modules_blocks_modules_9_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_9_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_9_modules_3_parameters_weight_, + l_self_modules_blocks_modules_9_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_72 = ( + l_self_modules_blocks_modules_9_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_9_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_9_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_9_modules_3_parameters_bias_ = None + input_74 = torch.conv2d( + input_73, + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1024, + ) + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_75 = torch._C._nn.gelu(input_74, approximate="none") + input_74 = None + input_76 = torch.nn.functional.batch_norm( + input_75, + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_75 = l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_77 = input_76 + input_73 + input_76 = input_73 = None + input_78 = torch.conv2d( + input_77, + l_self_modules_blocks_modules_10_modules_1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_77 = ( + l_self_modules_blocks_modules_10_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_10_modules_1_parameters_bias_ = None + input_79 = torch._C._nn.gelu(input_78, approximate="none") + input_78 = None + input_80 = torch.nn.functional.batch_norm( + input_79, + l_self_modules_blocks_modules_10_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_10_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_10_modules_3_parameters_weight_, + l_self_modules_blocks_modules_10_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_79 = ( + l_self_modules_blocks_modules_10_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_10_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_10_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_10_modules_3_parameters_bias_ = None + input_81 = torch.conv2d( + input_80, + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1024, + ) + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_82 = torch._C._nn.gelu(input_81, approximate="none") + input_81 = None + input_83 = torch.nn.functional.batch_norm( + input_82, + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_82 = l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_84 = input_83 + input_80 + input_83 = input_80 = None + input_85 = torch.conv2d( + input_84, + l_self_modules_blocks_modules_11_modules_1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_84 = ( + l_self_modules_blocks_modules_11_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_11_modules_1_parameters_bias_ = None + input_86 = torch._C._nn.gelu(input_85, approximate="none") + input_85 = None + input_87 = torch.nn.functional.batch_norm( + input_86, + l_self_modules_blocks_modules_11_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_11_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_11_modules_3_parameters_weight_, + l_self_modules_blocks_modules_11_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_86 = ( + l_self_modules_blocks_modules_11_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_11_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_11_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_11_modules_3_parameters_bias_ = None + input_88 = torch.conv2d( + input_87, + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1024, + ) + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_89 = torch._C._nn.gelu(input_88, approximate="none") + input_88 = None + input_90 = torch.nn.functional.batch_norm( + input_89, + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_89 = l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_91 = input_90 + input_87 + input_90 = input_87 = None + input_92 = torch.conv2d( + input_91, + l_self_modules_blocks_modules_12_modules_1_parameters_weight_, + l_self_modules_blocks_modules_12_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_91 = ( + l_self_modules_blocks_modules_12_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_12_modules_1_parameters_bias_ = None + input_93 = torch._C._nn.gelu(input_92, approximate="none") + input_92 = None + input_94 = torch.nn.functional.batch_norm( + input_93, + l_self_modules_blocks_modules_12_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_12_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_12_modules_3_parameters_weight_, + l_self_modules_blocks_modules_12_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_93 = ( + l_self_modules_blocks_modules_12_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_12_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_12_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_12_modules_3_parameters_bias_ = None + input_95 = torch.conv2d( + input_94, + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1024, + ) + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_96 = torch._C._nn.gelu(input_95, approximate="none") + input_95 = None + input_97 = torch.nn.functional.batch_norm( + input_96, + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_96 = l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_98 = input_97 + input_94 + input_97 = input_94 = None + input_99 = torch.conv2d( + input_98, + l_self_modules_blocks_modules_13_modules_1_parameters_weight_, + l_self_modules_blocks_modules_13_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_98 = ( + l_self_modules_blocks_modules_13_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_13_modules_1_parameters_bias_ = None + input_100 = torch._C._nn.gelu(input_99, approximate="none") + input_99 = None + input_101 = torch.nn.functional.batch_norm( + input_100, + l_self_modules_blocks_modules_13_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_13_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_13_modules_3_parameters_weight_, + l_self_modules_blocks_modules_13_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_100 = ( + l_self_modules_blocks_modules_13_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_13_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_13_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_13_modules_3_parameters_bias_ = None + input_102 = torch.conv2d( + input_101, + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1024, + ) + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_103 = torch._C._nn.gelu(input_102, approximate="none") + input_102 = None + input_104 = torch.nn.functional.batch_norm( + input_103, + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_103 = l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_105 = input_104 + input_101 + input_104 = input_101 = None + input_106 = torch.conv2d( + input_105, + l_self_modules_blocks_modules_14_modules_1_parameters_weight_, + l_self_modules_blocks_modules_14_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_105 = ( + l_self_modules_blocks_modules_14_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_14_modules_1_parameters_bias_ = None + input_107 = torch._C._nn.gelu(input_106, approximate="none") + input_106 = None + input_108 = torch.nn.functional.batch_norm( + input_107, + l_self_modules_blocks_modules_14_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_14_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_14_modules_3_parameters_weight_, + l_self_modules_blocks_modules_14_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_107 = ( + l_self_modules_blocks_modules_14_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_14_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_14_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_14_modules_3_parameters_bias_ = None + input_109 = torch.conv2d( + input_108, + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1024, + ) + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_110 = torch._C._nn.gelu(input_109, approximate="none") + input_109 = None + input_111 = torch.nn.functional.batch_norm( + input_110, + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_110 = l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_112 = input_111 + input_108 + input_111 = input_108 = None + input_113 = torch.conv2d( + input_112, + l_self_modules_blocks_modules_15_modules_1_parameters_weight_, + l_self_modules_blocks_modules_15_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_112 = ( + l_self_modules_blocks_modules_15_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_15_modules_1_parameters_bias_ = None + input_114 = torch._C._nn.gelu(input_113, approximate="none") + input_113 = None + input_115 = torch.nn.functional.batch_norm( + input_114, + l_self_modules_blocks_modules_15_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_15_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_15_modules_3_parameters_weight_, + l_self_modules_blocks_modules_15_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_114 = ( + l_self_modules_blocks_modules_15_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_15_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_15_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_15_modules_3_parameters_bias_ = None + input_116 = torch.conv2d( + input_115, + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1024, + ) + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_117 = torch._C._nn.gelu(input_116, approximate="none") + input_116 = None + input_118 = torch.nn.functional.batch_norm( + input_117, + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_117 = l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_119 = input_118 + input_115 + input_118 = input_115 = None + input_120 = torch.conv2d( + input_119, + l_self_modules_blocks_modules_16_modules_1_parameters_weight_, + l_self_modules_blocks_modules_16_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_119 = ( + l_self_modules_blocks_modules_16_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_16_modules_1_parameters_bias_ = None + input_121 = torch._C._nn.gelu(input_120, approximate="none") + input_120 = None + input_122 = torch.nn.functional.batch_norm( + input_121, + l_self_modules_blocks_modules_16_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_16_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_16_modules_3_parameters_weight_, + l_self_modules_blocks_modules_16_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_121 = ( + l_self_modules_blocks_modules_16_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_16_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_16_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_16_modules_3_parameters_bias_ = None + input_123 = torch.conv2d( + input_122, + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1024, + ) + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_124 = torch._C._nn.gelu(input_123, approximate="none") + input_123 = None + input_125 = torch.nn.functional.batch_norm( + input_124, + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_124 = l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_126 = input_125 + input_122 + input_125 = input_122 = None + input_127 = torch.conv2d( + input_126, + l_self_modules_blocks_modules_17_modules_1_parameters_weight_, + l_self_modules_blocks_modules_17_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_126 = ( + l_self_modules_blocks_modules_17_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_17_modules_1_parameters_bias_ = None + input_128 = torch._C._nn.gelu(input_127, approximate="none") + input_127 = None + input_129 = torch.nn.functional.batch_norm( + input_128, + l_self_modules_blocks_modules_17_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_17_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_17_modules_3_parameters_weight_, + l_self_modules_blocks_modules_17_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_128 = ( + l_self_modules_blocks_modules_17_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_17_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_17_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_17_modules_3_parameters_bias_ = None + input_130 = torch.conv2d( + input_129, + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1024, + ) + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_131 = torch._C._nn.gelu(input_130, approximate="none") + input_130 = None + input_132 = torch.nn.functional.batch_norm( + input_131, + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_131 = l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_133 = input_132 + input_129 + input_132 = input_129 = None + input_134 = torch.conv2d( + input_133, + l_self_modules_blocks_modules_18_modules_1_parameters_weight_, + l_self_modules_blocks_modules_18_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_133 = ( + l_self_modules_blocks_modules_18_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_18_modules_1_parameters_bias_ = None + input_135 = torch._C._nn.gelu(input_134, approximate="none") + input_134 = None + input_136 = torch.nn.functional.batch_norm( + input_135, + l_self_modules_blocks_modules_18_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_18_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_18_modules_3_parameters_weight_, + l_self_modules_blocks_modules_18_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_135 = ( + l_self_modules_blocks_modules_18_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_18_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_18_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_18_modules_3_parameters_bias_ = None + input_137 = torch.conv2d( + input_136, + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1024, + ) + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_138 = torch._C._nn.gelu(input_137, approximate="none") + input_137 = None + input_139 = torch.nn.functional.batch_norm( + input_138, + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_138 = l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_140 = input_139 + input_136 + input_139 = input_136 = None + input_141 = torch.conv2d( + input_140, + l_self_modules_blocks_modules_19_modules_1_parameters_weight_, + l_self_modules_blocks_modules_19_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_140 = ( + l_self_modules_blocks_modules_19_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_19_modules_1_parameters_bias_ = None + input_142 = torch._C._nn.gelu(input_141, approximate="none") + input_141 = None + input_143 = torch.nn.functional.batch_norm( + input_142, + l_self_modules_blocks_modules_19_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_19_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_19_modules_3_parameters_weight_, + l_self_modules_blocks_modules_19_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_142 = ( + l_self_modules_blocks_modules_19_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_19_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_19_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_19_modules_3_parameters_bias_ = None + x = torch.nn.functional.adaptive_avg_pool2d(input_143, 1) + input_143 = None + x_1 = x.flatten(1, -1) + x = None + x_2 = torch.nn.functional.dropout(x_1, 0.0, False, False) + x_1 = None + x_3 = torch._C._nn.linear( + x_2, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_2 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_3,) diff --git a/samples/timm/convmixer_1024_20_ks9_p14.in1k/weight_meta.py b/samples/timm/convmixer_1024_20_ks9_p14.in1k/weight_meta.py new file mode 100644 index 0000000000..97442fb3b1 --- /dev/null +++ b/samples/timm/convmixer_1024_20_ks9_p14.in1k/weight_meta.py @@ -0,0 +1,2498 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [1024, 3, 14, 14] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.227 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_buffers_running_mean_: + name = "L_self_modules_stem_modules_2_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_buffers_running_var_: + name = "L_self_modules_stem_modules_2_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_parameters_weight_: + name = "L_self_modules_stem_modules_2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_parameters_bias_: + name = "L_self_modules_stem_modules_2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1024, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_1_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_0_modules_3_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_0_modules_3_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_3_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_3_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1024, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_1_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_1_modules_3_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_1_modules_3_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_3_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_3_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1024, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_1_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_2_modules_3_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_2_modules_3_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_3_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_3_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1024, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_1_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_3_modules_3_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_3_modules_3_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_3_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_3_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1024, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_1_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_4_modules_3_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_4_modules_3_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_3_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_3_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1024, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_1_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_5_modules_3_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_5_modules_3_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_3_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_3_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1024, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_1_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_6_modules_3_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_6_modules_3_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_3_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_3_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1024, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_1_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_7_modules_3_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_7_modules_3_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_3_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_3_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1024, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_1_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_8_modules_3_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_8_modules_3_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_3_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_3_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1024, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_1_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_9_modules_3_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_9_modules_3_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_3_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_3_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1024, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_1_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_10_modules_3_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_10_modules_3_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_3_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_3_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1024, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_1_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_11_modules_3_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_11_modules_3_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_3_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_3_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1024, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_1_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_12_modules_3_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_12_modules_3_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_3_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_3_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1024, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.066 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_1_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_13_modules_3_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_13_modules_3_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_3_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_3_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1024, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_1_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_14_modules_3_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_14_modules_3_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_3_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_3_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1024, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_1_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_15_modules_3_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_15_modules_3_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_3_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_3_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1024, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_1_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_16_modules_3_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_16_modules_3_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_3_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_3_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1024, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_1_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_17_modules_3_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_17_modules_3_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_3_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_3_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1024, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.067 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_1_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_18_modules_3_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_18_modules_3_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_3_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_3_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1024, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.066 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_1_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_19_modules_3_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_19_modules_3_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_3_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_3_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None diff --git a/samples/timm/convmixer_1536_20.in1k/graph_hash.txt b/samples/timm/convmixer_1536_20.in1k/graph_hash.txt new file mode 100644 index 0000000000..c75062cf4b --- /dev/null +++ b/samples/timm/convmixer_1536_20.in1k/graph_hash.txt @@ -0,0 +1 @@ +459befeb660c46c23fdd89e5ceb8db0837d52f8d2e3a40b865554c8f93063598 \ No newline at end of file diff --git a/samples/timm/beit3_large_patch16_224.indomain_pt/graph_net.json b/samples/timm/convmixer_1536_20.in1k/graph_net.json similarity index 100% rename from samples/timm/beit3_large_patch16_224.indomain_pt/graph_net.json rename to samples/timm/convmixer_1536_20.in1k/graph_net.json diff --git a/samples/timm/beit3_large_patch16_224.indomain_pt/input_meta.py b/samples/timm/convmixer_1536_20.in1k/input_meta.py similarity index 100% rename from samples/timm/beit3_large_patch16_224.indomain_pt/input_meta.py rename to samples/timm/convmixer_1536_20.in1k/input_meta.py diff --git a/samples/timm/beit3_large_patch16_224.indomain_pt/input_tensor_constraints.py b/samples/timm/convmixer_1536_20.in1k/input_tensor_constraints.py similarity index 100% rename from samples/timm/beit3_large_patch16_224.indomain_pt/input_tensor_constraints.py rename to samples/timm/convmixer_1536_20.in1k/input_tensor_constraints.py diff --git a/samples/timm/convmixer_1536_20.in1k/model.py b/samples/timm/convmixer_1536_20.in1k/model.py new file mode 100644 index 0000000000..13a625cacb --- /dev/null +++ b/samples/timm/convmixer_1536_20.in1k/model.py @@ -0,0 +1,1924 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_0_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_0_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_1_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_1_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_2_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_2_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_3_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_3_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_4_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_4_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_5_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_5_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_6_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_6_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_7_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_7_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_8_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_8_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_9_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_9_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_10_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_10_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_11_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_11_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_12_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_12_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_13_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_13_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_14_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_14_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_15_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_15_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_16_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_16_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_17_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_17_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_18_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_18_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_19_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_19_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_2_buffers_running_mean_ = ( + L_self_modules_stem_modules_2_buffers_running_mean_ + ) + l_self_modules_stem_modules_2_buffers_running_var_ = ( + L_self_modules_stem_modules_2_buffers_running_var_ + ) + l_self_modules_stem_modules_2_parameters_weight_ = ( + L_self_modules_stem_modules_2_parameters_weight_ + ) + l_self_modules_stem_modules_2_parameters_bias_ = ( + L_self_modules_stem_modules_2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_0_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_0_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_0_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_0_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_1_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_1_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_1_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_1_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_2_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_2_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_2_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_2_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_3_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_3_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_3_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_3_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_3_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_4_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_4_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_4_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_4_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_4_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_5_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_5_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_5_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_5_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_5_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_6_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_6_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_6_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_6_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_6_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_7_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_7_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_7_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_7_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_7_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_8_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_8_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_8_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_8_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_8_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_9_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_9_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_9_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_9_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_9_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_10_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_10_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_10_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_10_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_10_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_11_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_11_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_11_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_11_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_11_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_12_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_12_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_12_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_12_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_12_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_13_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_13_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_13_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_13_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_13_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_14_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_14_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_14_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_14_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_14_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_15_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_15_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_15_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_15_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_15_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_16_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_16_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_16_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_16_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_16_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_17_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_17_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_17_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_17_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_17_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_18_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_18_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_18_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_18_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_18_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_19_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_19_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_19_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_19_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_19_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_3_parameters_bias_ + ) + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (7, 7), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + input_2 = torch._C._nn.gelu(input_1, approximate="none") + input_1 = None + input_3 = torch.nn.functional.batch_norm( + input_2, + l_self_modules_stem_modules_2_buffers_running_mean_, + l_self_modules_stem_modules_2_buffers_running_var_, + l_self_modules_stem_modules_2_parameters_weight_, + l_self_modules_stem_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_2 = ( + l_self_modules_stem_modules_2_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_2_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_2_parameters_weight_ + ) = l_self_modules_stem_modules_2_parameters_bias_ = None + input_4 = torch.conv2d( + input_3, + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1536, + ) + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_5 = torch._C._nn.gelu(input_4, approximate="none") + input_4 = None + input_6 = torch.nn.functional.batch_norm( + input_5, + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_5 = l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_7 = input_6 + input_3 + input_6 = input_3 = None + input_8 = torch.conv2d( + input_7, + l_self_modules_blocks_modules_0_modules_1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_7 = ( + l_self_modules_blocks_modules_0_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_0_modules_1_parameters_bias_ = None + input_9 = torch._C._nn.gelu(input_8, approximate="none") + input_8 = None + input_10 = torch.nn.functional.batch_norm( + input_9, + l_self_modules_blocks_modules_0_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_0_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_0_modules_3_parameters_weight_, + l_self_modules_blocks_modules_0_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_9 = ( + l_self_modules_blocks_modules_0_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_0_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_0_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_0_modules_3_parameters_bias_ = None + input_11 = torch.conv2d( + input_10, + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1536, + ) + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_12 = torch._C._nn.gelu(input_11, approximate="none") + input_11 = None + input_13 = torch.nn.functional.batch_norm( + input_12, + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_12 = l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_14 = input_13 + input_10 + input_13 = input_10 = None + input_15 = torch.conv2d( + input_14, + l_self_modules_blocks_modules_1_modules_1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_14 = ( + l_self_modules_blocks_modules_1_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_1_modules_1_parameters_bias_ = None + input_16 = torch._C._nn.gelu(input_15, approximate="none") + input_15 = None + input_17 = torch.nn.functional.batch_norm( + input_16, + l_self_modules_blocks_modules_1_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_1_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_1_modules_3_parameters_weight_, + l_self_modules_blocks_modules_1_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_16 = ( + l_self_modules_blocks_modules_1_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_1_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_1_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_1_modules_3_parameters_bias_ = None + input_18 = torch.conv2d( + input_17, + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1536, + ) + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_19 = torch._C._nn.gelu(input_18, approximate="none") + input_18 = None + input_20 = torch.nn.functional.batch_norm( + input_19, + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_19 = l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_21 = input_20 + input_17 + input_20 = input_17 = None + input_22 = torch.conv2d( + input_21, + l_self_modules_blocks_modules_2_modules_1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_21 = ( + l_self_modules_blocks_modules_2_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_2_modules_1_parameters_bias_ = None + input_23 = torch._C._nn.gelu(input_22, approximate="none") + input_22 = None + input_24 = torch.nn.functional.batch_norm( + input_23, + l_self_modules_blocks_modules_2_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_2_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_2_modules_3_parameters_weight_, + l_self_modules_blocks_modules_2_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_23 = ( + l_self_modules_blocks_modules_2_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_2_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_2_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_2_modules_3_parameters_bias_ = None + input_25 = torch.conv2d( + input_24, + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1536, + ) + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_26 = torch._C._nn.gelu(input_25, approximate="none") + input_25 = None + input_27 = torch.nn.functional.batch_norm( + input_26, + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_26 = l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_28 = input_27 + input_24 + input_27 = input_24 = None + input_29 = torch.conv2d( + input_28, + l_self_modules_blocks_modules_3_modules_1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_28 = ( + l_self_modules_blocks_modules_3_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_3_modules_1_parameters_bias_ = None + input_30 = torch._C._nn.gelu(input_29, approximate="none") + input_29 = None + input_31 = torch.nn.functional.batch_norm( + input_30, + l_self_modules_blocks_modules_3_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_3_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_3_modules_3_parameters_weight_, + l_self_modules_blocks_modules_3_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_30 = ( + l_self_modules_blocks_modules_3_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_3_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_3_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_3_modules_3_parameters_bias_ = None + input_32 = torch.conv2d( + input_31, + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1536, + ) + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_33 = torch._C._nn.gelu(input_32, approximate="none") + input_32 = None + input_34 = torch.nn.functional.batch_norm( + input_33, + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_33 = l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_35 = input_34 + input_31 + input_34 = input_31 = None + input_36 = torch.conv2d( + input_35, + l_self_modules_blocks_modules_4_modules_1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_35 = ( + l_self_modules_blocks_modules_4_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_4_modules_1_parameters_bias_ = None + input_37 = torch._C._nn.gelu(input_36, approximate="none") + input_36 = None + input_38 = torch.nn.functional.batch_norm( + input_37, + l_self_modules_blocks_modules_4_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_4_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_4_modules_3_parameters_weight_, + l_self_modules_blocks_modules_4_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_37 = ( + l_self_modules_blocks_modules_4_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_4_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_4_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_4_modules_3_parameters_bias_ = None + input_39 = torch.conv2d( + input_38, + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1536, + ) + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_40 = torch._C._nn.gelu(input_39, approximate="none") + input_39 = None + input_41 = torch.nn.functional.batch_norm( + input_40, + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_40 = l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_42 = input_41 + input_38 + input_41 = input_38 = None + input_43 = torch.conv2d( + input_42, + l_self_modules_blocks_modules_5_modules_1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_42 = ( + l_self_modules_blocks_modules_5_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_5_modules_1_parameters_bias_ = None + input_44 = torch._C._nn.gelu(input_43, approximate="none") + input_43 = None + input_45 = torch.nn.functional.batch_norm( + input_44, + l_self_modules_blocks_modules_5_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_5_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_5_modules_3_parameters_weight_, + l_self_modules_blocks_modules_5_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_44 = ( + l_self_modules_blocks_modules_5_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_5_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_5_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_5_modules_3_parameters_bias_ = None + input_46 = torch.conv2d( + input_45, + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1536, + ) + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_47 = torch._C._nn.gelu(input_46, approximate="none") + input_46 = None + input_48 = torch.nn.functional.batch_norm( + input_47, + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_47 = l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_49 = input_48 + input_45 + input_48 = input_45 = None + input_50 = torch.conv2d( + input_49, + l_self_modules_blocks_modules_6_modules_1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_49 = ( + l_self_modules_blocks_modules_6_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_6_modules_1_parameters_bias_ = None + input_51 = torch._C._nn.gelu(input_50, approximate="none") + input_50 = None + input_52 = torch.nn.functional.batch_norm( + input_51, + l_self_modules_blocks_modules_6_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_6_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_6_modules_3_parameters_weight_, + l_self_modules_blocks_modules_6_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_51 = ( + l_self_modules_blocks_modules_6_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_6_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_6_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_6_modules_3_parameters_bias_ = None + input_53 = torch.conv2d( + input_52, + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1536, + ) + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_54 = torch._C._nn.gelu(input_53, approximate="none") + input_53 = None + input_55 = torch.nn.functional.batch_norm( + input_54, + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_54 = l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_56 = input_55 + input_52 + input_55 = input_52 = None + input_57 = torch.conv2d( + input_56, + l_self_modules_blocks_modules_7_modules_1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_56 = ( + l_self_modules_blocks_modules_7_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_7_modules_1_parameters_bias_ = None + input_58 = torch._C._nn.gelu(input_57, approximate="none") + input_57 = None + input_59 = torch.nn.functional.batch_norm( + input_58, + l_self_modules_blocks_modules_7_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_7_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_7_modules_3_parameters_weight_, + l_self_modules_blocks_modules_7_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_58 = ( + l_self_modules_blocks_modules_7_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_7_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_7_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_7_modules_3_parameters_bias_ = None + input_60 = torch.conv2d( + input_59, + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1536, + ) + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_61 = torch._C._nn.gelu(input_60, approximate="none") + input_60 = None + input_62 = torch.nn.functional.batch_norm( + input_61, + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_61 = l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_63 = input_62 + input_59 + input_62 = input_59 = None + input_64 = torch.conv2d( + input_63, + l_self_modules_blocks_modules_8_modules_1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_63 = ( + l_self_modules_blocks_modules_8_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_8_modules_1_parameters_bias_ = None + input_65 = torch._C._nn.gelu(input_64, approximate="none") + input_64 = None + input_66 = torch.nn.functional.batch_norm( + input_65, + l_self_modules_blocks_modules_8_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_8_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_8_modules_3_parameters_weight_, + l_self_modules_blocks_modules_8_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_65 = ( + l_self_modules_blocks_modules_8_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_8_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_8_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_8_modules_3_parameters_bias_ = None + input_67 = torch.conv2d( + input_66, + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1536, + ) + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_68 = torch._C._nn.gelu(input_67, approximate="none") + input_67 = None + input_69 = torch.nn.functional.batch_norm( + input_68, + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_68 = l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_70 = input_69 + input_66 + input_69 = input_66 = None + input_71 = torch.conv2d( + input_70, + l_self_modules_blocks_modules_9_modules_1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_70 = ( + l_self_modules_blocks_modules_9_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_9_modules_1_parameters_bias_ = None + input_72 = torch._C._nn.gelu(input_71, approximate="none") + input_71 = None + input_73 = torch.nn.functional.batch_norm( + input_72, + l_self_modules_blocks_modules_9_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_9_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_9_modules_3_parameters_weight_, + l_self_modules_blocks_modules_9_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_72 = ( + l_self_modules_blocks_modules_9_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_9_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_9_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_9_modules_3_parameters_bias_ = None + input_74 = torch.conv2d( + input_73, + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1536, + ) + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_75 = torch._C._nn.gelu(input_74, approximate="none") + input_74 = None + input_76 = torch.nn.functional.batch_norm( + input_75, + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_75 = l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_77 = input_76 + input_73 + input_76 = input_73 = None + input_78 = torch.conv2d( + input_77, + l_self_modules_blocks_modules_10_modules_1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_77 = ( + l_self_modules_blocks_modules_10_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_10_modules_1_parameters_bias_ = None + input_79 = torch._C._nn.gelu(input_78, approximate="none") + input_78 = None + input_80 = torch.nn.functional.batch_norm( + input_79, + l_self_modules_blocks_modules_10_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_10_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_10_modules_3_parameters_weight_, + l_self_modules_blocks_modules_10_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_79 = ( + l_self_modules_blocks_modules_10_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_10_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_10_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_10_modules_3_parameters_bias_ = None + input_81 = torch.conv2d( + input_80, + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1536, + ) + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_82 = torch._C._nn.gelu(input_81, approximate="none") + input_81 = None + input_83 = torch.nn.functional.batch_norm( + input_82, + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_82 = l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_84 = input_83 + input_80 + input_83 = input_80 = None + input_85 = torch.conv2d( + input_84, + l_self_modules_blocks_modules_11_modules_1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_84 = ( + l_self_modules_blocks_modules_11_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_11_modules_1_parameters_bias_ = None + input_86 = torch._C._nn.gelu(input_85, approximate="none") + input_85 = None + input_87 = torch.nn.functional.batch_norm( + input_86, + l_self_modules_blocks_modules_11_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_11_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_11_modules_3_parameters_weight_, + l_self_modules_blocks_modules_11_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_86 = ( + l_self_modules_blocks_modules_11_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_11_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_11_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_11_modules_3_parameters_bias_ = None + input_88 = torch.conv2d( + input_87, + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1536, + ) + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_89 = torch._C._nn.gelu(input_88, approximate="none") + input_88 = None + input_90 = torch.nn.functional.batch_norm( + input_89, + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_89 = l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_91 = input_90 + input_87 + input_90 = input_87 = None + input_92 = torch.conv2d( + input_91, + l_self_modules_blocks_modules_12_modules_1_parameters_weight_, + l_self_modules_blocks_modules_12_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_91 = ( + l_self_modules_blocks_modules_12_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_12_modules_1_parameters_bias_ = None + input_93 = torch._C._nn.gelu(input_92, approximate="none") + input_92 = None + input_94 = torch.nn.functional.batch_norm( + input_93, + l_self_modules_blocks_modules_12_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_12_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_12_modules_3_parameters_weight_, + l_self_modules_blocks_modules_12_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_93 = ( + l_self_modules_blocks_modules_12_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_12_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_12_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_12_modules_3_parameters_bias_ = None + input_95 = torch.conv2d( + input_94, + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1536, + ) + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_96 = torch._C._nn.gelu(input_95, approximate="none") + input_95 = None + input_97 = torch.nn.functional.batch_norm( + input_96, + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_96 = l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_98 = input_97 + input_94 + input_97 = input_94 = None + input_99 = torch.conv2d( + input_98, + l_self_modules_blocks_modules_13_modules_1_parameters_weight_, + l_self_modules_blocks_modules_13_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_98 = ( + l_self_modules_blocks_modules_13_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_13_modules_1_parameters_bias_ = None + input_100 = torch._C._nn.gelu(input_99, approximate="none") + input_99 = None + input_101 = torch.nn.functional.batch_norm( + input_100, + l_self_modules_blocks_modules_13_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_13_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_13_modules_3_parameters_weight_, + l_self_modules_blocks_modules_13_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_100 = ( + l_self_modules_blocks_modules_13_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_13_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_13_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_13_modules_3_parameters_bias_ = None + input_102 = torch.conv2d( + input_101, + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1536, + ) + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_103 = torch._C._nn.gelu(input_102, approximate="none") + input_102 = None + input_104 = torch.nn.functional.batch_norm( + input_103, + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_103 = l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_105 = input_104 + input_101 + input_104 = input_101 = None + input_106 = torch.conv2d( + input_105, + l_self_modules_blocks_modules_14_modules_1_parameters_weight_, + l_self_modules_blocks_modules_14_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_105 = ( + l_self_modules_blocks_modules_14_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_14_modules_1_parameters_bias_ = None + input_107 = torch._C._nn.gelu(input_106, approximate="none") + input_106 = None + input_108 = torch.nn.functional.batch_norm( + input_107, + l_self_modules_blocks_modules_14_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_14_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_14_modules_3_parameters_weight_, + l_self_modules_blocks_modules_14_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_107 = ( + l_self_modules_blocks_modules_14_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_14_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_14_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_14_modules_3_parameters_bias_ = None + input_109 = torch.conv2d( + input_108, + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1536, + ) + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_110 = torch._C._nn.gelu(input_109, approximate="none") + input_109 = None + input_111 = torch.nn.functional.batch_norm( + input_110, + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_110 = l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_112 = input_111 + input_108 + input_111 = input_108 = None + input_113 = torch.conv2d( + input_112, + l_self_modules_blocks_modules_15_modules_1_parameters_weight_, + l_self_modules_blocks_modules_15_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_112 = ( + l_self_modules_blocks_modules_15_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_15_modules_1_parameters_bias_ = None + input_114 = torch._C._nn.gelu(input_113, approximate="none") + input_113 = None + input_115 = torch.nn.functional.batch_norm( + input_114, + l_self_modules_blocks_modules_15_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_15_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_15_modules_3_parameters_weight_, + l_self_modules_blocks_modules_15_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_114 = ( + l_self_modules_blocks_modules_15_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_15_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_15_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_15_modules_3_parameters_bias_ = None + input_116 = torch.conv2d( + input_115, + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1536, + ) + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_117 = torch._C._nn.gelu(input_116, approximate="none") + input_116 = None + input_118 = torch.nn.functional.batch_norm( + input_117, + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_117 = l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_119 = input_118 + input_115 + input_118 = input_115 = None + input_120 = torch.conv2d( + input_119, + l_self_modules_blocks_modules_16_modules_1_parameters_weight_, + l_self_modules_blocks_modules_16_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_119 = ( + l_self_modules_blocks_modules_16_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_16_modules_1_parameters_bias_ = None + input_121 = torch._C._nn.gelu(input_120, approximate="none") + input_120 = None + input_122 = torch.nn.functional.batch_norm( + input_121, + l_self_modules_blocks_modules_16_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_16_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_16_modules_3_parameters_weight_, + l_self_modules_blocks_modules_16_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_121 = ( + l_self_modules_blocks_modules_16_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_16_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_16_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_16_modules_3_parameters_bias_ = None + input_123 = torch.conv2d( + input_122, + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1536, + ) + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_124 = torch._C._nn.gelu(input_123, approximate="none") + input_123 = None + input_125 = torch.nn.functional.batch_norm( + input_124, + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_124 = l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_126 = input_125 + input_122 + input_125 = input_122 = None + input_127 = torch.conv2d( + input_126, + l_self_modules_blocks_modules_17_modules_1_parameters_weight_, + l_self_modules_blocks_modules_17_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_126 = ( + l_self_modules_blocks_modules_17_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_17_modules_1_parameters_bias_ = None + input_128 = torch._C._nn.gelu(input_127, approximate="none") + input_127 = None + input_129 = torch.nn.functional.batch_norm( + input_128, + l_self_modules_blocks_modules_17_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_17_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_17_modules_3_parameters_weight_, + l_self_modules_blocks_modules_17_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_128 = ( + l_self_modules_blocks_modules_17_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_17_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_17_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_17_modules_3_parameters_bias_ = None + input_130 = torch.conv2d( + input_129, + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1536, + ) + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_131 = torch._C._nn.gelu(input_130, approximate="none") + input_130 = None + input_132 = torch.nn.functional.batch_norm( + input_131, + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_131 = l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_133 = input_132 + input_129 + input_132 = input_129 = None + input_134 = torch.conv2d( + input_133, + l_self_modules_blocks_modules_18_modules_1_parameters_weight_, + l_self_modules_blocks_modules_18_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_133 = ( + l_self_modules_blocks_modules_18_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_18_modules_1_parameters_bias_ = None + input_135 = torch._C._nn.gelu(input_134, approximate="none") + input_134 = None + input_136 = torch.nn.functional.batch_norm( + input_135, + l_self_modules_blocks_modules_18_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_18_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_18_modules_3_parameters_weight_, + l_self_modules_blocks_modules_18_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_135 = ( + l_self_modules_blocks_modules_18_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_18_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_18_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_18_modules_3_parameters_bias_ = None + input_137 = torch.conv2d( + input_136, + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 1536, + ) + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_138 = torch._C._nn.gelu(input_137, approximate="none") + input_137 = None + input_139 = torch.nn.functional.batch_norm( + input_138, + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_138 = l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_140 = input_139 + input_136 + input_139 = input_136 = None + input_141 = torch.conv2d( + input_140, + l_self_modules_blocks_modules_19_modules_1_parameters_weight_, + l_self_modules_blocks_modules_19_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_140 = ( + l_self_modules_blocks_modules_19_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_19_modules_1_parameters_bias_ = None + input_142 = torch._C._nn.gelu(input_141, approximate="none") + input_141 = None + input_143 = torch.nn.functional.batch_norm( + input_142, + l_self_modules_blocks_modules_19_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_19_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_19_modules_3_parameters_weight_, + l_self_modules_blocks_modules_19_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_142 = ( + l_self_modules_blocks_modules_19_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_19_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_19_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_19_modules_3_parameters_bias_ = None + x = torch.nn.functional.adaptive_avg_pool2d(input_143, 1) + input_143 = None + x_1 = x.flatten(1, -1) + x = None + x_2 = torch.nn.functional.dropout(x_1, 0.0, False, False) + x_1 = None + x_3 = torch._C._nn.linear( + x_2, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_2 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_3,) diff --git a/samples/timm/convmixer_1536_20.in1k/weight_meta.py b/samples/timm/convmixer_1536_20.in1k/weight_meta.py new file mode 100644 index 0000000000..4f2db7d617 --- /dev/null +++ b/samples/timm/convmixer_1536_20.in1k/weight_meta.py @@ -0,0 +1,2498 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [1536, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.231 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_buffers_running_mean_: + name = "L_self_modules_stem_modules_2_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_buffers_running_var_: + name = "L_self_modules_stem_modules_2_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_parameters_weight_: + name = "L_self_modules_stem_modules_2_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_parameters_bias_: + name = "L_self_modules_stem_modules_2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1536, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.066 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_1_parameters_weight_" + shape = [1536, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.014 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_0_modules_3_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_0_modules_3_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_3_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_3_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1536, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_1_parameters_weight_" + shape = [1536, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_1_modules_3_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_1_modules_3_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_3_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_3_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1536, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_1_parameters_weight_" + shape = [1536, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_2_modules_3_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_2_modules_3_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_3_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_3_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1536, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_1_parameters_weight_" + shape = [1536, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_3_modules_3_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_3_modules_3_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_3_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_3_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1536, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_1_parameters_weight_" + shape = [1536, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.014 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_4_modules_3_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_4_modules_3_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_3_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_3_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1536, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_1_parameters_weight_" + shape = [1536, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_5_modules_3_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_5_modules_3_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_3_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_3_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1536, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_1_parameters_weight_" + shape = [1536, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_6_modules_3_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_6_modules_3_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_3_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_3_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1536, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_1_parameters_weight_" + shape = [1536, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_7_modules_3_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_7_modules_3_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_3_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_3_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1536, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_1_parameters_weight_" + shape = [1536, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.014 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_8_modules_3_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_8_modules_3_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_3_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_3_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1536, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_1_parameters_weight_" + shape = [1536, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_9_modules_3_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_9_modules_3_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_3_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_3_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1536, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_1_parameters_weight_" + shape = [1536, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_10_modules_3_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_10_modules_3_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_3_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_3_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1536, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.066 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_1_parameters_weight_" + shape = [1536, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_11_modules_3_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_11_modules_3_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_3_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_3_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1536, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_1_parameters_weight_" + shape = [1536, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_12_modules_3_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_12_modules_3_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_3_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_3_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1536, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_1_parameters_weight_" + shape = [1536, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_13_modules_3_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_13_modules_3_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_3_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_3_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1536, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_1_parameters_weight_" + shape = [1536, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_14_modules_3_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_14_modules_3_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_3_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_3_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1536, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_1_parameters_weight_" + shape = [1536, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_15_modules_3_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_15_modules_3_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_3_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_3_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1536, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_1_parameters_weight_" + shape = [1536, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_16_modules_3_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_16_modules_3_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_3_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_3_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1536, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_1_parameters_weight_" + shape = [1536, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_17_modules_3_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_17_modules_3_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_3_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_3_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1536, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_1_parameters_weight_" + shape = [1536, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_18_modules_3_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_18_modules_3_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_3_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_3_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [1536, 1, 9, 9] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_1_parameters_weight_" + shape = [1536, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_19_modules_3_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_19_modules_3_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_3_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_3_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None diff --git a/samples/timm/convmixer_768_32.in1k/graph_hash.txt b/samples/timm/convmixer_768_32.in1k/graph_hash.txt new file mode 100644 index 0000000000..2b5460c93e --- /dev/null +++ b/samples/timm/convmixer_768_32.in1k/graph_hash.txt @@ -0,0 +1 @@ +a5995aa6a86d9bad67ca93668dd266a0487abe7f4513c124c4966124061c39ac \ No newline at end of file diff --git a/samples/timm/beit3_large_patch16_224.pt/graph_net.json b/samples/timm/convmixer_768_32.in1k/graph_net.json similarity index 100% rename from samples/timm/beit3_large_patch16_224.pt/graph_net.json rename to samples/timm/convmixer_768_32.in1k/graph_net.json diff --git a/samples/timm/beit3_large_patch16_224.pt/input_meta.py b/samples/timm/convmixer_768_32.in1k/input_meta.py similarity index 100% rename from samples/timm/beit3_large_patch16_224.pt/input_meta.py rename to samples/timm/convmixer_768_32.in1k/input_meta.py diff --git a/samples/timm/beit3_large_patch16_224.pt/input_tensor_constraints.py b/samples/timm/convmixer_768_32.in1k/input_tensor_constraints.py similarity index 100% rename from samples/timm/beit3_large_patch16_224.pt/input_tensor_constraints.py rename to samples/timm/convmixer_768_32.in1k/input_tensor_constraints.py diff --git a/samples/timm/convmixer_768_32.in1k/model.py b/samples/timm/convmixer_768_32.in1k/model.py new file mode 100644 index 0000000000..f759c4bac4 --- /dev/null +++ b/samples/timm/convmixer_768_32.in1k/model.py @@ -0,0 +1,3028 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_0_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_0_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_1_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_1_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_2_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_2_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_3_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_3_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_4_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_4_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_5_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_5_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_6_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_6_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_7_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_7_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_8_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_8_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_9_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_9_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_10_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_10_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_11_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_11_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_12_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_12_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_13_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_13_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_14_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_14_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_15_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_15_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_16_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_16_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_17_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_17_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_18_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_18_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_19_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_19_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_20_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_20_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_21_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_21_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_22_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_22_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_23_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_23_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_24_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_24_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_25_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_25_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_26_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_26_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_27_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_27_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_28_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_28_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_29_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_29_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_30_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_30_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_0_modules_fn_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_0_modules_fn_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_3_buffers_running_mean_: torch.Tensor, + L_self_modules_blocks_modules_31_modules_3_buffers_running_var_: torch.Tensor, + L_self_modules_blocks_modules_31_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_3_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_2_buffers_running_mean_ = ( + L_self_modules_stem_modules_2_buffers_running_mean_ + ) + l_self_modules_stem_modules_2_buffers_running_var_ = ( + L_self_modules_stem_modules_2_buffers_running_var_ + ) + l_self_modules_stem_modules_2_parameters_weight_ = ( + L_self_modules_stem_modules_2_parameters_weight_ + ) + l_self_modules_stem_modules_2_parameters_bias_ = ( + L_self_modules_stem_modules_2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_0_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_0_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_0_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_0_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_1_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_1_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_1_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_1_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_2_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_2_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_2_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_2_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_3_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_3_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_3_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_3_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_3_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_4_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_4_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_4_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_4_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_4_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_5_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_5_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_5_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_5_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_5_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_6_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_6_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_6_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_6_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_6_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_7_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_7_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_7_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_7_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_7_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_8_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_8_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_8_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_8_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_8_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_9_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_9_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_9_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_9_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_9_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_10_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_10_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_10_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_10_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_10_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_11_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_11_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_11_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_11_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_11_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_12_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_12_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_12_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_12_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_12_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_13_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_13_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_13_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_13_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_13_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_14_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_14_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_14_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_14_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_14_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_15_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_15_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_15_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_15_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_15_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_16_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_16_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_16_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_16_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_16_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_17_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_17_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_17_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_17_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_17_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_18_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_18_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_18_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_18_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_18_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_19_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_19_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_19_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_19_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_19_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_20_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_20_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_20_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_20_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_20_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_20_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_20_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_20_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_21_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_21_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_21_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_21_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_21_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_21_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_21_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_21_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_22_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_22_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_22_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_22_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_22_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_22_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_22_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_22_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_23_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_23_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_23_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_23_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_23_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_23_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_23_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_23_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_24_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_24_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_24_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_24_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_24_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_24_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_24_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_24_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_24_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_24_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_24_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_24_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_24_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_24_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_24_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_24_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_25_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_25_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_25_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_25_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_25_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_25_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_25_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_25_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_25_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_25_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_25_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_25_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_25_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_25_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_25_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_25_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_26_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_26_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_26_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_26_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_26_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_26_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_26_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_26_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_26_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_26_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_26_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_26_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_26_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_26_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_26_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_26_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_27_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_27_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_27_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_27_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_27_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_27_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_27_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_27_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_27_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_27_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_27_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_27_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_27_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_27_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_27_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_27_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_28_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_28_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_28_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_28_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_28_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_28_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_28_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_28_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_28_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_28_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_28_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_28_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_28_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_28_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_28_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_28_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_29_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_29_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_29_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_29_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_29_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_29_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_29_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_29_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_29_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_29_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_29_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_29_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_29_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_29_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_29_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_29_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_30_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_30_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_30_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_30_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_30_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_30_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_30_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_30_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_30_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_30_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_30_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_30_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_30_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_30_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_30_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_30_modules_3_parameters_bias_ + ) + l_self_modules_blocks_modules_31_modules_0_modules_fn_modules_0_parameters_weight_ = L_self_modules_blocks_modules_31_modules_0_modules_fn_modules_0_parameters_weight_ + l_self_modules_blocks_modules_31_modules_0_modules_fn_modules_0_parameters_bias_ = L_self_modules_blocks_modules_31_modules_0_modules_fn_modules_0_parameters_bias_ + l_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_buffers_running_mean_ = L_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_buffers_running_mean_ + l_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_buffers_running_var_ = L_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_buffers_running_var_ + l_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_parameters_weight_ = L_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_parameters_weight_ + l_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_parameters_bias_ = L_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_parameters_bias_ + l_self_modules_blocks_modules_31_modules_1_parameters_weight_ = ( + L_self_modules_blocks_modules_31_modules_1_parameters_weight_ + ) + l_self_modules_blocks_modules_31_modules_1_parameters_bias_ = ( + L_self_modules_blocks_modules_31_modules_1_parameters_bias_ + ) + l_self_modules_blocks_modules_31_modules_3_buffers_running_mean_ = ( + L_self_modules_blocks_modules_31_modules_3_buffers_running_mean_ + ) + l_self_modules_blocks_modules_31_modules_3_buffers_running_var_ = ( + L_self_modules_blocks_modules_31_modules_3_buffers_running_var_ + ) + l_self_modules_blocks_modules_31_modules_3_parameters_weight_ = ( + L_self_modules_blocks_modules_31_modules_3_parameters_weight_ + ) + l_self_modules_blocks_modules_31_modules_3_parameters_bias_ = ( + L_self_modules_blocks_modules_31_modules_3_parameters_bias_ + ) + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (7, 7), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + input_2 = torch.nn.functional.relu(input_1, inplace=False) + input_1 = None + input_3 = torch.nn.functional.batch_norm( + input_2, + l_self_modules_stem_modules_2_buffers_running_mean_, + l_self_modules_stem_modules_2_buffers_running_var_, + l_self_modules_stem_modules_2_parameters_weight_, + l_self_modules_stem_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_2 = ( + l_self_modules_stem_modules_2_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_2_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_2_parameters_weight_ + ) = l_self_modules_stem_modules_2_parameters_bias_ = None + input_4 = torch.conv2d( + input_3, + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_5 = torch.nn.functional.relu(input_4, inplace=False) + input_4 = None + input_6 = torch.nn.functional.batch_norm( + input_5, + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_5 = l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_7 = input_6 + input_3 + input_6 = input_3 = None + input_8 = torch.conv2d( + input_7, + l_self_modules_blocks_modules_0_modules_1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_7 = ( + l_self_modules_blocks_modules_0_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_0_modules_1_parameters_bias_ = None + input_9 = torch.nn.functional.relu(input_8, inplace=False) + input_8 = None + input_10 = torch.nn.functional.batch_norm( + input_9, + l_self_modules_blocks_modules_0_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_0_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_0_modules_3_parameters_weight_, + l_self_modules_blocks_modules_0_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_9 = ( + l_self_modules_blocks_modules_0_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_0_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_0_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_0_modules_3_parameters_bias_ = None + input_11 = torch.conv2d( + input_10, + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_12 = torch.nn.functional.relu(input_11, inplace=False) + input_11 = None + input_13 = torch.nn.functional.batch_norm( + input_12, + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_12 = l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_14 = input_13 + input_10 + input_13 = input_10 = None + input_15 = torch.conv2d( + input_14, + l_self_modules_blocks_modules_1_modules_1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_14 = ( + l_self_modules_blocks_modules_1_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_1_modules_1_parameters_bias_ = None + input_16 = torch.nn.functional.relu(input_15, inplace=False) + input_15 = None + input_17 = torch.nn.functional.batch_norm( + input_16, + l_self_modules_blocks_modules_1_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_1_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_1_modules_3_parameters_weight_, + l_self_modules_blocks_modules_1_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_16 = ( + l_self_modules_blocks_modules_1_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_1_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_1_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_1_modules_3_parameters_bias_ = None + input_18 = torch.conv2d( + input_17, + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_19 = torch.nn.functional.relu(input_18, inplace=False) + input_18 = None + input_20 = torch.nn.functional.batch_norm( + input_19, + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_19 = l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_21 = input_20 + input_17 + input_20 = input_17 = None + input_22 = torch.conv2d( + input_21, + l_self_modules_blocks_modules_2_modules_1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_21 = ( + l_self_modules_blocks_modules_2_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_2_modules_1_parameters_bias_ = None + input_23 = torch.nn.functional.relu(input_22, inplace=False) + input_22 = None + input_24 = torch.nn.functional.batch_norm( + input_23, + l_self_modules_blocks_modules_2_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_2_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_2_modules_3_parameters_weight_, + l_self_modules_blocks_modules_2_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_23 = ( + l_self_modules_blocks_modules_2_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_2_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_2_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_2_modules_3_parameters_bias_ = None + input_25 = torch.conv2d( + input_24, + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_26 = torch.nn.functional.relu(input_25, inplace=False) + input_25 = None + input_27 = torch.nn.functional.batch_norm( + input_26, + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_26 = l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_28 = input_27 + input_24 + input_27 = input_24 = None + input_29 = torch.conv2d( + input_28, + l_self_modules_blocks_modules_3_modules_1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_28 = ( + l_self_modules_blocks_modules_3_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_3_modules_1_parameters_bias_ = None + input_30 = torch.nn.functional.relu(input_29, inplace=False) + input_29 = None + input_31 = torch.nn.functional.batch_norm( + input_30, + l_self_modules_blocks_modules_3_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_3_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_3_modules_3_parameters_weight_, + l_self_modules_blocks_modules_3_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_30 = ( + l_self_modules_blocks_modules_3_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_3_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_3_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_3_modules_3_parameters_bias_ = None + input_32 = torch.conv2d( + input_31, + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_33 = torch.nn.functional.relu(input_32, inplace=False) + input_32 = None + input_34 = torch.nn.functional.batch_norm( + input_33, + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_33 = l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_35 = input_34 + input_31 + input_34 = input_31 = None + input_36 = torch.conv2d( + input_35, + l_self_modules_blocks_modules_4_modules_1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_35 = ( + l_self_modules_blocks_modules_4_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_4_modules_1_parameters_bias_ = None + input_37 = torch.nn.functional.relu(input_36, inplace=False) + input_36 = None + input_38 = torch.nn.functional.batch_norm( + input_37, + l_self_modules_blocks_modules_4_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_4_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_4_modules_3_parameters_weight_, + l_self_modules_blocks_modules_4_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_37 = ( + l_self_modules_blocks_modules_4_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_4_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_4_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_4_modules_3_parameters_bias_ = None + input_39 = torch.conv2d( + input_38, + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_40 = torch.nn.functional.relu(input_39, inplace=False) + input_39 = None + input_41 = torch.nn.functional.batch_norm( + input_40, + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_40 = l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_42 = input_41 + input_38 + input_41 = input_38 = None + input_43 = torch.conv2d( + input_42, + l_self_modules_blocks_modules_5_modules_1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_42 = ( + l_self_modules_blocks_modules_5_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_5_modules_1_parameters_bias_ = None + input_44 = torch.nn.functional.relu(input_43, inplace=False) + input_43 = None + input_45 = torch.nn.functional.batch_norm( + input_44, + l_self_modules_blocks_modules_5_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_5_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_5_modules_3_parameters_weight_, + l_self_modules_blocks_modules_5_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_44 = ( + l_self_modules_blocks_modules_5_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_5_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_5_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_5_modules_3_parameters_bias_ = None + input_46 = torch.conv2d( + input_45, + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_47 = torch.nn.functional.relu(input_46, inplace=False) + input_46 = None + input_48 = torch.nn.functional.batch_norm( + input_47, + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_47 = l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_49 = input_48 + input_45 + input_48 = input_45 = None + input_50 = torch.conv2d( + input_49, + l_self_modules_blocks_modules_6_modules_1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_49 = ( + l_self_modules_blocks_modules_6_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_6_modules_1_parameters_bias_ = None + input_51 = torch.nn.functional.relu(input_50, inplace=False) + input_50 = None + input_52 = torch.nn.functional.batch_norm( + input_51, + l_self_modules_blocks_modules_6_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_6_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_6_modules_3_parameters_weight_, + l_self_modules_blocks_modules_6_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_51 = ( + l_self_modules_blocks_modules_6_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_6_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_6_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_6_modules_3_parameters_bias_ = None + input_53 = torch.conv2d( + input_52, + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_54 = torch.nn.functional.relu(input_53, inplace=False) + input_53 = None + input_55 = torch.nn.functional.batch_norm( + input_54, + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_54 = l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_56 = input_55 + input_52 + input_55 = input_52 = None + input_57 = torch.conv2d( + input_56, + l_self_modules_blocks_modules_7_modules_1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_56 = ( + l_self_modules_blocks_modules_7_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_7_modules_1_parameters_bias_ = None + input_58 = torch.nn.functional.relu(input_57, inplace=False) + input_57 = None + input_59 = torch.nn.functional.batch_norm( + input_58, + l_self_modules_blocks_modules_7_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_7_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_7_modules_3_parameters_weight_, + l_self_modules_blocks_modules_7_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_58 = ( + l_self_modules_blocks_modules_7_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_7_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_7_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_7_modules_3_parameters_bias_ = None + input_60 = torch.conv2d( + input_59, + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_61 = torch.nn.functional.relu(input_60, inplace=False) + input_60 = None + input_62 = torch.nn.functional.batch_norm( + input_61, + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_61 = l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_63 = input_62 + input_59 + input_62 = input_59 = None + input_64 = torch.conv2d( + input_63, + l_self_modules_blocks_modules_8_modules_1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_63 = ( + l_self_modules_blocks_modules_8_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_8_modules_1_parameters_bias_ = None + input_65 = torch.nn.functional.relu(input_64, inplace=False) + input_64 = None + input_66 = torch.nn.functional.batch_norm( + input_65, + l_self_modules_blocks_modules_8_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_8_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_8_modules_3_parameters_weight_, + l_self_modules_blocks_modules_8_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_65 = ( + l_self_modules_blocks_modules_8_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_8_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_8_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_8_modules_3_parameters_bias_ = None + input_67 = torch.conv2d( + input_66, + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_68 = torch.nn.functional.relu(input_67, inplace=False) + input_67 = None + input_69 = torch.nn.functional.batch_norm( + input_68, + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_68 = l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_70 = input_69 + input_66 + input_69 = input_66 = None + input_71 = torch.conv2d( + input_70, + l_self_modules_blocks_modules_9_modules_1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_70 = ( + l_self_modules_blocks_modules_9_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_9_modules_1_parameters_bias_ = None + input_72 = torch.nn.functional.relu(input_71, inplace=False) + input_71 = None + input_73 = torch.nn.functional.batch_norm( + input_72, + l_self_modules_blocks_modules_9_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_9_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_9_modules_3_parameters_weight_, + l_self_modules_blocks_modules_9_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_72 = ( + l_self_modules_blocks_modules_9_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_9_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_9_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_9_modules_3_parameters_bias_ = None + input_74 = torch.conv2d( + input_73, + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_75 = torch.nn.functional.relu(input_74, inplace=False) + input_74 = None + input_76 = torch.nn.functional.batch_norm( + input_75, + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_75 = l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_77 = input_76 + input_73 + input_76 = input_73 = None + input_78 = torch.conv2d( + input_77, + l_self_modules_blocks_modules_10_modules_1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_77 = ( + l_self_modules_blocks_modules_10_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_10_modules_1_parameters_bias_ = None + input_79 = torch.nn.functional.relu(input_78, inplace=False) + input_78 = None + input_80 = torch.nn.functional.batch_norm( + input_79, + l_self_modules_blocks_modules_10_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_10_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_10_modules_3_parameters_weight_, + l_self_modules_blocks_modules_10_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_79 = ( + l_self_modules_blocks_modules_10_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_10_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_10_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_10_modules_3_parameters_bias_ = None + input_81 = torch.conv2d( + input_80, + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_82 = torch.nn.functional.relu(input_81, inplace=False) + input_81 = None + input_83 = torch.nn.functional.batch_norm( + input_82, + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_82 = l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_84 = input_83 + input_80 + input_83 = input_80 = None + input_85 = torch.conv2d( + input_84, + l_self_modules_blocks_modules_11_modules_1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_84 = ( + l_self_modules_blocks_modules_11_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_11_modules_1_parameters_bias_ = None + input_86 = torch.nn.functional.relu(input_85, inplace=False) + input_85 = None + input_87 = torch.nn.functional.batch_norm( + input_86, + l_self_modules_blocks_modules_11_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_11_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_11_modules_3_parameters_weight_, + l_self_modules_blocks_modules_11_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_86 = ( + l_self_modules_blocks_modules_11_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_11_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_11_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_11_modules_3_parameters_bias_ = None + input_88 = torch.conv2d( + input_87, + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_89 = torch.nn.functional.relu(input_88, inplace=False) + input_88 = None + input_90 = torch.nn.functional.batch_norm( + input_89, + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_89 = l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_91 = input_90 + input_87 + input_90 = input_87 = None + input_92 = torch.conv2d( + input_91, + l_self_modules_blocks_modules_12_modules_1_parameters_weight_, + l_self_modules_blocks_modules_12_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_91 = ( + l_self_modules_blocks_modules_12_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_12_modules_1_parameters_bias_ = None + input_93 = torch.nn.functional.relu(input_92, inplace=False) + input_92 = None + input_94 = torch.nn.functional.batch_norm( + input_93, + l_self_modules_blocks_modules_12_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_12_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_12_modules_3_parameters_weight_, + l_self_modules_blocks_modules_12_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_93 = ( + l_self_modules_blocks_modules_12_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_12_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_12_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_12_modules_3_parameters_bias_ = None + input_95 = torch.conv2d( + input_94, + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_96 = torch.nn.functional.relu(input_95, inplace=False) + input_95 = None + input_97 = torch.nn.functional.batch_norm( + input_96, + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_96 = l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_98 = input_97 + input_94 + input_97 = input_94 = None + input_99 = torch.conv2d( + input_98, + l_self_modules_blocks_modules_13_modules_1_parameters_weight_, + l_self_modules_blocks_modules_13_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_98 = ( + l_self_modules_blocks_modules_13_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_13_modules_1_parameters_bias_ = None + input_100 = torch.nn.functional.relu(input_99, inplace=False) + input_99 = None + input_101 = torch.nn.functional.batch_norm( + input_100, + l_self_modules_blocks_modules_13_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_13_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_13_modules_3_parameters_weight_, + l_self_modules_blocks_modules_13_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_100 = ( + l_self_modules_blocks_modules_13_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_13_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_13_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_13_modules_3_parameters_bias_ = None + input_102 = torch.conv2d( + input_101, + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_103 = torch.nn.functional.relu(input_102, inplace=False) + input_102 = None + input_104 = torch.nn.functional.batch_norm( + input_103, + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_103 = l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_105 = input_104 + input_101 + input_104 = input_101 = None + input_106 = torch.conv2d( + input_105, + l_self_modules_blocks_modules_14_modules_1_parameters_weight_, + l_self_modules_blocks_modules_14_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_105 = ( + l_self_modules_blocks_modules_14_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_14_modules_1_parameters_bias_ = None + input_107 = torch.nn.functional.relu(input_106, inplace=False) + input_106 = None + input_108 = torch.nn.functional.batch_norm( + input_107, + l_self_modules_blocks_modules_14_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_14_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_14_modules_3_parameters_weight_, + l_self_modules_blocks_modules_14_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_107 = ( + l_self_modules_blocks_modules_14_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_14_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_14_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_14_modules_3_parameters_bias_ = None + input_109 = torch.conv2d( + input_108, + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_110 = torch.nn.functional.relu(input_109, inplace=False) + input_109 = None + input_111 = torch.nn.functional.batch_norm( + input_110, + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_110 = l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_112 = input_111 + input_108 + input_111 = input_108 = None + input_113 = torch.conv2d( + input_112, + l_self_modules_blocks_modules_15_modules_1_parameters_weight_, + l_self_modules_blocks_modules_15_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_112 = ( + l_self_modules_blocks_modules_15_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_15_modules_1_parameters_bias_ = None + input_114 = torch.nn.functional.relu(input_113, inplace=False) + input_113 = None + input_115 = torch.nn.functional.batch_norm( + input_114, + l_self_modules_blocks_modules_15_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_15_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_15_modules_3_parameters_weight_, + l_self_modules_blocks_modules_15_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_114 = ( + l_self_modules_blocks_modules_15_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_15_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_15_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_15_modules_3_parameters_bias_ = None + input_116 = torch.conv2d( + input_115, + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_117 = torch.nn.functional.relu(input_116, inplace=False) + input_116 = None + input_118 = torch.nn.functional.batch_norm( + input_117, + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_117 = l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_119 = input_118 + input_115 + input_118 = input_115 = None + input_120 = torch.conv2d( + input_119, + l_self_modules_blocks_modules_16_modules_1_parameters_weight_, + l_self_modules_blocks_modules_16_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_119 = ( + l_self_modules_blocks_modules_16_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_16_modules_1_parameters_bias_ = None + input_121 = torch.nn.functional.relu(input_120, inplace=False) + input_120 = None + input_122 = torch.nn.functional.batch_norm( + input_121, + l_self_modules_blocks_modules_16_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_16_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_16_modules_3_parameters_weight_, + l_self_modules_blocks_modules_16_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_121 = ( + l_self_modules_blocks_modules_16_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_16_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_16_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_16_modules_3_parameters_bias_ = None + input_123 = torch.conv2d( + input_122, + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_124 = torch.nn.functional.relu(input_123, inplace=False) + input_123 = None + input_125 = torch.nn.functional.batch_norm( + input_124, + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_124 = l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_126 = input_125 + input_122 + input_125 = input_122 = None + input_127 = torch.conv2d( + input_126, + l_self_modules_blocks_modules_17_modules_1_parameters_weight_, + l_self_modules_blocks_modules_17_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_126 = ( + l_self_modules_blocks_modules_17_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_17_modules_1_parameters_bias_ = None + input_128 = torch.nn.functional.relu(input_127, inplace=False) + input_127 = None + input_129 = torch.nn.functional.batch_norm( + input_128, + l_self_modules_blocks_modules_17_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_17_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_17_modules_3_parameters_weight_, + l_self_modules_blocks_modules_17_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_128 = ( + l_self_modules_blocks_modules_17_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_17_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_17_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_17_modules_3_parameters_bias_ = None + input_130 = torch.conv2d( + input_129, + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_131 = torch.nn.functional.relu(input_130, inplace=False) + input_130 = None + input_132 = torch.nn.functional.batch_norm( + input_131, + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_131 = l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_133 = input_132 + input_129 + input_132 = input_129 = None + input_134 = torch.conv2d( + input_133, + l_self_modules_blocks_modules_18_modules_1_parameters_weight_, + l_self_modules_blocks_modules_18_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_133 = ( + l_self_modules_blocks_modules_18_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_18_modules_1_parameters_bias_ = None + input_135 = torch.nn.functional.relu(input_134, inplace=False) + input_134 = None + input_136 = torch.nn.functional.batch_norm( + input_135, + l_self_modules_blocks_modules_18_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_18_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_18_modules_3_parameters_weight_, + l_self_modules_blocks_modules_18_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_135 = ( + l_self_modules_blocks_modules_18_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_18_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_18_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_18_modules_3_parameters_bias_ = None + input_137 = torch.conv2d( + input_136, + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_138 = torch.nn.functional.relu(input_137, inplace=False) + input_137 = None + input_139 = torch.nn.functional.batch_norm( + input_138, + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_138 = l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_140 = input_139 + input_136 + input_139 = input_136 = None + input_141 = torch.conv2d( + input_140, + l_self_modules_blocks_modules_19_modules_1_parameters_weight_, + l_self_modules_blocks_modules_19_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_140 = ( + l_self_modules_blocks_modules_19_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_19_modules_1_parameters_bias_ = None + input_142 = torch.nn.functional.relu(input_141, inplace=False) + input_141 = None + input_143 = torch.nn.functional.batch_norm( + input_142, + l_self_modules_blocks_modules_19_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_19_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_19_modules_3_parameters_weight_, + l_self_modules_blocks_modules_19_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_142 = ( + l_self_modules_blocks_modules_19_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_19_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_19_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_19_modules_3_parameters_bias_ = None + input_144 = torch.conv2d( + input_143, + l_self_modules_blocks_modules_20_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_20_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_20_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_20_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_145 = torch.nn.functional.relu(input_144, inplace=False) + input_144 = None + input_146 = torch.nn.functional.batch_norm( + input_145, + l_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_145 = l_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_147 = input_146 + input_143 + input_146 = input_143 = None + input_148 = torch.conv2d( + input_147, + l_self_modules_blocks_modules_20_modules_1_parameters_weight_, + l_self_modules_blocks_modules_20_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_147 = ( + l_self_modules_blocks_modules_20_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_20_modules_1_parameters_bias_ = None + input_149 = torch.nn.functional.relu(input_148, inplace=False) + input_148 = None + input_150 = torch.nn.functional.batch_norm( + input_149, + l_self_modules_blocks_modules_20_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_20_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_20_modules_3_parameters_weight_, + l_self_modules_blocks_modules_20_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_149 = ( + l_self_modules_blocks_modules_20_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_20_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_20_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_20_modules_3_parameters_bias_ = None + input_151 = torch.conv2d( + input_150, + l_self_modules_blocks_modules_21_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_21_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_21_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_21_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_152 = torch.nn.functional.relu(input_151, inplace=False) + input_151 = None + input_153 = torch.nn.functional.batch_norm( + input_152, + l_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_152 = l_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_154 = input_153 + input_150 + input_153 = input_150 = None + input_155 = torch.conv2d( + input_154, + l_self_modules_blocks_modules_21_modules_1_parameters_weight_, + l_self_modules_blocks_modules_21_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_154 = ( + l_self_modules_blocks_modules_21_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_21_modules_1_parameters_bias_ = None + input_156 = torch.nn.functional.relu(input_155, inplace=False) + input_155 = None + input_157 = torch.nn.functional.batch_norm( + input_156, + l_self_modules_blocks_modules_21_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_21_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_21_modules_3_parameters_weight_, + l_self_modules_blocks_modules_21_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_156 = ( + l_self_modules_blocks_modules_21_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_21_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_21_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_21_modules_3_parameters_bias_ = None + input_158 = torch.conv2d( + input_157, + l_self_modules_blocks_modules_22_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_22_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_22_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_22_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_159 = torch.nn.functional.relu(input_158, inplace=False) + input_158 = None + input_160 = torch.nn.functional.batch_norm( + input_159, + l_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_159 = l_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_161 = input_160 + input_157 + input_160 = input_157 = None + input_162 = torch.conv2d( + input_161, + l_self_modules_blocks_modules_22_modules_1_parameters_weight_, + l_self_modules_blocks_modules_22_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_161 = ( + l_self_modules_blocks_modules_22_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_22_modules_1_parameters_bias_ = None + input_163 = torch.nn.functional.relu(input_162, inplace=False) + input_162 = None + input_164 = torch.nn.functional.batch_norm( + input_163, + l_self_modules_blocks_modules_22_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_22_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_22_modules_3_parameters_weight_, + l_self_modules_blocks_modules_22_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_163 = ( + l_self_modules_blocks_modules_22_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_22_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_22_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_22_modules_3_parameters_bias_ = None + input_165 = torch.conv2d( + input_164, + l_self_modules_blocks_modules_23_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_23_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_23_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_23_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_166 = torch.nn.functional.relu(input_165, inplace=False) + input_165 = None + input_167 = torch.nn.functional.batch_norm( + input_166, + l_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_166 = l_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_168 = input_167 + input_164 + input_167 = input_164 = None + input_169 = torch.conv2d( + input_168, + l_self_modules_blocks_modules_23_modules_1_parameters_weight_, + l_self_modules_blocks_modules_23_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_168 = ( + l_self_modules_blocks_modules_23_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_23_modules_1_parameters_bias_ = None + input_170 = torch.nn.functional.relu(input_169, inplace=False) + input_169 = None + input_171 = torch.nn.functional.batch_norm( + input_170, + l_self_modules_blocks_modules_23_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_23_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_23_modules_3_parameters_weight_, + l_self_modules_blocks_modules_23_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_170 = ( + l_self_modules_blocks_modules_23_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_23_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_23_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_23_modules_3_parameters_bias_ = None + input_172 = torch.conv2d( + input_171, + l_self_modules_blocks_modules_24_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_24_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_24_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_24_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_173 = torch.nn.functional.relu(input_172, inplace=False) + input_172 = None + input_174 = torch.nn.functional.batch_norm( + input_173, + l_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_173 = l_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_175 = input_174 + input_171 + input_174 = input_171 = None + input_176 = torch.conv2d( + input_175, + l_self_modules_blocks_modules_24_modules_1_parameters_weight_, + l_self_modules_blocks_modules_24_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_175 = ( + l_self_modules_blocks_modules_24_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_24_modules_1_parameters_bias_ = None + input_177 = torch.nn.functional.relu(input_176, inplace=False) + input_176 = None + input_178 = torch.nn.functional.batch_norm( + input_177, + l_self_modules_blocks_modules_24_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_24_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_24_modules_3_parameters_weight_, + l_self_modules_blocks_modules_24_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_177 = ( + l_self_modules_blocks_modules_24_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_24_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_24_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_24_modules_3_parameters_bias_ = None + input_179 = torch.conv2d( + input_178, + l_self_modules_blocks_modules_25_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_25_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_25_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_25_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_180 = torch.nn.functional.relu(input_179, inplace=False) + input_179 = None + input_181 = torch.nn.functional.batch_norm( + input_180, + l_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_180 = l_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_182 = input_181 + input_178 + input_181 = input_178 = None + input_183 = torch.conv2d( + input_182, + l_self_modules_blocks_modules_25_modules_1_parameters_weight_, + l_self_modules_blocks_modules_25_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_182 = ( + l_self_modules_blocks_modules_25_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_25_modules_1_parameters_bias_ = None + input_184 = torch.nn.functional.relu(input_183, inplace=False) + input_183 = None + input_185 = torch.nn.functional.batch_norm( + input_184, + l_self_modules_blocks_modules_25_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_25_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_25_modules_3_parameters_weight_, + l_self_modules_blocks_modules_25_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_184 = ( + l_self_modules_blocks_modules_25_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_25_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_25_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_25_modules_3_parameters_bias_ = None + input_186 = torch.conv2d( + input_185, + l_self_modules_blocks_modules_26_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_26_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_26_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_26_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_187 = torch.nn.functional.relu(input_186, inplace=False) + input_186 = None + input_188 = torch.nn.functional.batch_norm( + input_187, + l_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_187 = l_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_189 = input_188 + input_185 + input_188 = input_185 = None + input_190 = torch.conv2d( + input_189, + l_self_modules_blocks_modules_26_modules_1_parameters_weight_, + l_self_modules_blocks_modules_26_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_189 = ( + l_self_modules_blocks_modules_26_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_26_modules_1_parameters_bias_ = None + input_191 = torch.nn.functional.relu(input_190, inplace=False) + input_190 = None + input_192 = torch.nn.functional.batch_norm( + input_191, + l_self_modules_blocks_modules_26_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_26_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_26_modules_3_parameters_weight_, + l_self_modules_blocks_modules_26_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_191 = ( + l_self_modules_blocks_modules_26_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_26_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_26_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_26_modules_3_parameters_bias_ = None + input_193 = torch.conv2d( + input_192, + l_self_modules_blocks_modules_27_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_27_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_27_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_27_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_194 = torch.nn.functional.relu(input_193, inplace=False) + input_193 = None + input_195 = torch.nn.functional.batch_norm( + input_194, + l_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_194 = l_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_196 = input_195 + input_192 + input_195 = input_192 = None + input_197 = torch.conv2d( + input_196, + l_self_modules_blocks_modules_27_modules_1_parameters_weight_, + l_self_modules_blocks_modules_27_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_196 = ( + l_self_modules_blocks_modules_27_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_27_modules_1_parameters_bias_ = None + input_198 = torch.nn.functional.relu(input_197, inplace=False) + input_197 = None + input_199 = torch.nn.functional.batch_norm( + input_198, + l_self_modules_blocks_modules_27_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_27_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_27_modules_3_parameters_weight_, + l_self_modules_blocks_modules_27_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_198 = ( + l_self_modules_blocks_modules_27_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_27_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_27_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_27_modules_3_parameters_bias_ = None + input_200 = torch.conv2d( + input_199, + l_self_modules_blocks_modules_28_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_28_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_28_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_28_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_201 = torch.nn.functional.relu(input_200, inplace=False) + input_200 = None + input_202 = torch.nn.functional.batch_norm( + input_201, + l_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_201 = l_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_203 = input_202 + input_199 + input_202 = input_199 = None + input_204 = torch.conv2d( + input_203, + l_self_modules_blocks_modules_28_modules_1_parameters_weight_, + l_self_modules_blocks_modules_28_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_203 = ( + l_self_modules_blocks_modules_28_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_28_modules_1_parameters_bias_ = None + input_205 = torch.nn.functional.relu(input_204, inplace=False) + input_204 = None + input_206 = torch.nn.functional.batch_norm( + input_205, + l_self_modules_blocks_modules_28_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_28_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_28_modules_3_parameters_weight_, + l_self_modules_blocks_modules_28_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_205 = ( + l_self_modules_blocks_modules_28_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_28_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_28_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_28_modules_3_parameters_bias_ = None + input_207 = torch.conv2d( + input_206, + l_self_modules_blocks_modules_29_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_29_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_29_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_29_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_208 = torch.nn.functional.relu(input_207, inplace=False) + input_207 = None + input_209 = torch.nn.functional.batch_norm( + input_208, + l_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_208 = l_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_210 = input_209 + input_206 + input_209 = input_206 = None + input_211 = torch.conv2d( + input_210, + l_self_modules_blocks_modules_29_modules_1_parameters_weight_, + l_self_modules_blocks_modules_29_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_210 = ( + l_self_modules_blocks_modules_29_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_29_modules_1_parameters_bias_ = None + input_212 = torch.nn.functional.relu(input_211, inplace=False) + input_211 = None + input_213 = torch.nn.functional.batch_norm( + input_212, + l_self_modules_blocks_modules_29_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_29_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_29_modules_3_parameters_weight_, + l_self_modules_blocks_modules_29_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_212 = ( + l_self_modules_blocks_modules_29_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_29_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_29_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_29_modules_3_parameters_bias_ = None + input_214 = torch.conv2d( + input_213, + l_self_modules_blocks_modules_30_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_30_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_30_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_30_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_215 = torch.nn.functional.relu(input_214, inplace=False) + input_214 = None + input_216 = torch.nn.functional.batch_norm( + input_215, + l_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_215 = l_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_217 = input_216 + input_213 + input_216 = input_213 = None + input_218 = torch.conv2d( + input_217, + l_self_modules_blocks_modules_30_modules_1_parameters_weight_, + l_self_modules_blocks_modules_30_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_217 = ( + l_self_modules_blocks_modules_30_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_30_modules_1_parameters_bias_ = None + input_219 = torch.nn.functional.relu(input_218, inplace=False) + input_218 = None + input_220 = torch.nn.functional.batch_norm( + input_219, + l_self_modules_blocks_modules_30_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_30_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_30_modules_3_parameters_weight_, + l_self_modules_blocks_modules_30_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_219 = ( + l_self_modules_blocks_modules_30_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_30_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_30_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_30_modules_3_parameters_bias_ = None + input_221 = torch.conv2d( + input_220, + l_self_modules_blocks_modules_31_modules_0_modules_fn_modules_0_parameters_weight_, + l_self_modules_blocks_modules_31_modules_0_modules_fn_modules_0_parameters_bias_, + (1, 1), + "same", + (1, 1), + 768, + ) + l_self_modules_blocks_modules_31_modules_0_modules_fn_modules_0_parameters_weight_ = l_self_modules_blocks_modules_31_modules_0_modules_fn_modules_0_parameters_bias_ = (None) + input_222 = torch.nn.functional.relu(input_221, inplace=False) + input_221 = None + input_223 = torch.nn.functional.batch_norm( + input_222, + l_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_buffers_running_mean_, + l_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_buffers_running_var_, + l_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_parameters_weight_, + l_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_222 = l_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_buffers_running_mean_ = l_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_buffers_running_var_ = l_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_parameters_weight_ = l_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_parameters_bias_ = (None) + input_224 = input_223 + input_220 + input_223 = input_220 = None + input_225 = torch.conv2d( + input_224, + l_self_modules_blocks_modules_31_modules_1_parameters_weight_, + l_self_modules_blocks_modules_31_modules_1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_224 = ( + l_self_modules_blocks_modules_31_modules_1_parameters_weight_ + ) = l_self_modules_blocks_modules_31_modules_1_parameters_bias_ = None + input_226 = torch.nn.functional.relu(input_225, inplace=False) + input_225 = None + input_227 = torch.nn.functional.batch_norm( + input_226, + l_self_modules_blocks_modules_31_modules_3_buffers_running_mean_, + l_self_modules_blocks_modules_31_modules_3_buffers_running_var_, + l_self_modules_blocks_modules_31_modules_3_parameters_weight_, + l_self_modules_blocks_modules_31_modules_3_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_226 = ( + l_self_modules_blocks_modules_31_modules_3_buffers_running_mean_ + ) = ( + l_self_modules_blocks_modules_31_modules_3_buffers_running_var_ + ) = ( + l_self_modules_blocks_modules_31_modules_3_parameters_weight_ + ) = l_self_modules_blocks_modules_31_modules_3_parameters_bias_ = None + x = torch.nn.functional.adaptive_avg_pool2d(input_227, 1) + input_227 = None + x_1 = x.flatten(1, -1) + x = None + x_2 = torch.nn.functional.dropout(x_1, 0.0, False, False) + x_1 = None + x_3 = torch._C._nn.linear( + x_2, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_2 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_3,) diff --git a/samples/timm/convmixer_768_32.in1k/weight_meta.py b/samples/timm/convmixer_768_32.in1k/weight_meta.py new file mode 100644 index 0000000000..0f2a57dad9 --- /dev/null +++ b/samples/timm/convmixer_768_32.in1k/weight_meta.py @@ -0,0 +1,3938 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [768, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.232 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_buffers_running_mean_: + name = "L_self_modules_stem_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_buffers_running_var_: + name = "L_self_modules_stem_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_parameters_weight_: + name = "L_self_modules_stem_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_parameters_bias_: + name = "L_self_modules_stem_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.086 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_0_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_0_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_1_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_1_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_2_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_2_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_3_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_3_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_4_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_4_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_5_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_5_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.084 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_6_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_6_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.004 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_7_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_7_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_8_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_8_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.084 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_9_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_9_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.084 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_10_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_10_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_11_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_11_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.084 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_12_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_12_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_13_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_13_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.081 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_14_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_14_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.005 + std = 0.081 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_15_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_15_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_16_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_16_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.084 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_17_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_17_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_18_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_18_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_19_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_19_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_20_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_20_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_21_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_21_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_22_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_22_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.080 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_23_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_23_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_24_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_24_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.084 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_24_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_24_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_24_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_24_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_24_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_24_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_24_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_25_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_25_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.084 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_25_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_25_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_25_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_25_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_25_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_25_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_25_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_26_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_26_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_26_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_26_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_26_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_26_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_26_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_26_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_26_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_27_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_27_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.085 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_27_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_27_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_27_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_27_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_27_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_27_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_27_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_28_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_28_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.004 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_28_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_28_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_28_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_28_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_28_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_28_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_28_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_29_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_29_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.005 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_29_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_29_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_29_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_29_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_29_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_29_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_29_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_30_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_30_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.085 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_30_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_30_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_30_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_30_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_30_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_30_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_30_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_0_modules_fn_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_31_modules_0_modules_fn_modules_0_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.082 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_0_modules_fn_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_31_modules_0_modules_fn_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.085 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_buffers_running_mean_: + name = "L_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_buffers_running_var_: + name = "L_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_31_modules_0_modules_fn_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_1_parameters_weight_: + name = "L_self_modules_blocks_modules_31_modules_1_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_1_parameters_bias_: + name = "L_self_modules_blocks_modules_31_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_3_buffers_running_mean_: + name = "L_self_modules_blocks_modules_31_modules_3_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_3_buffers_running_var_: + name = "L_self_modules_blocks_modules_31_modules_3_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_3_parameters_weight_: + name = "L_self_modules_blocks_modules_31_modules_3_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_3_parameters_bias_: + name = "L_self_modules_blocks_modules_31_modules_3_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None diff --git a/samples/timm/convnext_atto.d2_in1k/graph_hash.txt b/samples/timm/convnext_atto.d2_in1k/graph_hash.txt new file mode 100644 index 0000000000..eedf65667f --- /dev/null +++ b/samples/timm/convnext_atto.d2_in1k/graph_hash.txt @@ -0,0 +1 @@ +84d3d5845f8731d40d97c48baa7a6700f9f61d9504a55d27b952141745e34104 \ No newline at end of file diff --git a/samples/timm/convnext_atto.d2_in1k/graph_net.json b/samples/timm/convnext_atto.d2_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnext_atto.d2_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_atto.d2_in1k/input_meta.py b/samples/timm/convnext_atto.d2_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_atto.d2_in1k/input_tensor_constraints.py b/samples/timm/convnext_atto.d2_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_atto.d2_in1k/model.py b/samples/timm/convnext_atto.d2_in1k/model.py new file mode 100644 index 0000000000..e4dfcc936c --- /dev/null +++ b/samples/timm/convnext_atto.d2_in1k/model.py @@ -0,0 +1,1123 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (40,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 40, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (40,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 40, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (40,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_21 = torch.conv2d( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (40,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_2 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (80,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + reshape_2 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_37 = x_36.mul(reshape_2) + x_36 = reshape_2 = None + x_38 = x_37 + input_2 + x_37 = input_2 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (80,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = torch.conv2d( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (80,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_3 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_60 = torch.conv2d( + x_59, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_59 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + reshape_4 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_62 = x_61.mul(reshape_4) + x_61 = reshape_4 = None + x_63 = x_62 + input_3 + x_62 = input_3 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_71 = torch.conv2d( + x_70, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_70 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + reshape_5 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_73 = x_72.mul(reshape_5) + x_72 = reshape_5 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_84 = x_83.mul(reshape_6) + x_83 = reshape_6 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_115 = torch.conv2d( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = x_118.permute(0, 2, 3, 1) + x_118 = None + x_120 = torch.nn.functional.layer_norm( + x_119, + (160,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_119 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_121 = x_120.permute(0, 3, 1, 2) + x_120 = None + input_4 = torch.conv2d( + x_121, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_122 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_123 = x_122.permute(0, 2, 3, 1) + x_122 = None + x_124 = torch.nn.functional.layer_norm( + x_123, + (320,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_125 = x_124.permute(0, 3, 1, 2) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126) + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_129 = torch.conv2d( + x_128, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_128 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) + x_129 = None + reshape_10 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_131 = x_130.mul(reshape_10) + x_130 = reshape_10 = None + x_132 = x_131 + input_4 + x_131 = input_4 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_134 = x_133.permute(0, 2, 3, 1) + x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (320,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_134 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_136 = x_135.permute(0, 3, 1, 2) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137) + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = torch.conv2d( + x_139, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_139 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_141 = torch.nn.functional.dropout(x_140, 0.0, False, False) + x_140 = None + reshape_11 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_142 = x_141.mul(reshape_11) + x_141 = reshape_11 = None + x_143 = x_142 + x_132 + x_142 = x_132 = None + x_144 = torch.nn.functional.adaptive_avg_pool2d(x_143, 1) + x_143 = None + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (320,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = x_147.flatten(1, -1) + x_147 = None + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = torch._C._nn.linear( + x_149, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_149 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_150,) diff --git a/samples/timm/convnext_atto.d2_in1k/weight_meta.py b/samples/timm/convnext_atto.d2_in1k/weight_meta.py new file mode 100644 index 0000000000..2c1d5a91f4 --- /dev/null +++ b/samples/timm/convnext_atto.d2_in1k/weight_meta.py @@ -0,0 +1,1310 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [40, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.227 + std = 1.282 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [40, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [160, 40, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [40, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [40, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [160, 40, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [40, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [80, 40, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [160, 80, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [320, 160, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_atto_ols.a2_in1k/graph_hash.txt b/samples/timm/convnext_atto_ols.a2_in1k/graph_hash.txt new file mode 100644 index 0000000000..1e9acf5e79 --- /dev/null +++ b/samples/timm/convnext_atto_ols.a2_in1k/graph_hash.txt @@ -0,0 +1 @@ +f61a87fb1a9abdffa812f56e0ebb94352465f2b1de8f3f6f6124fff252dc8d1a \ No newline at end of file diff --git a/samples/timm/convnext_atto_ols.a2_in1k/graph_net.json b/samples/timm/convnext_atto_ols.a2_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnext_atto_ols.a2_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_atto_ols.a2_in1k/input_meta.py b/samples/timm/convnext_atto_ols.a2_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_atto_ols.a2_in1k/input_tensor_constraints.py b/samples/timm/convnext_atto_ols.a2_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_atto_ols.a2_in1k/model.py b/samples/timm/convnext_atto_ols.a2_in1k/model.py new file mode 100644 index 0000000000..e117a5c35b --- /dev/null +++ b/samples/timm/convnext_atto_ols.a2_in1k/model.py @@ -0,0 +1,1143 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stem_modules_2_parameters_weight_ = ( + L_self_modules_stem_modules_2_parameters_weight_ + ) + l_self_modules_stem_modules_2_parameters_bias_ = ( + L_self_modules_stem_modules_2_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + input_2 = torch.conv2d( + input_1, + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_1 = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x = input_2.permute(0, 2, 3, 1) + input_2 = None + x_1 = torch.nn.functional.layer_norm( + x, + (40,), + l_self_modules_stem_modules_2_parameters_weight_, + l_self_modules_stem_modules_2_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_2_parameters_weight_ + ) = l_self_modules_stem_modules_2_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 40, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (40,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 40, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (40,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_21 = torch.conv2d( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (40,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_3 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_3, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (80,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + reshape_2 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_37 = x_36.mul(reshape_2) + x_36 = reshape_2 = None + x_38 = x_37 + input_3 + x_37 = input_3 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (80,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = torch.conv2d( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (80,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_4 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_4, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_60 = torch.conv2d( + x_59, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_59 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + reshape_4 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_62 = x_61.mul(reshape_4) + x_61 = reshape_4 = None + x_63 = x_62 + input_4 + x_62 = input_4 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_71 = torch.conv2d( + x_70, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_70 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + reshape_5 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_73 = x_72.mul(reshape_5) + x_72 = reshape_5 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_84 = x_83.mul(reshape_6) + x_83 = reshape_6 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_115 = torch.conv2d( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = x_118.permute(0, 2, 3, 1) + x_118 = None + x_120 = torch.nn.functional.layer_norm( + x_119, + (160,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_119 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_121 = x_120.permute(0, 3, 1, 2) + x_120 = None + input_5 = torch.conv2d( + x_121, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_122 = torch.conv2d( + input_5, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_123 = x_122.permute(0, 2, 3, 1) + x_122 = None + x_124 = torch.nn.functional.layer_norm( + x_123, + (320,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_125 = x_124.permute(0, 3, 1, 2) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126) + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_129 = torch.conv2d( + x_128, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_128 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) + x_129 = None + reshape_10 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_131 = x_130.mul(reshape_10) + x_130 = reshape_10 = None + x_132 = x_131 + input_5 + x_131 = input_5 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_134 = x_133.permute(0, 2, 3, 1) + x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (320,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_134 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_136 = x_135.permute(0, 3, 1, 2) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137) + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = torch.conv2d( + x_139, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_139 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_141 = torch.nn.functional.dropout(x_140, 0.0, False, False) + x_140 = None + reshape_11 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_142 = x_141.mul(reshape_11) + x_141 = reshape_11 = None + x_143 = x_142 + x_132 + x_142 = x_132 = None + x_144 = torch.nn.functional.adaptive_avg_pool2d(x_143, 1) + x_143 = None + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (320,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = x_147.flatten(1, -1) + x_147 = None + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = torch._C._nn.linear( + x_149, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_149 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_150,) diff --git a/samples/timm/convnext_atto_ols.a2_in1k/weight_meta.py b/samples/timm/convnext_atto_ols.a2_in1k/weight_meta.py new file mode 100644 index 0000000000..900544cfe2 --- /dev/null +++ b/samples/timm/convnext_atto_ols.a2_in1k/weight_meta.py @@ -0,0 +1,1330 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [24, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.232 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [40, 24, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_parameters_weight_: + name = "L_self_modules_stem_modules_2_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_parameters_bias_: + name = "L_self_modules_stem_modules_2_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [40, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [160, 40, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [40, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [40, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [160, 40, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [40, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [80, 40, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [160, 80, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [320, 160, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_base.clip_laion2b/graph_hash.txt b/samples/timm/convnext_base.clip_laion2b/graph_hash.txt new file mode 100644 index 0000000000..f5b53a89e0 --- /dev/null +++ b/samples/timm/convnext_base.clip_laion2b/graph_hash.txt @@ -0,0 +1 @@ +152e97e25e351ecaec13d28602c6382db8c6a1cab30e50ec944d857bd1ee7960 \ No newline at end of file diff --git a/samples/timm/convnext_base.clip_laion2b/graph_net.json b/samples/timm/convnext_base.clip_laion2b/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnext_base.clip_laion2b/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_base.clip_laion2b/input_meta.py b/samples/timm/convnext_base.clip_laion2b/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_base.clip_laion2b/input_tensor_constraints.py b/samples/timm/convnext_base.clip_laion2b/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_base.clip_laion2b/model.py b/samples/timm/convnext_base.clip_laion2b/model.py new file mode 100644 index 0000000000..9d5b5c1072 --- /dev/null +++ b/samples/timm/convnext_base.clip_laion2b/model.py @@ -0,0 +1,2707 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (128,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_9 = torch._C._nn.linear( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = x_10.permute(0, 3, 1, 2) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = torch._C._nn.linear( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_19 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = x_21.permute(0, 3, 1, 2) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_31 = torch._C._nn.linear( + x_30, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_30 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = x_32.permute(0, 3, 1, 2) + x_32 = None + reshape_2 = l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_34 = x_33.mul(reshape_2) + x_33 = reshape_2 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (128,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_45 = torch._C._nn.linear( + x_44, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_44 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_46 = torch.nn.functional.dropout(x_45, 0.0, False, False) + x_45 = None + x_47 = x_46.permute(0, 3, 1, 2) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = torch._C._nn.linear( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_55 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_57 = torch.nn.functional.dropout(x_56, 0.0, False, False) + x_56 = None + x_58 = x_57.permute(0, 3, 1, 2) + x_57 = None + reshape_4 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_59 = x_58.mul(reshape_4) + x_58 = reshape_4 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + reshape_5 = l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_70 = x_69.mul(reshape_5) + x_69 = reshape_5 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (256,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = torch._C._nn.linear( + x_80, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_80 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_82 = torch.nn.functional.dropout(x_81, 0.0, False, False) + x_81 = None + x_83 = x_82.permute(0, 3, 1, 2) + x_82 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_84 = x_83.mul(reshape_6) + x_83 = reshape_6 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = torch._C._nn.linear( + x_91, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_91 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_93 = torch.nn.functional.dropout(x_92, 0.0, False, False) + x_92 = None + x_94 = x_93.permute(0, 3, 1, 2) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_104 = torch.nn.functional.dropout(x_103, 0.0, False, False) + x_103 = None + x_105 = x_104.permute(0, 3, 1, 2) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = torch._C._nn.linear( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_113 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_115 = torch.nn.functional.dropout(x_114, 0.0, False, False) + x_114 = None + x_116 = x_115.permute(0, 3, 1, 2) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_124 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_126.permute(0, 3, 1, 2) + x_126 = None + reshape_10 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_128 = x_127.mul(reshape_10) + x_127 = reshape_10 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = torch._C._nn.linear( + x_135, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_135 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_137 = torch.nn.functional.dropout(x_136, 0.0, False, False) + x_136 = None + x_138 = x_137.permute(0, 3, 1, 2) + x_137 = None + reshape_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_139 = x_138.mul(reshape_11) + x_138 = reshape_11 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = torch._C._nn.linear( + x_146, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_146 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_148 = torch.nn.functional.dropout(x_147, 0.0, False, False) + x_147 = None + x_149 = x_148.permute(0, 3, 1, 2) + x_148 = None + reshape_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + None + ) + x_150 = x_149.mul(reshape_12) + x_149 = reshape_12 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_159.permute(0, 3, 1, 2) + x_159 = None + reshape_13 = l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + None + ) + x_161 = x_160.mul(reshape_13) + x_160 = reshape_13 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_169 = torch._C._nn.linear( + x_168, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_168 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_170 = torch.nn.functional.dropout(x_169, 0.0, False, False) + x_169 = None + x_171 = x_170.permute(0, 3, 1, 2) + x_170 = None + reshape_14 = l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + None + ) + x_172 = x_171.mul(reshape_14) + x_171 = reshape_14 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = (None) + x_175 = x_174.permute(0, 2, 3, 1) + x_174 = None + x_176 = torch.nn.functional.layer_norm( + x_175, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_, + 1e-06, + ) + x_175 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = (None) + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_176 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_178 = torch._C._nn.gelu(x_177) + x_177 = None + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_180 = torch._C._nn.linear( + x_179, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_179 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_181 = torch.nn.functional.dropout(x_180, 0.0, False, False) + x_180 = None + x_182 = x_181.permute(0, 3, 1, 2) + x_181 = None + reshape_15 = l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ = ( + None + ) + x_183 = x_182.mul(reshape_15) + x_182 = reshape_15 = None + x_184 = x_183 + x_173 + x_183 = x_173 = None + x_185 = torch.conv2d( + x_184, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = (None) + x_186 = x_185.permute(0, 2, 3, 1) + x_185 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = (None) + x_188 = torch._C._nn.linear( + x_187, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_187 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_189 = torch._C._nn.gelu(x_188) + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = torch._C._nn.linear( + x_190, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_190 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_192 = torch.nn.functional.dropout(x_191, 0.0, False, False) + x_191 = None + x_193 = x_192.permute(0, 3, 1, 2) + x_192 = None + reshape_16 = l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ = ( + None + ) + x_194 = x_193.mul(reshape_16) + x_193 = reshape_16 = None + x_195 = x_194 + x_184 + x_194 = x_184 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = (None) + x_197 = x_196.permute(0, 2, 3, 1) + x_196 = None + x_198 = torch.nn.functional.layer_norm( + x_197, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_, + 1e-06, + ) + x_197 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = (None) + x_199 = torch._C._nn.linear( + x_198, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_198 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_200 = torch._C._nn.gelu(x_199) + x_199 = None + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_202 = torch._C._nn.linear( + x_201, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_201 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_203 = torch.nn.functional.dropout(x_202, 0.0, False, False) + x_202 = None + x_204 = x_203.permute(0, 3, 1, 2) + x_203 = None + reshape_17 = l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ = ( + None + ) + x_205 = x_204.mul(reshape_17) + x_204 = reshape_17 = None + x_206 = x_205 + x_195 + x_205 = x_195 = None + x_207 = torch.conv2d( + x_206, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = (None) + x_208 = x_207.permute(0, 2, 3, 1) + x_207 = None + x_209 = torch.nn.functional.layer_norm( + x_208, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_, + 1e-06, + ) + x_208 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = (None) + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_209 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_211 = torch._C._nn.gelu(x_210) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_213 = torch._C._nn.linear( + x_212, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_212 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_214 = torch.nn.functional.dropout(x_213, 0.0, False, False) + x_213 = None + x_215 = x_214.permute(0, 3, 1, 2) + x_214 = None + reshape_18 = l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ = ( + None + ) + x_216 = x_215.mul(reshape_18) + x_215 = reshape_18 = None + x_217 = x_216 + x_206 + x_216 = x_206 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = (None) + x_219 = x_218.permute(0, 2, 3, 1) + x_218 = None + x_220 = torch.nn.functional.layer_norm( + x_219, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_, + 1e-06, + ) + x_219 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = (None) + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_220 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_222 = torch._C._nn.gelu(x_221) + x_221 = None + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_224 = torch._C._nn.linear( + x_223, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_223 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = x_225.permute(0, 3, 1, 2) + x_225 = None + reshape_19 = l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ = ( + None + ) + x_227 = x_226.mul(reshape_19) + x_226 = reshape_19 = None + x_228 = x_227 + x_217 + x_227 = x_217 = None + x_229 = torch.conv2d( + x_228, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = (None) + x_230 = x_229.permute(0, 2, 3, 1) + x_229 = None + x_231 = torch.nn.functional.layer_norm( + x_230, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_, + 1e-06, + ) + x_230 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = (None) + x_232 = torch._C._nn.linear( + x_231, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_231 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_233 = torch._C._nn.gelu(x_232) + x_232 = None + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_235 = torch._C._nn.linear( + x_234, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_234 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_236 = torch.nn.functional.dropout(x_235, 0.0, False, False) + x_235 = None + x_237 = x_236.permute(0, 3, 1, 2) + x_236 = None + reshape_20 = l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ = ( + None + ) + x_238 = x_237.mul(reshape_20) + x_237 = reshape_20 = None + x_239 = x_238 + x_228 + x_238 = x_228 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = (None) + x_241 = x_240.permute(0, 2, 3, 1) + x_240 = None + x_242 = torch.nn.functional.layer_norm( + x_241, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_, + 1e-06, + ) + x_241 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = (None) + x_243 = torch._C._nn.linear( + x_242, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_242 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_244 = torch._C._nn.gelu(x_243) + x_243 = None + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_246 = torch._C._nn.linear( + x_245, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_245 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_247 = torch.nn.functional.dropout(x_246, 0.0, False, False) + x_246 = None + x_248 = x_247.permute(0, 3, 1, 2) + x_247 = None + reshape_21 = l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ = ( + None + ) + x_249 = x_248.mul(reshape_21) + x_248 = reshape_21 = None + x_250 = x_249 + x_239 + x_249 = x_239 = None + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = (None) + x_252 = x_251.permute(0, 2, 3, 1) + x_251 = None + x_253 = torch.nn.functional.layer_norm( + x_252, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_, + 1e-06, + ) + x_252 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = (None) + x_254 = torch._C._nn.linear( + x_253, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_253 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_255 = torch._C._nn.gelu(x_254) + x_254 = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_257 = torch._C._nn.linear( + x_256, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_256 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_258 = torch.nn.functional.dropout(x_257, 0.0, False, False) + x_257 = None + x_259 = x_258.permute(0, 3, 1, 2) + x_258 = None + reshape_22 = l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ = ( + None + ) + x_260 = x_259.mul(reshape_22) + x_259 = reshape_22 = None + x_261 = x_260 + x_250 + x_260 = x_250 = None + x_262 = torch.conv2d( + x_261, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = (None) + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = (None) + x_265 = torch._C._nn.linear( + x_264, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_264 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_266 = torch._C._nn.gelu(x_265) + x_265 = None + x_267 = torch.nn.functional.dropout(x_266, 0.0, False, False) + x_266 = None + x_268 = torch._C._nn.linear( + x_267, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_269 = torch.nn.functional.dropout(x_268, 0.0, False, False) + x_268 = None + x_270 = x_269.permute(0, 3, 1, 2) + x_269 = None + reshape_23 = l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ = ( + None + ) + x_271 = x_270.mul(reshape_23) + x_270 = reshape_23 = None + x_272 = x_271 + x_261 + x_271 = x_261 = None + x_273 = torch.conv2d( + x_272, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = (None) + x_274 = x_273.permute(0, 2, 3, 1) + x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_, + 1e-06, + ) + x_274 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = (None) + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_277 = torch._C._nn.gelu(x_276) + x_276 = None + x_278 = torch.nn.functional.dropout(x_277, 0.0, False, False) + x_277 = None + x_279 = torch._C._nn.linear( + x_278, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_280 = torch.nn.functional.dropout(x_279, 0.0, False, False) + x_279 = None + x_281 = x_280.permute(0, 3, 1, 2) + x_280 = None + reshape_24 = l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ = ( + None + ) + x_282 = x_281.mul(reshape_24) + x_281 = reshape_24 = None + x_283 = x_282 + x_272 + x_282 = x_272 = None + x_284 = torch.conv2d( + x_283, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = (None) + x_285 = x_284.permute(0, 2, 3, 1) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_, + 1e-06, + ) + x_285 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = (None) + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_288 = torch._C._nn.gelu(x_287) + x_287 = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + x_290 = torch._C._nn.linear( + x_289, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_291 = torch.nn.functional.dropout(x_290, 0.0, False, False) + x_290 = None + x_292 = x_291.permute(0, 3, 1, 2) + x_291 = None + reshape_25 = l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ = ( + None + ) + x_293 = x_292.mul(reshape_25) + x_292 = reshape_25 = None + x_294 = x_293 + x_283 + x_293 = x_283 = None + x_295 = torch.conv2d( + x_294, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = (None) + x_296 = x_295.permute(0, 2, 3, 1) + x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_, + 1e-06, + ) + x_296 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = (None) + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_299 = torch._C._nn.gelu(x_298) + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_301 = torch._C._nn.linear( + x_300, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_302 = torch.nn.functional.dropout(x_301, 0.0, False, False) + x_301 = None + x_303 = x_302.permute(0, 3, 1, 2) + x_302 = None + reshape_26 = l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ = ( + None + ) + x_304 = x_303.mul(reshape_26) + x_303 = reshape_26 = None + x_305 = x_304 + x_294 + x_304 = x_294 = None + x_306 = torch.conv2d( + x_305, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = (None) + x_307 = x_306.permute(0, 2, 3, 1) + x_306 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_, + 1e-06, + ) + x_307 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = (None) + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_310 = torch._C._nn.gelu(x_309) + x_309 = None + x_311 = torch.nn.functional.dropout(x_310, 0.0, False, False) + x_310 = None + x_312 = torch._C._nn.linear( + x_311, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_313 = torch.nn.functional.dropout(x_312, 0.0, False, False) + x_312 = None + x_314 = x_313.permute(0, 3, 1, 2) + x_313 = None + reshape_27 = l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ = ( + None + ) + x_315 = x_314.mul(reshape_27) + x_314 = reshape_27 = None + x_316 = x_315 + x_305 + x_315 = x_305 = None + x_317 = torch.conv2d( + x_316, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = (None) + x_318 = x_317.permute(0, 2, 3, 1) + x_317 = None + x_319 = torch.nn.functional.layer_norm( + x_318, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_, + 1e-06, + ) + x_318 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = (None) + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_321 = torch._C._nn.gelu(x_320) + x_320 = None + x_322 = torch.nn.functional.dropout(x_321, 0.0, False, False) + x_321 = None + x_323 = torch._C._nn.linear( + x_322, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_324 = torch.nn.functional.dropout(x_323, 0.0, False, False) + x_323 = None + x_325 = x_324.permute(0, 3, 1, 2) + x_324 = None + reshape_28 = l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ = ( + None + ) + x_326 = x_325.mul(reshape_28) + x_325 = reshape_28 = None + x_327 = x_326 + x_316 + x_326 = x_316 = None + x_328 = torch.conv2d( + x_327, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = (None) + x_329 = x_328.permute(0, 2, 3, 1) + x_328 = None + x_330 = torch.nn.functional.layer_norm( + x_329, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_, + 1e-06, + ) + x_329 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = (None) + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_332 = torch._C._nn.gelu(x_331) + x_331 = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_334 = torch._C._nn.linear( + x_333, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_335 = torch.nn.functional.dropout(x_334, 0.0, False, False) + x_334 = None + x_336 = x_335.permute(0, 3, 1, 2) + x_335 = None + reshape_29 = l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ = ( + None + ) + x_337 = x_336.mul(reshape_29) + x_336 = reshape_29 = None + x_338 = x_337 + x_327 + x_337 = x_327 = None + x_339 = torch.conv2d( + x_338, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = (None) + x_340 = x_339.permute(0, 2, 3, 1) + x_339 = None + x_341 = torch.nn.functional.layer_norm( + x_340, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_, + 1e-06, + ) + x_340 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = (None) + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_343 = torch._C._nn.gelu(x_342) + x_342 = None + x_344 = torch.nn.functional.dropout(x_343, 0.0, False, False) + x_343 = None + x_345 = torch._C._nn.linear( + x_344, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_346 = torch.nn.functional.dropout(x_345, 0.0, False, False) + x_345 = None + x_347 = x_346.permute(0, 3, 1, 2) + x_346 = None + reshape_30 = l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ = ( + None + ) + x_348 = x_347.mul(reshape_30) + x_347 = reshape_30 = None + x_349 = x_348 + x_338 + x_348 = x_338 = None + x_350 = torch.conv2d( + x_349, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = (None) + x_351 = x_350.permute(0, 2, 3, 1) + x_350 = None + x_352 = torch.nn.functional.layer_norm( + x_351, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_, + 1e-06, + ) + x_351 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = (None) + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_354 = torch._C._nn.gelu(x_353) + x_353 = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_356 = torch._C._nn.linear( + x_355, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_357 = torch.nn.functional.dropout(x_356, 0.0, False, False) + x_356 = None + x_358 = x_357.permute(0, 3, 1, 2) + x_357 = None + reshape_31 = l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ = ( + None + ) + x_359 = x_358.mul(reshape_31) + x_358 = reshape_31 = None + x_360 = x_359 + x_349 + x_359 = x_349 = None + x_361 = torch.conv2d( + x_360, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = (None) + x_362 = x_361.permute(0, 2, 3, 1) + x_361 = None + x_363 = torch.nn.functional.layer_norm( + x_362, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_, + 1e-06, + ) + x_362 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = (None) + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_365 = torch._C._nn.gelu(x_364) + x_364 = None + x_366 = torch.nn.functional.dropout(x_365, 0.0, False, False) + x_365 = None + x_367 = torch._C._nn.linear( + x_366, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_368 = torch.nn.functional.dropout(x_367, 0.0, False, False) + x_367 = None + x_369 = x_368.permute(0, 3, 1, 2) + x_368 = None + reshape_32 = l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ = ( + None + ) + x_370 = x_369.mul(reshape_32) + x_369 = reshape_32 = None + x_371 = x_370 + x_360 + x_370 = x_360 = None + x_372 = x_371.permute(0, 2, 3, 1) + x_371 = None + x_373 = torch.nn.functional.layer_norm( + x_372, + (512,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_372 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_374 = x_373.permute(0, 3, 1, 2) + x_373 = None + input_4 = torch.conv2d( + x_374, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_374 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_375 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_376 = x_375.permute(0, 2, 3, 1) + x_375 = None + x_377 = torch.nn.functional.layer_norm( + x_376, + (1024,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_376 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_378 = torch._C._nn.linear( + x_377, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_377 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_379 = torch._C._nn.gelu(x_378) + x_378 = None + x_380 = torch.nn.functional.dropout(x_379, 0.0, False, False) + x_379 = None + x_381 = torch._C._nn.linear( + x_380, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_380 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_382 = torch.nn.functional.dropout(x_381, 0.0, False, False) + x_381 = None + x_383 = x_382.permute(0, 3, 1, 2) + x_382 = None + reshape_33 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_384 = x_383.mul(reshape_33) + x_383 = reshape_33 = None + x_385 = x_384 + input_4 + x_384 = input_4 = None + x_386 = torch.conv2d( + x_385, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_387 = x_386.permute(0, 2, 3, 1) + x_386 = None + x_388 = torch.nn.functional.layer_norm( + x_387, + (1024,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_387 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_389 = torch._C._nn.linear( + x_388, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_388 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_390 = torch._C._nn.gelu(x_389) + x_389 = None + x_391 = torch.nn.functional.dropout(x_390, 0.0, False, False) + x_390 = None + x_392 = torch._C._nn.linear( + x_391, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_391 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_393 = torch.nn.functional.dropout(x_392, 0.0, False, False) + x_392 = None + x_394 = x_393.permute(0, 3, 1, 2) + x_393 = None + reshape_34 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_395 = x_394.mul(reshape_34) + x_394 = reshape_34 = None + x_396 = x_395 + x_385 + x_395 = x_385 = None + x_397 = torch.conv2d( + x_396, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_398 = x_397.permute(0, 2, 3, 1) + x_397 = None + x_399 = torch.nn.functional.layer_norm( + x_398, + (1024,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_398 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_400 = torch._C._nn.linear( + x_399, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_399 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_401 = torch._C._nn.gelu(x_400) + x_400 = None + x_402 = torch.nn.functional.dropout(x_401, 0.0, False, False) + x_401 = None + x_403 = torch._C._nn.linear( + x_402, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_402 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_404 = torch.nn.functional.dropout(x_403, 0.0, False, False) + x_403 = None + x_405 = x_404.permute(0, 3, 1, 2) + x_404 = None + reshape_35 = l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_406 = x_405.mul(reshape_35) + x_405 = reshape_35 = None + x_407 = x_406 + x_396 + x_406 = x_396 = None + x_408 = torch.nn.functional.adaptive_avg_pool2d(x_407, 1) + x_407 = None + x_409 = x_408.permute(0, 2, 3, 1) + x_408 = None + x_410 = torch.nn.functional.layer_norm( + x_409, + (1024,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_409 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_411 = x_410.permute(0, 3, 1, 2) + x_410 = None + x_412 = x_411.flatten(1, -1) + x_411 = None + x_413 = torch.nn.functional.dropout(x_412, 0.0, False, False) + x_412 = None + x_414 = torch._C._nn.linear( + x_413, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_413 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_414,) diff --git a/samples/timm/convnext_base.clip_laion2b/weight_meta.py b/samples/timm/convnext_base.clip_laion2b/weight_meta.py new file mode 100644 index 0000000000..566dc3a68a --- /dev/null +++ b/samples/timm/convnext_base.clip_laion2b/weight_meta.py @@ -0,0 +1,3470 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [128, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.227 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [256, 128, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [512, 256, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [1024, 512, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [640, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None From d49328c6dc4ba3e3dd54ebb58231ca14fda19067 Mon Sep 17 00:00:00 2001 From: Liyulingyue <852433440@qq.com> Date: Mon, 4 Aug 2025 23:32:14 +0800 Subject: [PATCH 6/7] timm group --- .../convnext_femto.d1_in1k/graph_hash.txt | 1 + .../convnext_femto.d1_in1k/graph_net.json | 5 + .../timm/convnext_femto.d1_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/convnext_femto.d1_in1k/model.py | 1123 ++++++ .../convnext_femto.d1_in1k/weight_meta.py | 1310 +++++++ .../convnext_femto_ols.d1_in1k/graph_hash.txt | 1 + .../convnext_femto_ols.d1_in1k/graph_net.json | 5 + .../convnext_femto_ols.d1_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/convnext_femto_ols.d1_in1k/model.py | 1143 ++++++ .../convnext_femto_ols.d1_in1k/weight_meta.py | 1330 +++++++ .../convnext_large.fb_in1k/graph_hash.txt | 1 + .../convnext_large.fb_in1k/graph_net.json | 5 + .../timm/convnext_large.fb_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/convnext_large.fb_in1k/model.py | 2707 +++++++++++++ .../convnext_large.fb_in1k/weight_meta.py | 3470 ++++++++++++++++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../model.py | 2725 +++++++++++++ .../weight_meta.py | 3490 +++++++++++++++++ 24 files changed, 17322 insertions(+) create mode 100644 samples/timm/convnext_femto.d1_in1k/graph_hash.txt create mode 100644 samples/timm/convnext_femto.d1_in1k/graph_net.json create mode 100644 samples/timm/convnext_femto.d1_in1k/input_meta.py create mode 100644 samples/timm/convnext_femto.d1_in1k/input_tensor_constraints.py create mode 100644 samples/timm/convnext_femto.d1_in1k/model.py create mode 100644 samples/timm/convnext_femto.d1_in1k/weight_meta.py create mode 100644 samples/timm/convnext_femto_ols.d1_in1k/graph_hash.txt create mode 100644 samples/timm/convnext_femto_ols.d1_in1k/graph_net.json create mode 100644 samples/timm/convnext_femto_ols.d1_in1k/input_meta.py create mode 100644 samples/timm/convnext_femto_ols.d1_in1k/input_tensor_constraints.py create mode 100644 samples/timm/convnext_femto_ols.d1_in1k/model.py create mode 100644 samples/timm/convnext_femto_ols.d1_in1k/weight_meta.py create mode 100644 samples/timm/convnext_large.fb_in1k/graph_hash.txt create mode 100644 samples/timm/convnext_large.fb_in1k/graph_net.json create mode 100644 samples/timm/convnext_large.fb_in1k/input_meta.py create mode 100644 samples/timm/convnext_large.fb_in1k/input_tensor_constraints.py create mode 100644 samples/timm/convnext_large.fb_in1k/model.py create mode 100644 samples/timm/convnext_large.fb_in1k/weight_meta.py create mode 100644 samples/timm/convnext_large_mlp.clip_laion2b_augreg/graph_hash.txt create mode 100644 samples/timm/convnext_large_mlp.clip_laion2b_augreg/graph_net.json create mode 100644 samples/timm/convnext_large_mlp.clip_laion2b_augreg/input_meta.py create mode 100644 samples/timm/convnext_large_mlp.clip_laion2b_augreg/input_tensor_constraints.py create mode 100644 samples/timm/convnext_large_mlp.clip_laion2b_augreg/model.py create mode 100644 samples/timm/convnext_large_mlp.clip_laion2b_augreg/weight_meta.py diff --git a/samples/timm/convnext_femto.d1_in1k/graph_hash.txt b/samples/timm/convnext_femto.d1_in1k/graph_hash.txt new file mode 100644 index 0000000000..179cbdda8b --- /dev/null +++ b/samples/timm/convnext_femto.d1_in1k/graph_hash.txt @@ -0,0 +1 @@ +85c5a30ca86bd7b64b940fa8564803e5044d2c766d79b9366c542559fe06217e \ No newline at end of file diff --git a/samples/timm/convnext_femto.d1_in1k/graph_net.json b/samples/timm/convnext_femto.d1_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnext_femto.d1_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_femto.d1_in1k/input_meta.py b/samples/timm/convnext_femto.d1_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_femto.d1_in1k/input_tensor_constraints.py b/samples/timm/convnext_femto.d1_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_femto.d1_in1k/model.py b/samples/timm/convnext_femto.d1_in1k/model.py new file mode 100644 index 0000000000..e98cdb69ec --- /dev/null +++ b/samples/timm/convnext_femto.d1_in1k/model.py @@ -0,0 +1,1123 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (48,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 48, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (48,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 48, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (48,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_21 = torch.conv2d( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (48,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_2 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (96,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + reshape_2 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_37 = x_36.mul(reshape_2) + x_36 = reshape_2 = None + x_38 = x_37 + input_2 + x_37 = input_2 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (96,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = torch.conv2d( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (96,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_3 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_60 = torch.conv2d( + x_59, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_59 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + reshape_4 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_62 = x_61.mul(reshape_4) + x_61 = reshape_4 = None + x_63 = x_62 + input_3 + x_62 = input_3 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_71 = torch.conv2d( + x_70, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_70 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + reshape_5 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_73 = x_72.mul(reshape_5) + x_72 = reshape_5 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_84 = x_83.mul(reshape_6) + x_83 = reshape_6 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_115 = torch.conv2d( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = x_118.permute(0, 2, 3, 1) + x_118 = None + x_120 = torch.nn.functional.layer_norm( + x_119, + (192,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_119 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_121 = x_120.permute(0, 3, 1, 2) + x_120 = None + input_4 = torch.conv2d( + x_121, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_122 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_123 = x_122.permute(0, 2, 3, 1) + x_122 = None + x_124 = torch.nn.functional.layer_norm( + x_123, + (384,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_125 = x_124.permute(0, 3, 1, 2) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126) + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_129 = torch.conv2d( + x_128, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_128 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) + x_129 = None + reshape_10 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_131 = x_130.mul(reshape_10) + x_130 = reshape_10 = None + x_132 = x_131 + input_4 + x_131 = input_4 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_134 = x_133.permute(0, 2, 3, 1) + x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (384,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_134 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_136 = x_135.permute(0, 3, 1, 2) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137) + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = torch.conv2d( + x_139, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_139 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_141 = torch.nn.functional.dropout(x_140, 0.0, False, False) + x_140 = None + reshape_11 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_142 = x_141.mul(reshape_11) + x_141 = reshape_11 = None + x_143 = x_142 + x_132 + x_142 = x_132 = None + x_144 = torch.nn.functional.adaptive_avg_pool2d(x_143, 1) + x_143 = None + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (384,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = x_147.flatten(1, -1) + x_147 = None + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = torch._C._nn.linear( + x_149, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_149 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_150,) diff --git a/samples/timm/convnext_femto.d1_in1k/weight_meta.py b/samples/timm/convnext_femto.d1_in1k/weight_meta.py new file mode 100644 index 0000000000..820bd32534 --- /dev/null +++ b/samples/timm/convnext_femto.d1_in1k/weight_meta.py @@ -0,0 +1,1310 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [48, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.224 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [48, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [192, 48, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [48, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [48, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [192, 48, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [48, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [96, 48, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [192, 96, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [384, 192, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_femto_ols.d1_in1k/graph_hash.txt b/samples/timm/convnext_femto_ols.d1_in1k/graph_hash.txt new file mode 100644 index 0000000000..3643c3de05 --- /dev/null +++ b/samples/timm/convnext_femto_ols.d1_in1k/graph_hash.txt @@ -0,0 +1 @@ +4320de679ab41491b17bf22496332b6f32de1e1b6e78c5fa72353e59ad697d11 \ No newline at end of file diff --git a/samples/timm/convnext_femto_ols.d1_in1k/graph_net.json b/samples/timm/convnext_femto_ols.d1_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnext_femto_ols.d1_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_femto_ols.d1_in1k/input_meta.py b/samples/timm/convnext_femto_ols.d1_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_femto_ols.d1_in1k/input_tensor_constraints.py b/samples/timm/convnext_femto_ols.d1_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_femto_ols.d1_in1k/model.py b/samples/timm/convnext_femto_ols.d1_in1k/model.py new file mode 100644 index 0000000000..34ad8b43c7 --- /dev/null +++ b/samples/timm/convnext_femto_ols.d1_in1k/model.py @@ -0,0 +1,1143 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stem_modules_2_parameters_weight_ = ( + L_self_modules_stem_modules_2_parameters_weight_ + ) + l_self_modules_stem_modules_2_parameters_bias_ = ( + L_self_modules_stem_modules_2_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + input_2 = torch.conv2d( + input_1, + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_1 = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x = input_2.permute(0, 2, 3, 1) + input_2 = None + x_1 = torch.nn.functional.layer_norm( + x, + (48,), + l_self_modules_stem_modules_2_parameters_weight_, + l_self_modules_stem_modules_2_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_2_parameters_weight_ + ) = l_self_modules_stem_modules_2_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 48, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (48,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 48, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (48,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_21 = torch.conv2d( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (48,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_3 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_3, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (96,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + reshape_2 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_37 = x_36.mul(reshape_2) + x_36 = reshape_2 = None + x_38 = x_37 + input_3 + x_37 = input_3 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (96,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = torch.conv2d( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (96,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_4 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_4, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_60 = torch.conv2d( + x_59, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_59 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + reshape_4 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_62 = x_61.mul(reshape_4) + x_61 = reshape_4 = None + x_63 = x_62 + input_4 + x_62 = input_4 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_71 = torch.conv2d( + x_70, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_70 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + reshape_5 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_73 = x_72.mul(reshape_5) + x_72 = reshape_5 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_84 = x_83.mul(reshape_6) + x_83 = reshape_6 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_115 = torch.conv2d( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = x_118.permute(0, 2, 3, 1) + x_118 = None + x_120 = torch.nn.functional.layer_norm( + x_119, + (192,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_119 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_121 = x_120.permute(0, 3, 1, 2) + x_120 = None + input_5 = torch.conv2d( + x_121, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_122 = torch.conv2d( + input_5, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_123 = x_122.permute(0, 2, 3, 1) + x_122 = None + x_124 = torch.nn.functional.layer_norm( + x_123, + (384,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_125 = x_124.permute(0, 3, 1, 2) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126) + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_129 = torch.conv2d( + x_128, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_128 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) + x_129 = None + reshape_10 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_131 = x_130.mul(reshape_10) + x_130 = reshape_10 = None + x_132 = x_131 + input_5 + x_131 = input_5 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_134 = x_133.permute(0, 2, 3, 1) + x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (384,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_134 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_136 = x_135.permute(0, 3, 1, 2) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137) + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = torch.conv2d( + x_139, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_139 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_141 = torch.nn.functional.dropout(x_140, 0.0, False, False) + x_140 = None + reshape_11 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_142 = x_141.mul(reshape_11) + x_141 = reshape_11 = None + x_143 = x_142 + x_132 + x_142 = x_132 = None + x_144 = torch.nn.functional.adaptive_avg_pool2d(x_143, 1) + x_143 = None + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (384,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = x_147.flatten(1, -1) + x_147 = None + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = torch._C._nn.linear( + x_149, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_149 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_150,) diff --git a/samples/timm/convnext_femto_ols.d1_in1k/weight_meta.py b/samples/timm/convnext_femto_ols.d1_in1k/weight_meta.py new file mode 100644 index 0000000000..acb01961d2 --- /dev/null +++ b/samples/timm/convnext_femto_ols.d1_in1k/weight_meta.py @@ -0,0 +1,1330 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [24, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.226 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [48, 24, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_parameters_weight_: + name = "L_self_modules_stem_modules_2_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_parameters_bias_: + name = "L_self_modules_stem_modules_2_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [48, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [192, 48, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [48, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [48, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [192, 48, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [48, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [96, 48, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [192, 96, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [384, 192, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_large.fb_in1k/graph_hash.txt b/samples/timm/convnext_large.fb_in1k/graph_hash.txt new file mode 100644 index 0000000000..8df7fba92a --- /dev/null +++ b/samples/timm/convnext_large.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +65dd2eb2a0251f0b9e6d8e0aed379b52889a974b4b76df6f41342441d0a7b477 \ No newline at end of file diff --git a/samples/timm/convnext_large.fb_in1k/graph_net.json b/samples/timm/convnext_large.fb_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnext_large.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_large.fb_in1k/input_meta.py b/samples/timm/convnext_large.fb_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_large.fb_in1k/input_tensor_constraints.py b/samples/timm/convnext_large.fb_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_large.fb_in1k/model.py b/samples/timm/convnext_large.fb_in1k/model.py new file mode 100644 index 0000000000..9d2199cede --- /dev/null +++ b/samples/timm/convnext_large.fb_in1k/model.py @@ -0,0 +1,2707 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (192,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (192,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_9 = torch._C._nn.linear( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = x_10.permute(0, 3, 1, 2) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (192,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = torch._C._nn.linear( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_19 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = x_21.permute(0, 3, 1, 2) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (192,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_31 = torch._C._nn.linear( + x_30, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_30 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = x_32.permute(0, 3, 1, 2) + x_32 = None + reshape_2 = l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_34 = x_33.mul(reshape_2) + x_33 = reshape_2 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (192,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (384,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_45 = torch._C._nn.linear( + x_44, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_44 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_46 = torch.nn.functional.dropout(x_45, 0.0, False, False) + x_45 = None + x_47 = x_46.permute(0, 3, 1, 2) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (384,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = torch._C._nn.linear( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_55 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_57 = torch.nn.functional.dropout(x_56, 0.0, False, False) + x_56 = None + x_58 = x_57.permute(0, 3, 1, 2) + x_57 = None + reshape_4 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_59 = x_58.mul(reshape_4) + x_58 = reshape_4 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (384,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + reshape_5 = l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_70 = x_69.mul(reshape_5) + x_69 = reshape_5 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (384,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = torch._C._nn.linear( + x_80, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_80 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_82 = torch.nn.functional.dropout(x_81, 0.0, False, False) + x_81 = None + x_83 = x_82.permute(0, 3, 1, 2) + x_82 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_84 = x_83.mul(reshape_6) + x_83 = reshape_6 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = torch._C._nn.linear( + x_91, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_91 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_93 = torch.nn.functional.dropout(x_92, 0.0, False, False) + x_92 = None + x_94 = x_93.permute(0, 3, 1, 2) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_104 = torch.nn.functional.dropout(x_103, 0.0, False, False) + x_103 = None + x_105 = x_104.permute(0, 3, 1, 2) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = torch._C._nn.linear( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_113 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_115 = torch.nn.functional.dropout(x_114, 0.0, False, False) + x_114 = None + x_116 = x_115.permute(0, 3, 1, 2) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_124 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_126.permute(0, 3, 1, 2) + x_126 = None + reshape_10 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_128 = x_127.mul(reshape_10) + x_127 = reshape_10 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = torch._C._nn.linear( + x_135, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_135 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_137 = torch.nn.functional.dropout(x_136, 0.0, False, False) + x_136 = None + x_138 = x_137.permute(0, 3, 1, 2) + x_137 = None + reshape_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_139 = x_138.mul(reshape_11) + x_138 = reshape_11 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = torch._C._nn.linear( + x_146, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_146 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_148 = torch.nn.functional.dropout(x_147, 0.0, False, False) + x_147 = None + x_149 = x_148.permute(0, 3, 1, 2) + x_148 = None + reshape_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + None + ) + x_150 = x_149.mul(reshape_12) + x_149 = reshape_12 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_159.permute(0, 3, 1, 2) + x_159 = None + reshape_13 = l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + None + ) + x_161 = x_160.mul(reshape_13) + x_160 = reshape_13 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_169 = torch._C._nn.linear( + x_168, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_168 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_170 = torch.nn.functional.dropout(x_169, 0.0, False, False) + x_169 = None + x_171 = x_170.permute(0, 3, 1, 2) + x_170 = None + reshape_14 = l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + None + ) + x_172 = x_171.mul(reshape_14) + x_171 = reshape_14 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = (None) + x_175 = x_174.permute(0, 2, 3, 1) + x_174 = None + x_176 = torch.nn.functional.layer_norm( + x_175, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_, + 1e-06, + ) + x_175 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = (None) + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_176 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_178 = torch._C._nn.gelu(x_177) + x_177 = None + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_180 = torch._C._nn.linear( + x_179, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_179 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_181 = torch.nn.functional.dropout(x_180, 0.0, False, False) + x_180 = None + x_182 = x_181.permute(0, 3, 1, 2) + x_181 = None + reshape_15 = l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ = ( + None + ) + x_183 = x_182.mul(reshape_15) + x_182 = reshape_15 = None + x_184 = x_183 + x_173 + x_183 = x_173 = None + x_185 = torch.conv2d( + x_184, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = (None) + x_186 = x_185.permute(0, 2, 3, 1) + x_185 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = (None) + x_188 = torch._C._nn.linear( + x_187, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_187 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_189 = torch._C._nn.gelu(x_188) + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = torch._C._nn.linear( + x_190, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_190 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_192 = torch.nn.functional.dropout(x_191, 0.0, False, False) + x_191 = None + x_193 = x_192.permute(0, 3, 1, 2) + x_192 = None + reshape_16 = l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ = ( + None + ) + x_194 = x_193.mul(reshape_16) + x_193 = reshape_16 = None + x_195 = x_194 + x_184 + x_194 = x_184 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = (None) + x_197 = x_196.permute(0, 2, 3, 1) + x_196 = None + x_198 = torch.nn.functional.layer_norm( + x_197, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_, + 1e-06, + ) + x_197 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = (None) + x_199 = torch._C._nn.linear( + x_198, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_198 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_200 = torch._C._nn.gelu(x_199) + x_199 = None + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_202 = torch._C._nn.linear( + x_201, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_201 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_203 = torch.nn.functional.dropout(x_202, 0.0, False, False) + x_202 = None + x_204 = x_203.permute(0, 3, 1, 2) + x_203 = None + reshape_17 = l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ = ( + None + ) + x_205 = x_204.mul(reshape_17) + x_204 = reshape_17 = None + x_206 = x_205 + x_195 + x_205 = x_195 = None + x_207 = torch.conv2d( + x_206, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = (None) + x_208 = x_207.permute(0, 2, 3, 1) + x_207 = None + x_209 = torch.nn.functional.layer_norm( + x_208, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_, + 1e-06, + ) + x_208 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = (None) + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_209 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_211 = torch._C._nn.gelu(x_210) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_213 = torch._C._nn.linear( + x_212, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_212 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_214 = torch.nn.functional.dropout(x_213, 0.0, False, False) + x_213 = None + x_215 = x_214.permute(0, 3, 1, 2) + x_214 = None + reshape_18 = l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ = ( + None + ) + x_216 = x_215.mul(reshape_18) + x_215 = reshape_18 = None + x_217 = x_216 + x_206 + x_216 = x_206 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = (None) + x_219 = x_218.permute(0, 2, 3, 1) + x_218 = None + x_220 = torch.nn.functional.layer_norm( + x_219, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_, + 1e-06, + ) + x_219 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = (None) + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_220 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_222 = torch._C._nn.gelu(x_221) + x_221 = None + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_224 = torch._C._nn.linear( + x_223, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_223 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = x_225.permute(0, 3, 1, 2) + x_225 = None + reshape_19 = l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ = ( + None + ) + x_227 = x_226.mul(reshape_19) + x_226 = reshape_19 = None + x_228 = x_227 + x_217 + x_227 = x_217 = None + x_229 = torch.conv2d( + x_228, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = (None) + x_230 = x_229.permute(0, 2, 3, 1) + x_229 = None + x_231 = torch.nn.functional.layer_norm( + x_230, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_, + 1e-06, + ) + x_230 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = (None) + x_232 = torch._C._nn.linear( + x_231, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_231 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_233 = torch._C._nn.gelu(x_232) + x_232 = None + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_235 = torch._C._nn.linear( + x_234, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_234 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_236 = torch.nn.functional.dropout(x_235, 0.0, False, False) + x_235 = None + x_237 = x_236.permute(0, 3, 1, 2) + x_236 = None + reshape_20 = l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ = ( + None + ) + x_238 = x_237.mul(reshape_20) + x_237 = reshape_20 = None + x_239 = x_238 + x_228 + x_238 = x_228 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = (None) + x_241 = x_240.permute(0, 2, 3, 1) + x_240 = None + x_242 = torch.nn.functional.layer_norm( + x_241, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_, + 1e-06, + ) + x_241 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = (None) + x_243 = torch._C._nn.linear( + x_242, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_242 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_244 = torch._C._nn.gelu(x_243) + x_243 = None + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_246 = torch._C._nn.linear( + x_245, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_245 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_247 = torch.nn.functional.dropout(x_246, 0.0, False, False) + x_246 = None + x_248 = x_247.permute(0, 3, 1, 2) + x_247 = None + reshape_21 = l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ = ( + None + ) + x_249 = x_248.mul(reshape_21) + x_248 = reshape_21 = None + x_250 = x_249 + x_239 + x_249 = x_239 = None + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = (None) + x_252 = x_251.permute(0, 2, 3, 1) + x_251 = None + x_253 = torch.nn.functional.layer_norm( + x_252, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_, + 1e-06, + ) + x_252 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = (None) + x_254 = torch._C._nn.linear( + x_253, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_253 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_255 = torch._C._nn.gelu(x_254) + x_254 = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_257 = torch._C._nn.linear( + x_256, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_256 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_258 = torch.nn.functional.dropout(x_257, 0.0, False, False) + x_257 = None + x_259 = x_258.permute(0, 3, 1, 2) + x_258 = None + reshape_22 = l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ = ( + None + ) + x_260 = x_259.mul(reshape_22) + x_259 = reshape_22 = None + x_261 = x_260 + x_250 + x_260 = x_250 = None + x_262 = torch.conv2d( + x_261, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = (None) + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = (None) + x_265 = torch._C._nn.linear( + x_264, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_264 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_266 = torch._C._nn.gelu(x_265) + x_265 = None + x_267 = torch.nn.functional.dropout(x_266, 0.0, False, False) + x_266 = None + x_268 = torch._C._nn.linear( + x_267, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_269 = torch.nn.functional.dropout(x_268, 0.0, False, False) + x_268 = None + x_270 = x_269.permute(0, 3, 1, 2) + x_269 = None + reshape_23 = l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ = ( + None + ) + x_271 = x_270.mul(reshape_23) + x_270 = reshape_23 = None + x_272 = x_271 + x_261 + x_271 = x_261 = None + x_273 = torch.conv2d( + x_272, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = (None) + x_274 = x_273.permute(0, 2, 3, 1) + x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_, + 1e-06, + ) + x_274 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = (None) + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_277 = torch._C._nn.gelu(x_276) + x_276 = None + x_278 = torch.nn.functional.dropout(x_277, 0.0, False, False) + x_277 = None + x_279 = torch._C._nn.linear( + x_278, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_280 = torch.nn.functional.dropout(x_279, 0.0, False, False) + x_279 = None + x_281 = x_280.permute(0, 3, 1, 2) + x_280 = None + reshape_24 = l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ = ( + None + ) + x_282 = x_281.mul(reshape_24) + x_281 = reshape_24 = None + x_283 = x_282 + x_272 + x_282 = x_272 = None + x_284 = torch.conv2d( + x_283, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = (None) + x_285 = x_284.permute(0, 2, 3, 1) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_, + 1e-06, + ) + x_285 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = (None) + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_288 = torch._C._nn.gelu(x_287) + x_287 = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + x_290 = torch._C._nn.linear( + x_289, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_291 = torch.nn.functional.dropout(x_290, 0.0, False, False) + x_290 = None + x_292 = x_291.permute(0, 3, 1, 2) + x_291 = None + reshape_25 = l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ = ( + None + ) + x_293 = x_292.mul(reshape_25) + x_292 = reshape_25 = None + x_294 = x_293 + x_283 + x_293 = x_283 = None + x_295 = torch.conv2d( + x_294, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = (None) + x_296 = x_295.permute(0, 2, 3, 1) + x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_, + 1e-06, + ) + x_296 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = (None) + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_299 = torch._C._nn.gelu(x_298) + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_301 = torch._C._nn.linear( + x_300, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_302 = torch.nn.functional.dropout(x_301, 0.0, False, False) + x_301 = None + x_303 = x_302.permute(0, 3, 1, 2) + x_302 = None + reshape_26 = l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ = ( + None + ) + x_304 = x_303.mul(reshape_26) + x_303 = reshape_26 = None + x_305 = x_304 + x_294 + x_304 = x_294 = None + x_306 = torch.conv2d( + x_305, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = (None) + x_307 = x_306.permute(0, 2, 3, 1) + x_306 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_, + 1e-06, + ) + x_307 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = (None) + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_310 = torch._C._nn.gelu(x_309) + x_309 = None + x_311 = torch.nn.functional.dropout(x_310, 0.0, False, False) + x_310 = None + x_312 = torch._C._nn.linear( + x_311, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_313 = torch.nn.functional.dropout(x_312, 0.0, False, False) + x_312 = None + x_314 = x_313.permute(0, 3, 1, 2) + x_313 = None + reshape_27 = l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ = ( + None + ) + x_315 = x_314.mul(reshape_27) + x_314 = reshape_27 = None + x_316 = x_315 + x_305 + x_315 = x_305 = None + x_317 = torch.conv2d( + x_316, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = (None) + x_318 = x_317.permute(0, 2, 3, 1) + x_317 = None + x_319 = torch.nn.functional.layer_norm( + x_318, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_, + 1e-06, + ) + x_318 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = (None) + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_321 = torch._C._nn.gelu(x_320) + x_320 = None + x_322 = torch.nn.functional.dropout(x_321, 0.0, False, False) + x_321 = None + x_323 = torch._C._nn.linear( + x_322, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_324 = torch.nn.functional.dropout(x_323, 0.0, False, False) + x_323 = None + x_325 = x_324.permute(0, 3, 1, 2) + x_324 = None + reshape_28 = l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ = ( + None + ) + x_326 = x_325.mul(reshape_28) + x_325 = reshape_28 = None + x_327 = x_326 + x_316 + x_326 = x_316 = None + x_328 = torch.conv2d( + x_327, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = (None) + x_329 = x_328.permute(0, 2, 3, 1) + x_328 = None + x_330 = torch.nn.functional.layer_norm( + x_329, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_, + 1e-06, + ) + x_329 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = (None) + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_332 = torch._C._nn.gelu(x_331) + x_331 = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_334 = torch._C._nn.linear( + x_333, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_335 = torch.nn.functional.dropout(x_334, 0.0, False, False) + x_334 = None + x_336 = x_335.permute(0, 3, 1, 2) + x_335 = None + reshape_29 = l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ = ( + None + ) + x_337 = x_336.mul(reshape_29) + x_336 = reshape_29 = None + x_338 = x_337 + x_327 + x_337 = x_327 = None + x_339 = torch.conv2d( + x_338, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = (None) + x_340 = x_339.permute(0, 2, 3, 1) + x_339 = None + x_341 = torch.nn.functional.layer_norm( + x_340, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_, + 1e-06, + ) + x_340 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = (None) + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_343 = torch._C._nn.gelu(x_342) + x_342 = None + x_344 = torch.nn.functional.dropout(x_343, 0.0, False, False) + x_343 = None + x_345 = torch._C._nn.linear( + x_344, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_346 = torch.nn.functional.dropout(x_345, 0.0, False, False) + x_345 = None + x_347 = x_346.permute(0, 3, 1, 2) + x_346 = None + reshape_30 = l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ = ( + None + ) + x_348 = x_347.mul(reshape_30) + x_347 = reshape_30 = None + x_349 = x_348 + x_338 + x_348 = x_338 = None + x_350 = torch.conv2d( + x_349, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = (None) + x_351 = x_350.permute(0, 2, 3, 1) + x_350 = None + x_352 = torch.nn.functional.layer_norm( + x_351, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_, + 1e-06, + ) + x_351 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = (None) + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_354 = torch._C._nn.gelu(x_353) + x_353 = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_356 = torch._C._nn.linear( + x_355, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_357 = torch.nn.functional.dropout(x_356, 0.0, False, False) + x_356 = None + x_358 = x_357.permute(0, 3, 1, 2) + x_357 = None + reshape_31 = l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ = ( + None + ) + x_359 = x_358.mul(reshape_31) + x_358 = reshape_31 = None + x_360 = x_359 + x_349 + x_359 = x_349 = None + x_361 = torch.conv2d( + x_360, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = (None) + x_362 = x_361.permute(0, 2, 3, 1) + x_361 = None + x_363 = torch.nn.functional.layer_norm( + x_362, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_, + 1e-06, + ) + x_362 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = (None) + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_365 = torch._C._nn.gelu(x_364) + x_364 = None + x_366 = torch.nn.functional.dropout(x_365, 0.0, False, False) + x_365 = None + x_367 = torch._C._nn.linear( + x_366, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_368 = torch.nn.functional.dropout(x_367, 0.0, False, False) + x_367 = None + x_369 = x_368.permute(0, 3, 1, 2) + x_368 = None + reshape_32 = l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ = ( + None + ) + x_370 = x_369.mul(reshape_32) + x_369 = reshape_32 = None + x_371 = x_370 + x_360 + x_370 = x_360 = None + x_372 = x_371.permute(0, 2, 3, 1) + x_371 = None + x_373 = torch.nn.functional.layer_norm( + x_372, + (768,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_372 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_374 = x_373.permute(0, 3, 1, 2) + x_373 = None + input_4 = torch.conv2d( + x_374, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_374 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_375 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_376 = x_375.permute(0, 2, 3, 1) + x_375 = None + x_377 = torch.nn.functional.layer_norm( + x_376, + (1536,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_376 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_378 = torch._C._nn.linear( + x_377, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_377 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_379 = torch._C._nn.gelu(x_378) + x_378 = None + x_380 = torch.nn.functional.dropout(x_379, 0.0, False, False) + x_379 = None + x_381 = torch._C._nn.linear( + x_380, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_380 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_382 = torch.nn.functional.dropout(x_381, 0.0, False, False) + x_381 = None + x_383 = x_382.permute(0, 3, 1, 2) + x_382 = None + reshape_33 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_384 = x_383.mul(reshape_33) + x_383 = reshape_33 = None + x_385 = x_384 + input_4 + x_384 = input_4 = None + x_386 = torch.conv2d( + x_385, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_387 = x_386.permute(0, 2, 3, 1) + x_386 = None + x_388 = torch.nn.functional.layer_norm( + x_387, + (1536,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_387 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_389 = torch._C._nn.linear( + x_388, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_388 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_390 = torch._C._nn.gelu(x_389) + x_389 = None + x_391 = torch.nn.functional.dropout(x_390, 0.0, False, False) + x_390 = None + x_392 = torch._C._nn.linear( + x_391, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_391 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_393 = torch.nn.functional.dropout(x_392, 0.0, False, False) + x_392 = None + x_394 = x_393.permute(0, 3, 1, 2) + x_393 = None + reshape_34 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_395 = x_394.mul(reshape_34) + x_394 = reshape_34 = None + x_396 = x_395 + x_385 + x_395 = x_385 = None + x_397 = torch.conv2d( + x_396, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_398 = x_397.permute(0, 2, 3, 1) + x_397 = None + x_399 = torch.nn.functional.layer_norm( + x_398, + (1536,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_398 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_400 = torch._C._nn.linear( + x_399, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_399 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_401 = torch._C._nn.gelu(x_400) + x_400 = None + x_402 = torch.nn.functional.dropout(x_401, 0.0, False, False) + x_401 = None + x_403 = torch._C._nn.linear( + x_402, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_402 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_404 = torch.nn.functional.dropout(x_403, 0.0, False, False) + x_403 = None + x_405 = x_404.permute(0, 3, 1, 2) + x_404 = None + reshape_35 = l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_406 = x_405.mul(reshape_35) + x_405 = reshape_35 = None + x_407 = x_406 + x_396 + x_406 = x_396 = None + x_408 = torch.nn.functional.adaptive_avg_pool2d(x_407, 1) + x_407 = None + x_409 = x_408.permute(0, 2, 3, 1) + x_408 = None + x_410 = torch.nn.functional.layer_norm( + x_409, + (1536,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_409 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_411 = x_410.permute(0, 3, 1, 2) + x_410 = None + x_412 = x_411.flatten(1, -1) + x_411 = None + x_413 = torch.nn.functional.dropout(x_412, 0.0, False, False) + x_412 = None + x_414 = torch._C._nn.linear( + x_413, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_413 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_414,) diff --git a/samples/timm/convnext_large.fb_in1k/weight_meta.py b/samples/timm/convnext_large.fb_in1k/weight_meta.py new file mode 100644 index 0000000000..8563e8e7b3 --- /dev/null +++ b/samples/timm/convnext_large.fb_in1k/weight_meta.py @@ -0,0 +1,3470 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [192, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.223 + std = 1.281 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [384, 192, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [768, 384, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [1536, 768, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_large_mlp.clip_laion2b_augreg/graph_hash.txt b/samples/timm/convnext_large_mlp.clip_laion2b_augreg/graph_hash.txt new file mode 100644 index 0000000000..c34790de67 --- /dev/null +++ b/samples/timm/convnext_large_mlp.clip_laion2b_augreg/graph_hash.txt @@ -0,0 +1 @@ +acec500cb817544f7db171a60c01b406cb5865eb37e554a6c0ee9591630a2de7 \ No newline at end of file diff --git a/samples/timm/convnext_large_mlp.clip_laion2b_augreg/graph_net.json b/samples/timm/convnext_large_mlp.clip_laion2b_augreg/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnext_large_mlp.clip_laion2b_augreg/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_large_mlp.clip_laion2b_augreg/input_meta.py b/samples/timm/convnext_large_mlp.clip_laion2b_augreg/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_large_mlp.clip_laion2b_augreg/input_tensor_constraints.py b/samples/timm/convnext_large_mlp.clip_laion2b_augreg/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_large_mlp.clip_laion2b_augreg/model.py b/samples/timm/convnext_large_mlp.clip_laion2b_augreg/model.py new file mode 100644 index 0000000000..5502ab1517 --- /dev/null +++ b/samples/timm/convnext_large_mlp.clip_laion2b_augreg/model.py @@ -0,0 +1,2725 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_pre_logits_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_pre_logits_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_pre_logits_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_pre_logits_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_pre_logits_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_pre_logits_modules_fc_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (192,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (192,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_9 = torch._C._nn.linear( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = x_10.permute(0, 3, 1, 2) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (192,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = torch._C._nn.linear( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_19 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = x_21.permute(0, 3, 1, 2) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (192,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_31 = torch._C._nn.linear( + x_30, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_30 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = x_32.permute(0, 3, 1, 2) + x_32 = None + reshape_2 = l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_34 = x_33.mul(reshape_2) + x_33 = reshape_2 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (192,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (384,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_45 = torch._C._nn.linear( + x_44, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_44 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_46 = torch.nn.functional.dropout(x_45, 0.0, False, False) + x_45 = None + x_47 = x_46.permute(0, 3, 1, 2) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (384,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = torch._C._nn.linear( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_55 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_57 = torch.nn.functional.dropout(x_56, 0.0, False, False) + x_56 = None + x_58 = x_57.permute(0, 3, 1, 2) + x_57 = None + reshape_4 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_59 = x_58.mul(reshape_4) + x_58 = reshape_4 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (384,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + reshape_5 = l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_70 = x_69.mul(reshape_5) + x_69 = reshape_5 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (384,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = torch._C._nn.linear( + x_80, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_80 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_82 = torch.nn.functional.dropout(x_81, 0.0, False, False) + x_81 = None + x_83 = x_82.permute(0, 3, 1, 2) + x_82 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_84 = x_83.mul(reshape_6) + x_83 = reshape_6 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = torch._C._nn.linear( + x_91, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_91 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_93 = torch.nn.functional.dropout(x_92, 0.0, False, False) + x_92 = None + x_94 = x_93.permute(0, 3, 1, 2) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_104 = torch.nn.functional.dropout(x_103, 0.0, False, False) + x_103 = None + x_105 = x_104.permute(0, 3, 1, 2) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = torch._C._nn.linear( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_113 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_115 = torch.nn.functional.dropout(x_114, 0.0, False, False) + x_114 = None + x_116 = x_115.permute(0, 3, 1, 2) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_124 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_126.permute(0, 3, 1, 2) + x_126 = None + reshape_10 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_128 = x_127.mul(reshape_10) + x_127 = reshape_10 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = torch._C._nn.linear( + x_135, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_135 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_137 = torch.nn.functional.dropout(x_136, 0.0, False, False) + x_136 = None + x_138 = x_137.permute(0, 3, 1, 2) + x_137 = None + reshape_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_139 = x_138.mul(reshape_11) + x_138 = reshape_11 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = torch._C._nn.linear( + x_146, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_146 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_148 = torch.nn.functional.dropout(x_147, 0.0, False, False) + x_147 = None + x_149 = x_148.permute(0, 3, 1, 2) + x_148 = None + reshape_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + None + ) + x_150 = x_149.mul(reshape_12) + x_149 = reshape_12 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_159.permute(0, 3, 1, 2) + x_159 = None + reshape_13 = l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + None + ) + x_161 = x_160.mul(reshape_13) + x_160 = reshape_13 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_169 = torch._C._nn.linear( + x_168, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_168 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_170 = torch.nn.functional.dropout(x_169, 0.0, False, False) + x_169 = None + x_171 = x_170.permute(0, 3, 1, 2) + x_170 = None + reshape_14 = l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + None + ) + x_172 = x_171.mul(reshape_14) + x_171 = reshape_14 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = (None) + x_175 = x_174.permute(0, 2, 3, 1) + x_174 = None + x_176 = torch.nn.functional.layer_norm( + x_175, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_, + 1e-06, + ) + x_175 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = (None) + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_176 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_178 = torch._C._nn.gelu(x_177) + x_177 = None + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_180 = torch._C._nn.linear( + x_179, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_179 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_181 = torch.nn.functional.dropout(x_180, 0.0, False, False) + x_180 = None + x_182 = x_181.permute(0, 3, 1, 2) + x_181 = None + reshape_15 = l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ = ( + None + ) + x_183 = x_182.mul(reshape_15) + x_182 = reshape_15 = None + x_184 = x_183 + x_173 + x_183 = x_173 = None + x_185 = torch.conv2d( + x_184, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = (None) + x_186 = x_185.permute(0, 2, 3, 1) + x_185 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = (None) + x_188 = torch._C._nn.linear( + x_187, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_187 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_189 = torch._C._nn.gelu(x_188) + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = torch._C._nn.linear( + x_190, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_190 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_192 = torch.nn.functional.dropout(x_191, 0.0, False, False) + x_191 = None + x_193 = x_192.permute(0, 3, 1, 2) + x_192 = None + reshape_16 = l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ = ( + None + ) + x_194 = x_193.mul(reshape_16) + x_193 = reshape_16 = None + x_195 = x_194 + x_184 + x_194 = x_184 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = (None) + x_197 = x_196.permute(0, 2, 3, 1) + x_196 = None + x_198 = torch.nn.functional.layer_norm( + x_197, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_, + 1e-06, + ) + x_197 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = (None) + x_199 = torch._C._nn.linear( + x_198, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_198 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_200 = torch._C._nn.gelu(x_199) + x_199 = None + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_202 = torch._C._nn.linear( + x_201, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_201 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_203 = torch.nn.functional.dropout(x_202, 0.0, False, False) + x_202 = None + x_204 = x_203.permute(0, 3, 1, 2) + x_203 = None + reshape_17 = l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ = ( + None + ) + x_205 = x_204.mul(reshape_17) + x_204 = reshape_17 = None + x_206 = x_205 + x_195 + x_205 = x_195 = None + x_207 = torch.conv2d( + x_206, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = (None) + x_208 = x_207.permute(0, 2, 3, 1) + x_207 = None + x_209 = torch.nn.functional.layer_norm( + x_208, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_, + 1e-06, + ) + x_208 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = (None) + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_209 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_211 = torch._C._nn.gelu(x_210) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_213 = torch._C._nn.linear( + x_212, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_212 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_214 = torch.nn.functional.dropout(x_213, 0.0, False, False) + x_213 = None + x_215 = x_214.permute(0, 3, 1, 2) + x_214 = None + reshape_18 = l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ = ( + None + ) + x_216 = x_215.mul(reshape_18) + x_215 = reshape_18 = None + x_217 = x_216 + x_206 + x_216 = x_206 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = (None) + x_219 = x_218.permute(0, 2, 3, 1) + x_218 = None + x_220 = torch.nn.functional.layer_norm( + x_219, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_, + 1e-06, + ) + x_219 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = (None) + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_220 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_222 = torch._C._nn.gelu(x_221) + x_221 = None + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_224 = torch._C._nn.linear( + x_223, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_223 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = x_225.permute(0, 3, 1, 2) + x_225 = None + reshape_19 = l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ = ( + None + ) + x_227 = x_226.mul(reshape_19) + x_226 = reshape_19 = None + x_228 = x_227 + x_217 + x_227 = x_217 = None + x_229 = torch.conv2d( + x_228, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = (None) + x_230 = x_229.permute(0, 2, 3, 1) + x_229 = None + x_231 = torch.nn.functional.layer_norm( + x_230, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_, + 1e-06, + ) + x_230 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = (None) + x_232 = torch._C._nn.linear( + x_231, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_231 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_233 = torch._C._nn.gelu(x_232) + x_232 = None + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_235 = torch._C._nn.linear( + x_234, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_234 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_236 = torch.nn.functional.dropout(x_235, 0.0, False, False) + x_235 = None + x_237 = x_236.permute(0, 3, 1, 2) + x_236 = None + reshape_20 = l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ = ( + None + ) + x_238 = x_237.mul(reshape_20) + x_237 = reshape_20 = None + x_239 = x_238 + x_228 + x_238 = x_228 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = (None) + x_241 = x_240.permute(0, 2, 3, 1) + x_240 = None + x_242 = torch.nn.functional.layer_norm( + x_241, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_, + 1e-06, + ) + x_241 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = (None) + x_243 = torch._C._nn.linear( + x_242, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_242 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_244 = torch._C._nn.gelu(x_243) + x_243 = None + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_246 = torch._C._nn.linear( + x_245, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_245 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_247 = torch.nn.functional.dropout(x_246, 0.0, False, False) + x_246 = None + x_248 = x_247.permute(0, 3, 1, 2) + x_247 = None + reshape_21 = l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ = ( + None + ) + x_249 = x_248.mul(reshape_21) + x_248 = reshape_21 = None + x_250 = x_249 + x_239 + x_249 = x_239 = None + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = (None) + x_252 = x_251.permute(0, 2, 3, 1) + x_251 = None + x_253 = torch.nn.functional.layer_norm( + x_252, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_, + 1e-06, + ) + x_252 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = (None) + x_254 = torch._C._nn.linear( + x_253, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_253 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_255 = torch._C._nn.gelu(x_254) + x_254 = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_257 = torch._C._nn.linear( + x_256, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_256 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_258 = torch.nn.functional.dropout(x_257, 0.0, False, False) + x_257 = None + x_259 = x_258.permute(0, 3, 1, 2) + x_258 = None + reshape_22 = l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ = ( + None + ) + x_260 = x_259.mul(reshape_22) + x_259 = reshape_22 = None + x_261 = x_260 + x_250 + x_260 = x_250 = None + x_262 = torch.conv2d( + x_261, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = (None) + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = (None) + x_265 = torch._C._nn.linear( + x_264, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_264 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_266 = torch._C._nn.gelu(x_265) + x_265 = None + x_267 = torch.nn.functional.dropout(x_266, 0.0, False, False) + x_266 = None + x_268 = torch._C._nn.linear( + x_267, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_269 = torch.nn.functional.dropout(x_268, 0.0, False, False) + x_268 = None + x_270 = x_269.permute(0, 3, 1, 2) + x_269 = None + reshape_23 = l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ = ( + None + ) + x_271 = x_270.mul(reshape_23) + x_270 = reshape_23 = None + x_272 = x_271 + x_261 + x_271 = x_261 = None + x_273 = torch.conv2d( + x_272, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = (None) + x_274 = x_273.permute(0, 2, 3, 1) + x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_, + 1e-06, + ) + x_274 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = (None) + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_277 = torch._C._nn.gelu(x_276) + x_276 = None + x_278 = torch.nn.functional.dropout(x_277, 0.0, False, False) + x_277 = None + x_279 = torch._C._nn.linear( + x_278, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_280 = torch.nn.functional.dropout(x_279, 0.0, False, False) + x_279 = None + x_281 = x_280.permute(0, 3, 1, 2) + x_280 = None + reshape_24 = l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ = ( + None + ) + x_282 = x_281.mul(reshape_24) + x_281 = reshape_24 = None + x_283 = x_282 + x_272 + x_282 = x_272 = None + x_284 = torch.conv2d( + x_283, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = (None) + x_285 = x_284.permute(0, 2, 3, 1) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_, + 1e-06, + ) + x_285 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = (None) + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_288 = torch._C._nn.gelu(x_287) + x_287 = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + x_290 = torch._C._nn.linear( + x_289, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_291 = torch.nn.functional.dropout(x_290, 0.0, False, False) + x_290 = None + x_292 = x_291.permute(0, 3, 1, 2) + x_291 = None + reshape_25 = l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ = ( + None + ) + x_293 = x_292.mul(reshape_25) + x_292 = reshape_25 = None + x_294 = x_293 + x_283 + x_293 = x_283 = None + x_295 = torch.conv2d( + x_294, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = (None) + x_296 = x_295.permute(0, 2, 3, 1) + x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_, + 1e-06, + ) + x_296 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = (None) + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_299 = torch._C._nn.gelu(x_298) + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_301 = torch._C._nn.linear( + x_300, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_302 = torch.nn.functional.dropout(x_301, 0.0, False, False) + x_301 = None + x_303 = x_302.permute(0, 3, 1, 2) + x_302 = None + reshape_26 = l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ = ( + None + ) + x_304 = x_303.mul(reshape_26) + x_303 = reshape_26 = None + x_305 = x_304 + x_294 + x_304 = x_294 = None + x_306 = torch.conv2d( + x_305, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = (None) + x_307 = x_306.permute(0, 2, 3, 1) + x_306 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_, + 1e-06, + ) + x_307 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = (None) + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_310 = torch._C._nn.gelu(x_309) + x_309 = None + x_311 = torch.nn.functional.dropout(x_310, 0.0, False, False) + x_310 = None + x_312 = torch._C._nn.linear( + x_311, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_313 = torch.nn.functional.dropout(x_312, 0.0, False, False) + x_312 = None + x_314 = x_313.permute(0, 3, 1, 2) + x_313 = None + reshape_27 = l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ = ( + None + ) + x_315 = x_314.mul(reshape_27) + x_314 = reshape_27 = None + x_316 = x_315 + x_305 + x_315 = x_305 = None + x_317 = torch.conv2d( + x_316, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = (None) + x_318 = x_317.permute(0, 2, 3, 1) + x_317 = None + x_319 = torch.nn.functional.layer_norm( + x_318, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_, + 1e-06, + ) + x_318 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = (None) + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_321 = torch._C._nn.gelu(x_320) + x_320 = None + x_322 = torch.nn.functional.dropout(x_321, 0.0, False, False) + x_321 = None + x_323 = torch._C._nn.linear( + x_322, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_324 = torch.nn.functional.dropout(x_323, 0.0, False, False) + x_323 = None + x_325 = x_324.permute(0, 3, 1, 2) + x_324 = None + reshape_28 = l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ = ( + None + ) + x_326 = x_325.mul(reshape_28) + x_325 = reshape_28 = None + x_327 = x_326 + x_316 + x_326 = x_316 = None + x_328 = torch.conv2d( + x_327, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = (None) + x_329 = x_328.permute(0, 2, 3, 1) + x_328 = None + x_330 = torch.nn.functional.layer_norm( + x_329, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_, + 1e-06, + ) + x_329 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = (None) + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_332 = torch._C._nn.gelu(x_331) + x_331 = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_334 = torch._C._nn.linear( + x_333, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_335 = torch.nn.functional.dropout(x_334, 0.0, False, False) + x_334 = None + x_336 = x_335.permute(0, 3, 1, 2) + x_335 = None + reshape_29 = l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ = ( + None + ) + x_337 = x_336.mul(reshape_29) + x_336 = reshape_29 = None + x_338 = x_337 + x_327 + x_337 = x_327 = None + x_339 = torch.conv2d( + x_338, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = (None) + x_340 = x_339.permute(0, 2, 3, 1) + x_339 = None + x_341 = torch.nn.functional.layer_norm( + x_340, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_, + 1e-06, + ) + x_340 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = (None) + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_343 = torch._C._nn.gelu(x_342) + x_342 = None + x_344 = torch.nn.functional.dropout(x_343, 0.0, False, False) + x_343 = None + x_345 = torch._C._nn.linear( + x_344, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_346 = torch.nn.functional.dropout(x_345, 0.0, False, False) + x_345 = None + x_347 = x_346.permute(0, 3, 1, 2) + x_346 = None + reshape_30 = l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ = ( + None + ) + x_348 = x_347.mul(reshape_30) + x_347 = reshape_30 = None + x_349 = x_348 + x_338 + x_348 = x_338 = None + x_350 = torch.conv2d( + x_349, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = (None) + x_351 = x_350.permute(0, 2, 3, 1) + x_350 = None + x_352 = torch.nn.functional.layer_norm( + x_351, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_, + 1e-06, + ) + x_351 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = (None) + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_354 = torch._C._nn.gelu(x_353) + x_353 = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_356 = torch._C._nn.linear( + x_355, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_357 = torch.nn.functional.dropout(x_356, 0.0, False, False) + x_356 = None + x_358 = x_357.permute(0, 3, 1, 2) + x_357 = None + reshape_31 = l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ = ( + None + ) + x_359 = x_358.mul(reshape_31) + x_358 = reshape_31 = None + x_360 = x_359 + x_349 + x_359 = x_349 = None + x_361 = torch.conv2d( + x_360, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = (None) + x_362 = x_361.permute(0, 2, 3, 1) + x_361 = None + x_363 = torch.nn.functional.layer_norm( + x_362, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_, + 1e-06, + ) + x_362 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = (None) + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_365 = torch._C._nn.gelu(x_364) + x_364 = None + x_366 = torch.nn.functional.dropout(x_365, 0.0, False, False) + x_365 = None + x_367 = torch._C._nn.linear( + x_366, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_368 = torch.nn.functional.dropout(x_367, 0.0, False, False) + x_367 = None + x_369 = x_368.permute(0, 3, 1, 2) + x_368 = None + reshape_32 = l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ = ( + None + ) + x_370 = x_369.mul(reshape_32) + x_369 = reshape_32 = None + x_371 = x_370 + x_360 + x_370 = x_360 = None + x_372 = x_371.permute(0, 2, 3, 1) + x_371 = None + x_373 = torch.nn.functional.layer_norm( + x_372, + (768,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_372 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_374 = x_373.permute(0, 3, 1, 2) + x_373 = None + input_4 = torch.conv2d( + x_374, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_374 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_375 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_376 = x_375.permute(0, 2, 3, 1) + x_375 = None + x_377 = torch.nn.functional.layer_norm( + x_376, + (1536,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_376 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_378 = torch._C._nn.linear( + x_377, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_377 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_379 = torch._C._nn.gelu(x_378) + x_378 = None + x_380 = torch.nn.functional.dropout(x_379, 0.0, False, False) + x_379 = None + x_381 = torch._C._nn.linear( + x_380, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_380 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_382 = torch.nn.functional.dropout(x_381, 0.0, False, False) + x_381 = None + x_383 = x_382.permute(0, 3, 1, 2) + x_382 = None + reshape_33 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_384 = x_383.mul(reshape_33) + x_383 = reshape_33 = None + x_385 = x_384 + input_4 + x_384 = input_4 = None + x_386 = torch.conv2d( + x_385, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_387 = x_386.permute(0, 2, 3, 1) + x_386 = None + x_388 = torch.nn.functional.layer_norm( + x_387, + (1536,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_387 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_389 = torch._C._nn.linear( + x_388, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_388 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_390 = torch._C._nn.gelu(x_389) + x_389 = None + x_391 = torch.nn.functional.dropout(x_390, 0.0, False, False) + x_390 = None + x_392 = torch._C._nn.linear( + x_391, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_391 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_393 = torch.nn.functional.dropout(x_392, 0.0, False, False) + x_392 = None + x_394 = x_393.permute(0, 3, 1, 2) + x_393 = None + reshape_34 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_395 = x_394.mul(reshape_34) + x_394 = reshape_34 = None + x_396 = x_395 + x_385 + x_395 = x_385 = None + x_397 = torch.conv2d( + x_396, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_398 = x_397.permute(0, 2, 3, 1) + x_397 = None + x_399 = torch.nn.functional.layer_norm( + x_398, + (1536,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_398 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_400 = torch._C._nn.linear( + x_399, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_399 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_401 = torch._C._nn.gelu(x_400) + x_400 = None + x_402 = torch.nn.functional.dropout(x_401, 0.0, False, False) + x_401 = None + x_403 = torch._C._nn.linear( + x_402, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_402 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_404 = torch.nn.functional.dropout(x_403, 0.0, False, False) + x_403 = None + x_405 = x_404.permute(0, 3, 1, 2) + x_404 = None + reshape_35 = l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_406 = x_405.mul(reshape_35) + x_405 = reshape_35 = None + x_407 = x_406 + x_396 + x_406 = x_396 = None + x_408 = torch.nn.functional.adaptive_avg_pool2d(x_407, 1) + x_407 = None + x_409 = x_408.permute(0, 2, 3, 1) + x_408 = None + x_410 = torch.nn.functional.layer_norm( + x_409, + (1536,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_409 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_411 = x_410.permute(0, 3, 1, 2) + x_410 = None + x_412 = x_411.flatten(1, -1) + x_411 = None + input_5 = torch._C._nn.linear( + x_412, + l_self_modules_head_modules_pre_logits_modules_fc_parameters_weight_, + l_self_modules_head_modules_pre_logits_modules_fc_parameters_bias_, + ) + x_412 = ( + l_self_modules_head_modules_pre_logits_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_pre_logits_modules_fc_parameters_bias_ = None + input_6 = torch._C._nn.gelu(input_5) + input_5 = None + x_413 = torch.nn.functional.dropout(input_6, 0.0, False, False) + input_6 = None + x_414 = torch._C._nn.linear( + x_413, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_413 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_414,) diff --git a/samples/timm/convnext_large_mlp.clip_laion2b_augreg/weight_meta.py b/samples/timm/convnext_large_mlp.clip_laion2b_augreg/weight_meta.py new file mode 100644 index 0000000000..adb9903b8f --- /dev/null +++ b/samples/timm/convnext_large_mlp.clip_laion2b_augreg/weight_meta.py @@ -0,0 +1,3490 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [192, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.227 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [384, 192, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [768, 384, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [1536, 768, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_pre_logits_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_pre_logits_modules_fc_parameters_weight_" + shape = [1536, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_pre_logits_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_pre_logits_modules_fc_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [768, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None From b32ad38940b141d84bde5ab4418b6ec1f284ecef Mon Sep 17 00:00:00 2001 From: Liyulingyue <852433440@qq.com> Date: Tue, 5 Aug 2025 07:36:48 +0800 Subject: [PATCH 7/7] timm group --- .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../model.py | 2725 +++++ .../weight_meta.py | 3490 ++++++ .../convnext_nano.d1h_in1k/graph_hash.txt | 1 + .../convnext_nano.d1h_in1k/graph_net.json | 5 + .../timm/convnext_nano.d1h_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/convnext_nano.d1h_in1k/model.py | 1279 ++ .../convnext_nano.d1h_in1k/weight_meta.py | 1490 +++ .../convnext_nano_ols.d1h_in1k/graph_hash.txt | 1 + .../convnext_nano_ols.d1h_in1k/graph_net.json | 5 + .../convnext_nano_ols.d1h_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/convnext_nano_ols.d1h_in1k/model.py | 1299 ++ .../convnext_nano_ols.d1h_in1k/weight_meta.py | 1510 +++ .../timm/convnext_pico.d1_in1k/graph_hash.txt | 1 + .../timm/convnext_pico.d1_in1k/graph_net.json | 5 + .../timm/convnext_pico.d1_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/convnext_pico.d1_in1k/model.py | 1123 ++ .../timm/convnext_pico.d1_in1k/weight_meta.py | 1310 ++ .../convnext_pico_ols.d1_in1k/graph_hash.txt | 1 + .../convnext_pico_ols.d1_in1k/graph_net.json | 5 + .../convnext_pico_ols.d1_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/convnext_pico_ols.d1_in1k/model.py | 1143 ++ .../convnext_pico_ols.d1_in1k/weight_meta.py | 1330 ++ .../convnext_small.fb_in1k/graph_hash.txt | 1 + .../convnext_small.fb_in1k/graph_net.json | 5 + .../timm/convnext_small.fb_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/convnext_small.fb_in1k/model.py | 2707 +++++ .../convnext_small.fb_in1k/weight_meta.py | 3470 ++++++ .../timm/convnext_tiny.fb_in1k/graph_hash.txt | 1 + .../timm/convnext_tiny.fb_in1k/graph_net.json | 5 + .../timm/convnext_tiny.fb_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/convnext_tiny.fb_in1k/model.py | 1447 +++ .../timm/convnext_tiny.fb_in1k/weight_meta.py | 1850 +++ .../convnext_tiny_hnf.a2h_in1k/graph_hash.txt | 1 + .../convnext_tiny_hnf.a2h_in1k/graph_net.json | 5 + .../convnext_tiny_hnf.a2h_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/convnext_tiny_hnf.a2h_in1k/model.py | 1591 +++ .../convnext_tiny_hnf.a2h_in1k/weight_meta.py | 1850 +++ .../convnext_xlarge.fb_in22k/graph_hash.txt | 1 + .../convnext_xlarge.fb_in22k/graph_net.json | 5 + .../convnext_xlarge.fb_in22k/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/convnext_xlarge.fb_in22k/model.py | 2707 +++++ .../convnext_xlarge.fb_in22k/weight_meta.py | 3470 ++++++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../model.py | 2987 +++++ .../weight_meta.py | 3830 ++++++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../model.py | 983 ++ .../weight_meta.py | 974 ++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../model.py | 1005 ++ .../weight_meta.py | 994 ++ .../timm/convnextv2_atto.fcmae/graph_hash.txt | 1 + .../timm/convnextv2_atto.fcmae/graph_net.json | 5 + .../timm/convnextv2_atto.fcmae/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/convnextv2_atto.fcmae/model.py | 1299 ++ .../timm/convnextv2_atto.fcmae/weight_meta.py | 1410 +++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../convnextv2_atto.fcmae_ft_in1k/model.py | 1315 ++ .../weight_meta.py | 1430 +++ .../timm/convnextv2_base.fcmae/graph_hash.txt | 1 + .../timm/convnextv2_base.fcmae/graph_net.json | 5 + .../timm/convnextv2_base.fcmae/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/convnextv2_base.fcmae/model.py | 3267 +++++ .../timm/convnextv2_base.fcmae/weight_meta.py | 3810 ++++++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../convnextv2_base.fcmae_ft_in1k/model.py | 3283 +++++ .../weight_meta.py | 3830 ++++++ .../convnextv2_femto.fcmae/graph_hash.txt | 1 + .../convnextv2_femto.fcmae/graph_net.json | 5 + .../timm/convnextv2_femto.fcmae/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/convnextv2_femto.fcmae/model.py | 1299 ++ .../convnextv2_femto.fcmae/weight_meta.py | 1410 +++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../convnextv2_femto.fcmae_ft_in1k/model.py | 1315 ++ .../weight_meta.py | 1430 +++ .../timm/convnextv2_huge.fcmae/graph_hash.txt | 1 + .../timm/convnextv2_huge.fcmae/graph_net.json | 5 + .../timm/convnextv2_huge.fcmae/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/convnextv2_huge.fcmae/model.py | 3267 +++++ .../timm/convnextv2_huge.fcmae/weight_meta.py | 3810 ++++++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../convnextv2_huge.fcmae_ft_in1k/model.py | 3283 +++++ .../weight_meta.py | 3830 ++++++ .../convnextv2_large.fcmae/graph_hash.txt | 1 + .../convnextv2_large.fcmae/graph_net.json | 5 + .../timm/convnextv2_large.fcmae/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/convnextv2_large.fcmae/model.py | 3267 +++++ .../convnextv2_large.fcmae/weight_meta.py | 3810 ++++++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../convnextv2_large.fcmae_ft_in1k/model.py | 3283 +++++ .../weight_meta.py | 3830 ++++++ .../timm/convnextv2_nano.fcmae/graph_hash.txt | 1 + .../timm/convnextv2_nano.fcmae/graph_net.json | 5 + .../timm/convnextv2_nano.fcmae/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/convnextv2_nano.fcmae/model.py | 1487 +++ .../timm/convnextv2_nano.fcmae/weight_meta.py | 1610 +++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../convnextv2_nano.fcmae_ft_in1k/model.py | 1503 +++ .../weight_meta.py | 1630 +++ .../timm/convnextv2_pico.fcmae/graph_hash.txt | 1 + .../timm/convnextv2_pico.fcmae/graph_net.json | 5 + .../timm/convnextv2_pico.fcmae/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/convnextv2_pico.fcmae/model.py | 1299 ++ .../timm/convnextv2_pico.fcmae/weight_meta.py | 1410 +++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../convnextv2_pico.fcmae_ft_in1k/model.py | 1315 ++ .../weight_meta.py | 1430 +++ .../timm/convnextv2_tiny.fcmae/graph_hash.txt | 1 + .../timm/convnextv2_tiny.fcmae/graph_net.json | 5 + .../timm/convnextv2_tiny.fcmae/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/convnextv2_tiny.fcmae/model.py | 1719 +++ .../timm/convnextv2_tiny.fcmae/weight_meta.py | 2010 +++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../convnextv2_tiny.fcmae_ft_in1k/model.py | 1735 +++ .../weight_meta.py | 2030 ++++ .../timm/crossvit_15_240.in1k/graph_hash.txt | 1 + .../timm/crossvit_15_240.in1k/graph_net.json | 5 + .../timm/crossvit_15_240.in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/crossvit_15_240.in1k/model.py | 2733 +++++ .../timm/crossvit_15_240.in1k/weight_meta.py | 3408 ++++++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../crossvit_15_dagger_240.in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/crossvit_15_dagger_240.in1k/model.py | 2809 +++++ .../weight_meta.py | 3512 ++++++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../crossvit_15_dagger_408.in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/crossvit_15_dagger_408.in1k/model.py | 2814 +++++ .../weight_meta.py | 3522 ++++++ .../timm/crossvit_18_240.in1k/graph_hash.txt | 1 + .../timm/crossvit_18_240.in1k/graph_net.json | 5 + .../timm/crossvit_18_240.in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/crossvit_18_240.in1k/model.py | 3015 +++++ .../timm/crossvit_18_240.in1k/weight_meta.py | 3768 ++++++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../crossvit_18_dagger_240.in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/crossvit_18_dagger_240.in1k/model.py | 3091 +++++ .../weight_meta.py | 3872 ++++++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../crossvit_18_dagger_408.in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/crossvit_18_dagger_408.in1k/model.py | 3096 +++++ .../weight_meta.py | 3882 ++++++ .../timm/crossvit_9_240.in1k/graph_hash.txt | 1 + .../timm/crossvit_9_240.in1k/graph_net.json | 5 + .../timm/crossvit_9_240.in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/crossvit_9_240.in1k/model.py | 2169 ++++ .../timm/crossvit_9_240.in1k/weight_meta.py | 2688 ++++ .../crossvit_9_dagger_240.in1k/graph_hash.txt | 1 + .../crossvit_9_dagger_240.in1k/graph_net.json | 5 + .../crossvit_9_dagger_240.in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/crossvit_9_dagger_240.in1k/model.py | 2245 ++++ .../crossvit_9_dagger_240.in1k/weight_meta.py | 2792 +++++ .../crossvit_base_240.in1k/graph_hash.txt | 1 + .../crossvit_base_240.in1k/graph_net.json | 5 + .../timm/crossvit_base_240.in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/crossvit_base_240.in1k/model.py | 2451 ++++ .../crossvit_base_240.in1k/weight_meta.py | 3048 +++++ .../crossvit_small_240.in1k/graph_hash.txt | 1 + .../crossvit_small_240.in1k/graph_net.json | 5 + .../crossvit_small_240.in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/crossvit_small_240.in1k/model.py | 2451 ++++ .../crossvit_small_240.in1k/weight_meta.py | 3048 +++++ .../crossvit_tiny_240.in1k/graph_hash.txt | 1 + .../crossvit_tiny_240.in1k/graph_net.json | 5 + .../timm/crossvit_tiny_240.in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/crossvit_tiny_240.in1k/model.py | 2451 ++++ .../crossvit_tiny_240.in1k/weight_meta.py | 3048 +++++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../cs3darknet_focus_l.c2ns_in1k/model.py | 1992 +++ .../weight_meta.py | 2812 +++++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../cs3darknet_focus_m.c2ns_in1k/model.py | 1502 +++ .../weight_meta.py | 2112 ++++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../model.py | 1152 ++ .../weight_meta.py | 1612 +++ .../cs3darknet_l.c2ns_in1k/graph_hash.txt | 1 + .../cs3darknet_l.c2ns_in1k/graph_net.json | 5 + .../timm/cs3darknet_l.c2ns_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/cs3darknet_l.c2ns_in1k/model.py | 2041 ++++ .../cs3darknet_l.c2ns_in1k/weight_meta.py | 2862 +++++ .../cs3darknet_m.c2ns_in1k/graph_hash.txt | 1 + .../cs3darknet_m.c2ns_in1k/graph_net.json | 5 + .../timm/cs3darknet_m.c2ns_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/cs3darknet_m.c2ns_in1k/model.py | 1551 +++ .../cs3darknet_m.c2ns_in1k/weight_meta.py | 2162 ++++ .../cs3darknet_x.c2ns_in1k/graph_hash.txt | 1 + .../cs3darknet_x.c2ns_in1k/graph_net.json | 5 + .../timm/cs3darknet_x.c2ns_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/cs3darknet_x.c2ns_in1k/model.py | 2251 ++++ .../cs3darknet_x.c2ns_in1k/weight_meta.py | 3162 +++++ .../timm/cs3edgenet_x.c2_in1k/graph_hash.txt | 1 + .../timm/cs3edgenet_x.c2_in1k/graph_net.json | 5 + .../timm/cs3edgenet_x.c2_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/cs3edgenet_x.c2_in1k/model.py | 2251 ++++ .../timm/cs3edgenet_x.c2_in1k/weight_meta.py | 3162 +++++ .../cs3se_edgenet_x.c2ns_in1k/graph_hash.txt | 1 + .../cs3se_edgenet_x.c2ns_in1k/graph_net.json | 5 + .../cs3se_edgenet_x.c2ns_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/cs3se_edgenet_x.c2ns_in1k/model.py | 3091 +++++ .../cs3se_edgenet_x.c2ns_in1k/weight_meta.py | 4122 +++++++ .../cs3sedarknet_l.c2ns_in1k/graph_hash.txt | 1 + .../cs3sedarknet_l.c2ns_in1k/graph_net.json | 5 + .../cs3sedarknet_l.c2ns_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/cs3sedarknet_l.c2ns_in1k/model.py | 2776 +++++ .../cs3sedarknet_l.c2ns_in1k/weight_meta.py | 3702 ++++++ .../cs3sedarknet_x.c2ns_in1k/graph_hash.txt | 1 + .../cs3sedarknet_x.c2ns_in1k/graph_net.json | 5 + .../cs3sedarknet_x.c2ns_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/cs3sedarknet_x.c2ns_in1k/model.py | 3091 +++++ .../cs3sedarknet_x.c2ns_in1k/weight_meta.py | 4122 +++++++ .../timm/cspdarknet53.ra_in1k/graph_hash.txt | 1 + .../timm/cspdarknet53.ra_in1k/graph_net.json | 5 + .../timm/cspdarknet53.ra_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/cspdarknet53.ra_in1k/model.py | 2417 ++++ .../timm/cspdarknet53.ra_in1k/weight_meta.py | 3418 ++++++ .../timm/cspresnet50.ra_in1k/graph_hash.txt | 1 + .../timm/cspresnet50.ra_in1k/graph_net.json | 5 + .../timm/cspresnet50.ra_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/cspresnet50.ra_in1k/model.py | 1964 +++ .../timm/cspresnet50.ra_in1k/weight_meta.py | 2810 +++++ .../timm/cspresnext50.ra_in1k/graph_hash.txt | 1 + .../timm/cspresnext50.ra_in1k/graph_net.json | 5 + .../timm/cspresnext50.ra_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/cspresnext50.ra_in1k/model.py | 1964 +++ .../timm/cspresnext50.ra_in1k/weight_meta.py | 2810 +++++ .../timm/darknet53.c2ns_in1k/graph_hash.txt | 1 + .../timm/darknet53.c2ns_in1k/graph_net.json | 5 + .../timm/darknet53.c2ns_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/darknet53.c2ns_in1k/model.py | 1857 +++ .../timm/darknet53.c2ns_in1k/weight_meta.py | 2648 ++++ .../timm/darknetaa53.c2ns_in1k/graph_hash.txt | 1 + .../timm/darknetaa53.c2ns_in1k/graph_net.json | 5 + .../timm/darknetaa53.c2ns_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/darknetaa53.c2ns_in1k/model.py | 1867 +++ .../timm/darknetaa53.c2ns_in1k/weight_meta.py | 2638 ++++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../deit3_base_patch16_224.fb_in1k/model.py | 1868 +++ .../weight_meta.py | 1796 +++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../deit3_huge_patch14_224.fb_in1k/model.py | 4788 ++++++++ .../weight_meta.py | 4676 +++++++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../deit3_large_patch16_224.fb_in1k/model.py | 3620 ++++++ .../weight_meta.py | 3524 ++++++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../deit3_medium_patch16_224.fb_in1k/model.py | 1868 +++ .../weight_meta.py | 1796 +++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../deit3_small_patch16_224.fb_in1k/model.py | 1868 +++ .../weight_meta.py | 1796 +++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../model.py | 1747 +++ .../weight_meta.py | 1586 +++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../deit_base_patch16_224.fb_in1k/model.py | 1724 +++ .../weight_meta.py | 1556 +++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../model.py | 1747 +++ .../weight_meta.py | 1586 +++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../deit_small_patch16_224.fb_in1k/model.py | 1724 +++ .../weight_meta.py | 1556 +++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../model.py | 1747 +++ .../weight_meta.py | 1586 +++ .../graph_hash.txt | 1 + .../graph_net.json | 5 + .../input_meta.py | 0 .../input_tensor_constraints.py | 0 .../deit_tiny_patch16_224.fb_in1k/model.py | 1724 +++ .../weight_meta.py | 1556 +++ .../timm/densenet121.ra_in1k/graph_hash.txt | 1 + .../timm/densenet121.ra_in1k/graph_net.json | 5 + .../timm/densenet121.ra_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/densenet121.ra_in1k/model.py | 4977 ++++++++ .../timm/densenet121.ra_in1k/weight_meta.py | 6090 ++++++++++ .../timm/densenet161.tv_in1k/graph_hash.txt | 1 + .../timm/densenet161.tv_in1k/graph_net.json | 5 + .../timm/densenet161.tv_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/densenet161.tv_in1k/model.py | 6987 +++++++++++ .../timm/densenet161.tv_in1k/weight_meta.py | 8090 +++++++++++++ .../timm/densenet169.tv_in1k/graph_hash.txt | 1 + .../timm/densenet169.tv_in1k/graph_net.json | 5 + .../timm/densenet169.tv_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/densenet169.tv_in1k/model.py | 7373 +++++++++++ .../timm/densenet169.tv_in1k/weight_meta.py | 8490 +++++++++++++ .../timm/densenet201.tv_in1k/graph_hash.txt | 1 + .../timm/densenet201.tv_in1k/graph_net.json | 5 + .../timm/densenet201.tv_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 samples/timm/densenet201.tv_in1k/model.py | 9205 ++++++++++++++ .../timm/densenet201.tv_in1k/weight_meta.py | 10090 ++++++++++++++++ .../densenetblur121d.ra_in1k/graph_hash.txt | 1 + .../densenetblur121d.ra_in1k/graph_net.json | 5 + .../densenetblur121d.ra_in1k/input_meta.py | 0 .../input_tensor_constraints.py | 0 .../timm/densenetblur121d.ra_in1k/model.py | 5090 ++++++++ .../densenetblur121d.ra_in1k/weight_meta.py | 6200 ++++++++++ samples/timm/dla34.in1k/graph_hash.txt | 1 + samples/timm/dla34.in1k/graph_net.json | 5 + samples/timm/dla34.in1k/input_meta.py | 0 .../dla34.in1k/input_tensor_constraints.py | 0 samples/timm/dla34.in1k/model.py | 1596 +++ samples/timm/dla34.in1k/weight_meta.py | 1940 +++ 426 files changed, 385047 insertions(+) create mode 100644 samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/graph_hash.txt create mode 100644 samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/graph_net.json create mode 100644 samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/input_meta.py create mode 100644 samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/input_tensor_constraints.py create mode 100644 samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/model.py create mode 100644 samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/weight_meta.py create mode 100644 samples/timm/convnext_nano.d1h_in1k/graph_hash.txt create mode 100644 samples/timm/convnext_nano.d1h_in1k/graph_net.json create mode 100644 samples/timm/convnext_nano.d1h_in1k/input_meta.py create mode 100644 samples/timm/convnext_nano.d1h_in1k/input_tensor_constraints.py create mode 100644 samples/timm/convnext_nano.d1h_in1k/model.py create mode 100644 samples/timm/convnext_nano.d1h_in1k/weight_meta.py create mode 100644 samples/timm/convnext_nano_ols.d1h_in1k/graph_hash.txt create mode 100644 samples/timm/convnext_nano_ols.d1h_in1k/graph_net.json create mode 100644 samples/timm/convnext_nano_ols.d1h_in1k/input_meta.py create mode 100644 samples/timm/convnext_nano_ols.d1h_in1k/input_tensor_constraints.py create mode 100644 samples/timm/convnext_nano_ols.d1h_in1k/model.py create mode 100644 samples/timm/convnext_nano_ols.d1h_in1k/weight_meta.py create mode 100644 samples/timm/convnext_pico.d1_in1k/graph_hash.txt create mode 100644 samples/timm/convnext_pico.d1_in1k/graph_net.json create mode 100644 samples/timm/convnext_pico.d1_in1k/input_meta.py create mode 100644 samples/timm/convnext_pico.d1_in1k/input_tensor_constraints.py create mode 100644 samples/timm/convnext_pico.d1_in1k/model.py create mode 100644 samples/timm/convnext_pico.d1_in1k/weight_meta.py create mode 100644 samples/timm/convnext_pico_ols.d1_in1k/graph_hash.txt create mode 100644 samples/timm/convnext_pico_ols.d1_in1k/graph_net.json create mode 100644 samples/timm/convnext_pico_ols.d1_in1k/input_meta.py create mode 100644 samples/timm/convnext_pico_ols.d1_in1k/input_tensor_constraints.py create mode 100644 samples/timm/convnext_pico_ols.d1_in1k/model.py create mode 100644 samples/timm/convnext_pico_ols.d1_in1k/weight_meta.py create mode 100644 samples/timm/convnext_small.fb_in1k/graph_hash.txt create mode 100644 samples/timm/convnext_small.fb_in1k/graph_net.json create mode 100644 samples/timm/convnext_small.fb_in1k/input_meta.py create mode 100644 samples/timm/convnext_small.fb_in1k/input_tensor_constraints.py create mode 100644 samples/timm/convnext_small.fb_in1k/model.py create mode 100644 samples/timm/convnext_small.fb_in1k/weight_meta.py create mode 100644 samples/timm/convnext_tiny.fb_in1k/graph_hash.txt create mode 100644 samples/timm/convnext_tiny.fb_in1k/graph_net.json create mode 100644 samples/timm/convnext_tiny.fb_in1k/input_meta.py create mode 100644 samples/timm/convnext_tiny.fb_in1k/input_tensor_constraints.py create mode 100644 samples/timm/convnext_tiny.fb_in1k/model.py create mode 100644 samples/timm/convnext_tiny.fb_in1k/weight_meta.py create mode 100644 samples/timm/convnext_tiny_hnf.a2h_in1k/graph_hash.txt create mode 100644 samples/timm/convnext_tiny_hnf.a2h_in1k/graph_net.json create mode 100644 samples/timm/convnext_tiny_hnf.a2h_in1k/input_meta.py create mode 100644 samples/timm/convnext_tiny_hnf.a2h_in1k/input_tensor_constraints.py create mode 100644 samples/timm/convnext_tiny_hnf.a2h_in1k/model.py create mode 100644 samples/timm/convnext_tiny_hnf.a2h_in1k/weight_meta.py create mode 100644 samples/timm/convnext_xlarge.fb_in22k/graph_hash.txt create mode 100644 samples/timm/convnext_xlarge.fb_in22k/graph_net.json create mode 100644 samples/timm/convnext_xlarge.fb_in22k/input_meta.py create mode 100644 samples/timm/convnext_xlarge.fb_in22k/input_tensor_constraints.py create mode 100644 samples/timm/convnext_xlarge.fb_in22k/model.py create mode 100644 samples/timm/convnext_xlarge.fb_in22k/weight_meta.py create mode 100644 samples/timm/convnext_xxlarge.clip_laion2b_rewind/graph_hash.txt create mode 100644 samples/timm/convnext_xxlarge.clip_laion2b_rewind/graph_net.json create mode 100644 samples/timm/convnext_xxlarge.clip_laion2b_rewind/input_meta.py create mode 100644 samples/timm/convnext_xxlarge.clip_laion2b_rewind/input_tensor_constraints.py create mode 100644 samples/timm/convnext_xxlarge.clip_laion2b_rewind/model.py create mode 100644 samples/timm/convnext_xxlarge.clip_laion2b_rewind/weight_meta.py create mode 100644 samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/graph_hash.txt create mode 100644 samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/graph_net.json create mode 100644 samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/input_meta.py create mode 100644 samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/input_tensor_constraints.py create mode 100644 samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/model.py create mode 100644 samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/weight_meta.py create mode 100644 samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/graph_hash.txt create mode 100644 samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/graph_net.json create mode 100644 samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/input_meta.py create mode 100644 samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/input_tensor_constraints.py create mode 100644 samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/model.py create mode 100644 samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/weight_meta.py create mode 100644 samples/timm/convnextv2_atto.fcmae/graph_hash.txt create mode 100644 samples/timm/convnextv2_atto.fcmae/graph_net.json create mode 100644 samples/timm/convnextv2_atto.fcmae/input_meta.py create mode 100644 samples/timm/convnextv2_atto.fcmae/input_tensor_constraints.py create mode 100644 samples/timm/convnextv2_atto.fcmae/model.py create mode 100644 samples/timm/convnextv2_atto.fcmae/weight_meta.py create mode 100644 samples/timm/convnextv2_atto.fcmae_ft_in1k/graph_hash.txt create mode 100644 samples/timm/convnextv2_atto.fcmae_ft_in1k/graph_net.json create mode 100644 samples/timm/convnextv2_atto.fcmae_ft_in1k/input_meta.py create mode 100644 samples/timm/convnextv2_atto.fcmae_ft_in1k/input_tensor_constraints.py create mode 100644 samples/timm/convnextv2_atto.fcmae_ft_in1k/model.py create mode 100644 samples/timm/convnextv2_atto.fcmae_ft_in1k/weight_meta.py create mode 100644 samples/timm/convnextv2_base.fcmae/graph_hash.txt create mode 100644 samples/timm/convnextv2_base.fcmae/graph_net.json create mode 100644 samples/timm/convnextv2_base.fcmae/input_meta.py create mode 100644 samples/timm/convnextv2_base.fcmae/input_tensor_constraints.py create mode 100644 samples/timm/convnextv2_base.fcmae/model.py create mode 100644 samples/timm/convnextv2_base.fcmae/weight_meta.py create mode 100644 samples/timm/convnextv2_base.fcmae_ft_in1k/graph_hash.txt create mode 100644 samples/timm/convnextv2_base.fcmae_ft_in1k/graph_net.json create mode 100644 samples/timm/convnextv2_base.fcmae_ft_in1k/input_meta.py create mode 100644 samples/timm/convnextv2_base.fcmae_ft_in1k/input_tensor_constraints.py create mode 100644 samples/timm/convnextv2_base.fcmae_ft_in1k/model.py create mode 100644 samples/timm/convnextv2_base.fcmae_ft_in1k/weight_meta.py create mode 100644 samples/timm/convnextv2_femto.fcmae/graph_hash.txt create mode 100644 samples/timm/convnextv2_femto.fcmae/graph_net.json create mode 100644 samples/timm/convnextv2_femto.fcmae/input_meta.py create mode 100644 samples/timm/convnextv2_femto.fcmae/input_tensor_constraints.py create mode 100644 samples/timm/convnextv2_femto.fcmae/model.py create mode 100644 samples/timm/convnextv2_femto.fcmae/weight_meta.py create mode 100644 samples/timm/convnextv2_femto.fcmae_ft_in1k/graph_hash.txt create mode 100644 samples/timm/convnextv2_femto.fcmae_ft_in1k/graph_net.json create mode 100644 samples/timm/convnextv2_femto.fcmae_ft_in1k/input_meta.py create mode 100644 samples/timm/convnextv2_femto.fcmae_ft_in1k/input_tensor_constraints.py create mode 100644 samples/timm/convnextv2_femto.fcmae_ft_in1k/model.py create mode 100644 samples/timm/convnextv2_femto.fcmae_ft_in1k/weight_meta.py create mode 100644 samples/timm/convnextv2_huge.fcmae/graph_hash.txt create mode 100644 samples/timm/convnextv2_huge.fcmae/graph_net.json create mode 100644 samples/timm/convnextv2_huge.fcmae/input_meta.py create mode 100644 samples/timm/convnextv2_huge.fcmae/input_tensor_constraints.py create mode 100644 samples/timm/convnextv2_huge.fcmae/model.py create mode 100644 samples/timm/convnextv2_huge.fcmae/weight_meta.py create mode 100644 samples/timm/convnextv2_huge.fcmae_ft_in1k/graph_hash.txt create mode 100644 samples/timm/convnextv2_huge.fcmae_ft_in1k/graph_net.json create mode 100644 samples/timm/convnextv2_huge.fcmae_ft_in1k/input_meta.py create mode 100644 samples/timm/convnextv2_huge.fcmae_ft_in1k/input_tensor_constraints.py create mode 100644 samples/timm/convnextv2_huge.fcmae_ft_in1k/model.py create mode 100644 samples/timm/convnextv2_huge.fcmae_ft_in1k/weight_meta.py create mode 100644 samples/timm/convnextv2_large.fcmae/graph_hash.txt create mode 100644 samples/timm/convnextv2_large.fcmae/graph_net.json create mode 100644 samples/timm/convnextv2_large.fcmae/input_meta.py create mode 100644 samples/timm/convnextv2_large.fcmae/input_tensor_constraints.py create mode 100644 samples/timm/convnextv2_large.fcmae/model.py create mode 100644 samples/timm/convnextv2_large.fcmae/weight_meta.py create mode 100644 samples/timm/convnextv2_large.fcmae_ft_in1k/graph_hash.txt create mode 100644 samples/timm/convnextv2_large.fcmae_ft_in1k/graph_net.json create mode 100644 samples/timm/convnextv2_large.fcmae_ft_in1k/input_meta.py create mode 100644 samples/timm/convnextv2_large.fcmae_ft_in1k/input_tensor_constraints.py create mode 100644 samples/timm/convnextv2_large.fcmae_ft_in1k/model.py create mode 100644 samples/timm/convnextv2_large.fcmae_ft_in1k/weight_meta.py create mode 100644 samples/timm/convnextv2_nano.fcmae/graph_hash.txt create mode 100644 samples/timm/convnextv2_nano.fcmae/graph_net.json create mode 100644 samples/timm/convnextv2_nano.fcmae/input_meta.py create mode 100644 samples/timm/convnextv2_nano.fcmae/input_tensor_constraints.py create mode 100644 samples/timm/convnextv2_nano.fcmae/model.py create mode 100644 samples/timm/convnextv2_nano.fcmae/weight_meta.py create mode 100644 samples/timm/convnextv2_nano.fcmae_ft_in1k/graph_hash.txt create mode 100644 samples/timm/convnextv2_nano.fcmae_ft_in1k/graph_net.json create mode 100644 samples/timm/convnextv2_nano.fcmae_ft_in1k/input_meta.py create mode 100644 samples/timm/convnextv2_nano.fcmae_ft_in1k/input_tensor_constraints.py create mode 100644 samples/timm/convnextv2_nano.fcmae_ft_in1k/model.py create mode 100644 samples/timm/convnextv2_nano.fcmae_ft_in1k/weight_meta.py create mode 100644 samples/timm/convnextv2_pico.fcmae/graph_hash.txt create mode 100644 samples/timm/convnextv2_pico.fcmae/graph_net.json create mode 100644 samples/timm/convnextv2_pico.fcmae/input_meta.py create mode 100644 samples/timm/convnextv2_pico.fcmae/input_tensor_constraints.py create mode 100644 samples/timm/convnextv2_pico.fcmae/model.py create mode 100644 samples/timm/convnextv2_pico.fcmae/weight_meta.py create mode 100644 samples/timm/convnextv2_pico.fcmae_ft_in1k/graph_hash.txt create mode 100644 samples/timm/convnextv2_pico.fcmae_ft_in1k/graph_net.json create mode 100644 samples/timm/convnextv2_pico.fcmae_ft_in1k/input_meta.py create mode 100644 samples/timm/convnextv2_pico.fcmae_ft_in1k/input_tensor_constraints.py create mode 100644 samples/timm/convnextv2_pico.fcmae_ft_in1k/model.py create mode 100644 samples/timm/convnextv2_pico.fcmae_ft_in1k/weight_meta.py create mode 100644 samples/timm/convnextv2_tiny.fcmae/graph_hash.txt create mode 100644 samples/timm/convnextv2_tiny.fcmae/graph_net.json create mode 100644 samples/timm/convnextv2_tiny.fcmae/input_meta.py create mode 100644 samples/timm/convnextv2_tiny.fcmae/input_tensor_constraints.py create mode 100644 samples/timm/convnextv2_tiny.fcmae/model.py create mode 100644 samples/timm/convnextv2_tiny.fcmae/weight_meta.py create mode 100644 samples/timm/convnextv2_tiny.fcmae_ft_in1k/graph_hash.txt create mode 100644 samples/timm/convnextv2_tiny.fcmae_ft_in1k/graph_net.json create mode 100644 samples/timm/convnextv2_tiny.fcmae_ft_in1k/input_meta.py create mode 100644 samples/timm/convnextv2_tiny.fcmae_ft_in1k/input_tensor_constraints.py create mode 100644 samples/timm/convnextv2_tiny.fcmae_ft_in1k/model.py create mode 100644 samples/timm/convnextv2_tiny.fcmae_ft_in1k/weight_meta.py create mode 100644 samples/timm/crossvit_15_240.in1k/graph_hash.txt create mode 100644 samples/timm/crossvit_15_240.in1k/graph_net.json create mode 100644 samples/timm/crossvit_15_240.in1k/input_meta.py create mode 100644 samples/timm/crossvit_15_240.in1k/input_tensor_constraints.py create mode 100644 samples/timm/crossvit_15_240.in1k/model.py create mode 100644 samples/timm/crossvit_15_240.in1k/weight_meta.py create mode 100644 samples/timm/crossvit_15_dagger_240.in1k/graph_hash.txt create mode 100644 samples/timm/crossvit_15_dagger_240.in1k/graph_net.json create mode 100644 samples/timm/crossvit_15_dagger_240.in1k/input_meta.py create mode 100644 samples/timm/crossvit_15_dagger_240.in1k/input_tensor_constraints.py create mode 100644 samples/timm/crossvit_15_dagger_240.in1k/model.py create mode 100644 samples/timm/crossvit_15_dagger_240.in1k/weight_meta.py create mode 100644 samples/timm/crossvit_15_dagger_408.in1k/graph_hash.txt create mode 100644 samples/timm/crossvit_15_dagger_408.in1k/graph_net.json create mode 100644 samples/timm/crossvit_15_dagger_408.in1k/input_meta.py create mode 100644 samples/timm/crossvit_15_dagger_408.in1k/input_tensor_constraints.py create mode 100644 samples/timm/crossvit_15_dagger_408.in1k/model.py create mode 100644 samples/timm/crossvit_15_dagger_408.in1k/weight_meta.py create mode 100644 samples/timm/crossvit_18_240.in1k/graph_hash.txt create mode 100644 samples/timm/crossvit_18_240.in1k/graph_net.json create mode 100644 samples/timm/crossvit_18_240.in1k/input_meta.py create mode 100644 samples/timm/crossvit_18_240.in1k/input_tensor_constraints.py create mode 100644 samples/timm/crossvit_18_240.in1k/model.py create mode 100644 samples/timm/crossvit_18_240.in1k/weight_meta.py create mode 100644 samples/timm/crossvit_18_dagger_240.in1k/graph_hash.txt create mode 100644 samples/timm/crossvit_18_dagger_240.in1k/graph_net.json create mode 100644 samples/timm/crossvit_18_dagger_240.in1k/input_meta.py create mode 100644 samples/timm/crossvit_18_dagger_240.in1k/input_tensor_constraints.py create mode 100644 samples/timm/crossvit_18_dagger_240.in1k/model.py create mode 100644 samples/timm/crossvit_18_dagger_240.in1k/weight_meta.py create mode 100644 samples/timm/crossvit_18_dagger_408.in1k/graph_hash.txt create mode 100644 samples/timm/crossvit_18_dagger_408.in1k/graph_net.json create mode 100644 samples/timm/crossvit_18_dagger_408.in1k/input_meta.py create mode 100644 samples/timm/crossvit_18_dagger_408.in1k/input_tensor_constraints.py create mode 100644 samples/timm/crossvit_18_dagger_408.in1k/model.py create mode 100644 samples/timm/crossvit_18_dagger_408.in1k/weight_meta.py create mode 100644 samples/timm/crossvit_9_240.in1k/graph_hash.txt create mode 100644 samples/timm/crossvit_9_240.in1k/graph_net.json create mode 100644 samples/timm/crossvit_9_240.in1k/input_meta.py create mode 100644 samples/timm/crossvit_9_240.in1k/input_tensor_constraints.py create mode 100644 samples/timm/crossvit_9_240.in1k/model.py create mode 100644 samples/timm/crossvit_9_240.in1k/weight_meta.py create mode 100644 samples/timm/crossvit_9_dagger_240.in1k/graph_hash.txt create mode 100644 samples/timm/crossvit_9_dagger_240.in1k/graph_net.json create mode 100644 samples/timm/crossvit_9_dagger_240.in1k/input_meta.py create mode 100644 samples/timm/crossvit_9_dagger_240.in1k/input_tensor_constraints.py create mode 100644 samples/timm/crossvit_9_dagger_240.in1k/model.py create mode 100644 samples/timm/crossvit_9_dagger_240.in1k/weight_meta.py create mode 100644 samples/timm/crossvit_base_240.in1k/graph_hash.txt create mode 100644 samples/timm/crossvit_base_240.in1k/graph_net.json create mode 100644 samples/timm/crossvit_base_240.in1k/input_meta.py create mode 100644 samples/timm/crossvit_base_240.in1k/input_tensor_constraints.py create mode 100644 samples/timm/crossvit_base_240.in1k/model.py create mode 100644 samples/timm/crossvit_base_240.in1k/weight_meta.py create mode 100644 samples/timm/crossvit_small_240.in1k/graph_hash.txt create mode 100644 samples/timm/crossvit_small_240.in1k/graph_net.json create mode 100644 samples/timm/crossvit_small_240.in1k/input_meta.py create mode 100644 samples/timm/crossvit_small_240.in1k/input_tensor_constraints.py create mode 100644 samples/timm/crossvit_small_240.in1k/model.py create mode 100644 samples/timm/crossvit_small_240.in1k/weight_meta.py create mode 100644 samples/timm/crossvit_tiny_240.in1k/graph_hash.txt create mode 100644 samples/timm/crossvit_tiny_240.in1k/graph_net.json create mode 100644 samples/timm/crossvit_tiny_240.in1k/input_meta.py create mode 100644 samples/timm/crossvit_tiny_240.in1k/input_tensor_constraints.py create mode 100644 samples/timm/crossvit_tiny_240.in1k/model.py create mode 100644 samples/timm/crossvit_tiny_240.in1k/weight_meta.py create mode 100644 samples/timm/cs3darknet_focus_l.c2ns_in1k/graph_hash.txt create mode 100644 samples/timm/cs3darknet_focus_l.c2ns_in1k/graph_net.json create mode 100644 samples/timm/cs3darknet_focus_l.c2ns_in1k/input_meta.py create mode 100644 samples/timm/cs3darknet_focus_l.c2ns_in1k/input_tensor_constraints.py create mode 100644 samples/timm/cs3darknet_focus_l.c2ns_in1k/model.py create mode 100644 samples/timm/cs3darknet_focus_l.c2ns_in1k/weight_meta.py create mode 100644 samples/timm/cs3darknet_focus_m.c2ns_in1k/graph_hash.txt create mode 100644 samples/timm/cs3darknet_focus_m.c2ns_in1k/graph_net.json create mode 100644 samples/timm/cs3darknet_focus_m.c2ns_in1k/input_meta.py create mode 100644 samples/timm/cs3darknet_focus_m.c2ns_in1k/input_tensor_constraints.py create mode 100644 samples/timm/cs3darknet_focus_m.c2ns_in1k/model.py create mode 100644 samples/timm/cs3darknet_focus_m.c2ns_in1k/weight_meta.py create mode 100644 samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/graph_hash.txt create mode 100644 samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/graph_net.json create mode 100644 samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/input_meta.py create mode 100644 samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/input_tensor_constraints.py create mode 100644 samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/model.py create mode 100644 samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/weight_meta.py create mode 100644 samples/timm/cs3darknet_l.c2ns_in1k/graph_hash.txt create mode 100644 samples/timm/cs3darknet_l.c2ns_in1k/graph_net.json create mode 100644 samples/timm/cs3darknet_l.c2ns_in1k/input_meta.py create mode 100644 samples/timm/cs3darknet_l.c2ns_in1k/input_tensor_constraints.py create mode 100644 samples/timm/cs3darknet_l.c2ns_in1k/model.py create mode 100644 samples/timm/cs3darknet_l.c2ns_in1k/weight_meta.py create mode 100644 samples/timm/cs3darknet_m.c2ns_in1k/graph_hash.txt create mode 100644 samples/timm/cs3darknet_m.c2ns_in1k/graph_net.json create mode 100644 samples/timm/cs3darknet_m.c2ns_in1k/input_meta.py create mode 100644 samples/timm/cs3darknet_m.c2ns_in1k/input_tensor_constraints.py create mode 100644 samples/timm/cs3darknet_m.c2ns_in1k/model.py create mode 100644 samples/timm/cs3darknet_m.c2ns_in1k/weight_meta.py create mode 100644 samples/timm/cs3darknet_x.c2ns_in1k/graph_hash.txt create mode 100644 samples/timm/cs3darknet_x.c2ns_in1k/graph_net.json create mode 100644 samples/timm/cs3darknet_x.c2ns_in1k/input_meta.py create mode 100644 samples/timm/cs3darknet_x.c2ns_in1k/input_tensor_constraints.py create mode 100644 samples/timm/cs3darknet_x.c2ns_in1k/model.py create mode 100644 samples/timm/cs3darknet_x.c2ns_in1k/weight_meta.py create mode 100644 samples/timm/cs3edgenet_x.c2_in1k/graph_hash.txt create mode 100644 samples/timm/cs3edgenet_x.c2_in1k/graph_net.json create mode 100644 samples/timm/cs3edgenet_x.c2_in1k/input_meta.py create mode 100644 samples/timm/cs3edgenet_x.c2_in1k/input_tensor_constraints.py create mode 100644 samples/timm/cs3edgenet_x.c2_in1k/model.py create mode 100644 samples/timm/cs3edgenet_x.c2_in1k/weight_meta.py create mode 100644 samples/timm/cs3se_edgenet_x.c2ns_in1k/graph_hash.txt create mode 100644 samples/timm/cs3se_edgenet_x.c2ns_in1k/graph_net.json create mode 100644 samples/timm/cs3se_edgenet_x.c2ns_in1k/input_meta.py create mode 100644 samples/timm/cs3se_edgenet_x.c2ns_in1k/input_tensor_constraints.py create mode 100644 samples/timm/cs3se_edgenet_x.c2ns_in1k/model.py create mode 100644 samples/timm/cs3se_edgenet_x.c2ns_in1k/weight_meta.py create mode 100644 samples/timm/cs3sedarknet_l.c2ns_in1k/graph_hash.txt create mode 100644 samples/timm/cs3sedarknet_l.c2ns_in1k/graph_net.json create mode 100644 samples/timm/cs3sedarknet_l.c2ns_in1k/input_meta.py create mode 100644 samples/timm/cs3sedarknet_l.c2ns_in1k/input_tensor_constraints.py create mode 100644 samples/timm/cs3sedarknet_l.c2ns_in1k/model.py create mode 100644 samples/timm/cs3sedarknet_l.c2ns_in1k/weight_meta.py create mode 100644 samples/timm/cs3sedarknet_x.c2ns_in1k/graph_hash.txt create mode 100644 samples/timm/cs3sedarknet_x.c2ns_in1k/graph_net.json create mode 100644 samples/timm/cs3sedarknet_x.c2ns_in1k/input_meta.py create mode 100644 samples/timm/cs3sedarknet_x.c2ns_in1k/input_tensor_constraints.py create mode 100644 samples/timm/cs3sedarknet_x.c2ns_in1k/model.py create mode 100644 samples/timm/cs3sedarknet_x.c2ns_in1k/weight_meta.py create mode 100644 samples/timm/cspdarknet53.ra_in1k/graph_hash.txt create mode 100644 samples/timm/cspdarknet53.ra_in1k/graph_net.json create mode 100644 samples/timm/cspdarknet53.ra_in1k/input_meta.py create mode 100644 samples/timm/cspdarknet53.ra_in1k/input_tensor_constraints.py create mode 100644 samples/timm/cspdarknet53.ra_in1k/model.py create mode 100644 samples/timm/cspdarknet53.ra_in1k/weight_meta.py create mode 100644 samples/timm/cspresnet50.ra_in1k/graph_hash.txt create mode 100644 samples/timm/cspresnet50.ra_in1k/graph_net.json create mode 100644 samples/timm/cspresnet50.ra_in1k/input_meta.py create mode 100644 samples/timm/cspresnet50.ra_in1k/input_tensor_constraints.py create mode 100644 samples/timm/cspresnet50.ra_in1k/model.py create mode 100644 samples/timm/cspresnet50.ra_in1k/weight_meta.py create mode 100644 samples/timm/cspresnext50.ra_in1k/graph_hash.txt create mode 100644 samples/timm/cspresnext50.ra_in1k/graph_net.json create mode 100644 samples/timm/cspresnext50.ra_in1k/input_meta.py create mode 100644 samples/timm/cspresnext50.ra_in1k/input_tensor_constraints.py create mode 100644 samples/timm/cspresnext50.ra_in1k/model.py create mode 100644 samples/timm/cspresnext50.ra_in1k/weight_meta.py create mode 100644 samples/timm/darknet53.c2ns_in1k/graph_hash.txt create mode 100644 samples/timm/darknet53.c2ns_in1k/graph_net.json create mode 100644 samples/timm/darknet53.c2ns_in1k/input_meta.py create mode 100644 samples/timm/darknet53.c2ns_in1k/input_tensor_constraints.py create mode 100644 samples/timm/darknet53.c2ns_in1k/model.py create mode 100644 samples/timm/darknet53.c2ns_in1k/weight_meta.py create mode 100644 samples/timm/darknetaa53.c2ns_in1k/graph_hash.txt create mode 100644 samples/timm/darknetaa53.c2ns_in1k/graph_net.json create mode 100644 samples/timm/darknetaa53.c2ns_in1k/input_meta.py create mode 100644 samples/timm/darknetaa53.c2ns_in1k/input_tensor_constraints.py create mode 100644 samples/timm/darknetaa53.c2ns_in1k/model.py create mode 100644 samples/timm/darknetaa53.c2ns_in1k/weight_meta.py create mode 100644 samples/timm/deit3_base_patch16_224.fb_in1k/graph_hash.txt create mode 100644 samples/timm/deit3_base_patch16_224.fb_in1k/graph_net.json create mode 100644 samples/timm/deit3_base_patch16_224.fb_in1k/input_meta.py create mode 100644 samples/timm/deit3_base_patch16_224.fb_in1k/input_tensor_constraints.py create mode 100644 samples/timm/deit3_base_patch16_224.fb_in1k/model.py create mode 100644 samples/timm/deit3_base_patch16_224.fb_in1k/weight_meta.py create mode 100644 samples/timm/deit3_huge_patch14_224.fb_in1k/graph_hash.txt create mode 100644 samples/timm/deit3_huge_patch14_224.fb_in1k/graph_net.json create mode 100644 samples/timm/deit3_huge_patch14_224.fb_in1k/input_meta.py create mode 100644 samples/timm/deit3_huge_patch14_224.fb_in1k/input_tensor_constraints.py create mode 100644 samples/timm/deit3_huge_patch14_224.fb_in1k/model.py create mode 100644 samples/timm/deit3_huge_patch14_224.fb_in1k/weight_meta.py create mode 100644 samples/timm/deit3_large_patch16_224.fb_in1k/graph_hash.txt create mode 100644 samples/timm/deit3_large_patch16_224.fb_in1k/graph_net.json create mode 100644 samples/timm/deit3_large_patch16_224.fb_in1k/input_meta.py create mode 100644 samples/timm/deit3_large_patch16_224.fb_in1k/input_tensor_constraints.py create mode 100644 samples/timm/deit3_large_patch16_224.fb_in1k/model.py create mode 100644 samples/timm/deit3_large_patch16_224.fb_in1k/weight_meta.py create mode 100644 samples/timm/deit3_medium_patch16_224.fb_in1k/graph_hash.txt create mode 100644 samples/timm/deit3_medium_patch16_224.fb_in1k/graph_net.json create mode 100644 samples/timm/deit3_medium_patch16_224.fb_in1k/input_meta.py create mode 100644 samples/timm/deit3_medium_patch16_224.fb_in1k/input_tensor_constraints.py create mode 100644 samples/timm/deit3_medium_patch16_224.fb_in1k/model.py create mode 100644 samples/timm/deit3_medium_patch16_224.fb_in1k/weight_meta.py create mode 100644 samples/timm/deit3_small_patch16_224.fb_in1k/graph_hash.txt create mode 100644 samples/timm/deit3_small_patch16_224.fb_in1k/graph_net.json create mode 100644 samples/timm/deit3_small_patch16_224.fb_in1k/input_meta.py create mode 100644 samples/timm/deit3_small_patch16_224.fb_in1k/input_tensor_constraints.py create mode 100644 samples/timm/deit3_small_patch16_224.fb_in1k/model.py create mode 100644 samples/timm/deit3_small_patch16_224.fb_in1k/weight_meta.py create mode 100644 samples/timm/deit_base_distilled_patch16_224.fb_in1k/graph_hash.txt create mode 100644 samples/timm/deit_base_distilled_patch16_224.fb_in1k/graph_net.json create mode 100644 samples/timm/deit_base_distilled_patch16_224.fb_in1k/input_meta.py create mode 100644 samples/timm/deit_base_distilled_patch16_224.fb_in1k/input_tensor_constraints.py create mode 100644 samples/timm/deit_base_distilled_patch16_224.fb_in1k/model.py create mode 100644 samples/timm/deit_base_distilled_patch16_224.fb_in1k/weight_meta.py create mode 100644 samples/timm/deit_base_patch16_224.fb_in1k/graph_hash.txt create mode 100644 samples/timm/deit_base_patch16_224.fb_in1k/graph_net.json create mode 100644 samples/timm/deit_base_patch16_224.fb_in1k/input_meta.py create mode 100644 samples/timm/deit_base_patch16_224.fb_in1k/input_tensor_constraints.py create mode 100644 samples/timm/deit_base_patch16_224.fb_in1k/model.py create mode 100644 samples/timm/deit_base_patch16_224.fb_in1k/weight_meta.py create mode 100644 samples/timm/deit_small_distilled_patch16_224.fb_in1k/graph_hash.txt create mode 100644 samples/timm/deit_small_distilled_patch16_224.fb_in1k/graph_net.json create mode 100644 samples/timm/deit_small_distilled_patch16_224.fb_in1k/input_meta.py create mode 100644 samples/timm/deit_small_distilled_patch16_224.fb_in1k/input_tensor_constraints.py create mode 100644 samples/timm/deit_small_distilled_patch16_224.fb_in1k/model.py create mode 100644 samples/timm/deit_small_distilled_patch16_224.fb_in1k/weight_meta.py create mode 100644 samples/timm/deit_small_patch16_224.fb_in1k/graph_hash.txt create mode 100644 samples/timm/deit_small_patch16_224.fb_in1k/graph_net.json create mode 100644 samples/timm/deit_small_patch16_224.fb_in1k/input_meta.py create mode 100644 samples/timm/deit_small_patch16_224.fb_in1k/input_tensor_constraints.py create mode 100644 samples/timm/deit_small_patch16_224.fb_in1k/model.py create mode 100644 samples/timm/deit_small_patch16_224.fb_in1k/weight_meta.py create mode 100644 samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/graph_hash.txt create mode 100644 samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/graph_net.json create mode 100644 samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/input_meta.py create mode 100644 samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/input_tensor_constraints.py create mode 100644 samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/model.py create mode 100644 samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/weight_meta.py create mode 100644 samples/timm/deit_tiny_patch16_224.fb_in1k/graph_hash.txt create mode 100644 samples/timm/deit_tiny_patch16_224.fb_in1k/graph_net.json create mode 100644 samples/timm/deit_tiny_patch16_224.fb_in1k/input_meta.py create mode 100644 samples/timm/deit_tiny_patch16_224.fb_in1k/input_tensor_constraints.py create mode 100644 samples/timm/deit_tiny_patch16_224.fb_in1k/model.py create mode 100644 samples/timm/deit_tiny_patch16_224.fb_in1k/weight_meta.py create mode 100644 samples/timm/densenet121.ra_in1k/graph_hash.txt create mode 100644 samples/timm/densenet121.ra_in1k/graph_net.json create mode 100644 samples/timm/densenet121.ra_in1k/input_meta.py create mode 100644 samples/timm/densenet121.ra_in1k/input_tensor_constraints.py create mode 100644 samples/timm/densenet121.ra_in1k/model.py create mode 100644 samples/timm/densenet121.ra_in1k/weight_meta.py create mode 100644 samples/timm/densenet161.tv_in1k/graph_hash.txt create mode 100644 samples/timm/densenet161.tv_in1k/graph_net.json create mode 100644 samples/timm/densenet161.tv_in1k/input_meta.py create mode 100644 samples/timm/densenet161.tv_in1k/input_tensor_constraints.py create mode 100644 samples/timm/densenet161.tv_in1k/model.py create mode 100644 samples/timm/densenet161.tv_in1k/weight_meta.py create mode 100644 samples/timm/densenet169.tv_in1k/graph_hash.txt create mode 100644 samples/timm/densenet169.tv_in1k/graph_net.json create mode 100644 samples/timm/densenet169.tv_in1k/input_meta.py create mode 100644 samples/timm/densenet169.tv_in1k/input_tensor_constraints.py create mode 100644 samples/timm/densenet169.tv_in1k/model.py create mode 100644 samples/timm/densenet169.tv_in1k/weight_meta.py create mode 100644 samples/timm/densenet201.tv_in1k/graph_hash.txt create mode 100644 samples/timm/densenet201.tv_in1k/graph_net.json create mode 100644 samples/timm/densenet201.tv_in1k/input_meta.py create mode 100644 samples/timm/densenet201.tv_in1k/input_tensor_constraints.py create mode 100644 samples/timm/densenet201.tv_in1k/model.py create mode 100644 samples/timm/densenet201.tv_in1k/weight_meta.py create mode 100644 samples/timm/densenetblur121d.ra_in1k/graph_hash.txt create mode 100644 samples/timm/densenetblur121d.ra_in1k/graph_net.json create mode 100644 samples/timm/densenetblur121d.ra_in1k/input_meta.py create mode 100644 samples/timm/densenetblur121d.ra_in1k/input_tensor_constraints.py create mode 100644 samples/timm/densenetblur121d.ra_in1k/model.py create mode 100644 samples/timm/densenetblur121d.ra_in1k/weight_meta.py create mode 100644 samples/timm/dla34.in1k/graph_hash.txt create mode 100644 samples/timm/dla34.in1k/graph_net.json create mode 100644 samples/timm/dla34.in1k/input_meta.py create mode 100644 samples/timm/dla34.in1k/input_tensor_constraints.py create mode 100644 samples/timm/dla34.in1k/model.py create mode 100644 samples/timm/dla34.in1k/weight_meta.py diff --git a/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/graph_hash.txt b/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/graph_hash.txt new file mode 100644 index 0000000000..7d4a1bd7d4 --- /dev/null +++ b/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/graph_hash.txt @@ -0,0 +1 @@ +c1ffff8904452cfbdcfa2438656350764e533fe4ad59f7b8859706fb27985867 \ No newline at end of file diff --git a/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/graph_net.json b/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/input_meta.py b/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/input_tensor_constraints.py b/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/model.py b/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/model.py new file mode 100644 index 0000000000..5502ab1517 --- /dev/null +++ b/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/model.py @@ -0,0 +1,2725 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_pre_logits_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_pre_logits_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_pre_logits_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_pre_logits_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_pre_logits_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_pre_logits_modules_fc_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (192,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (192,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_9 = torch._C._nn.linear( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = x_10.permute(0, 3, 1, 2) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (192,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = torch._C._nn.linear( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_19 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = x_21.permute(0, 3, 1, 2) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (192,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_31 = torch._C._nn.linear( + x_30, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_30 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = x_32.permute(0, 3, 1, 2) + x_32 = None + reshape_2 = l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_34 = x_33.mul(reshape_2) + x_33 = reshape_2 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (192,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (384,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_45 = torch._C._nn.linear( + x_44, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_44 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_46 = torch.nn.functional.dropout(x_45, 0.0, False, False) + x_45 = None + x_47 = x_46.permute(0, 3, 1, 2) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (384,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = torch._C._nn.linear( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_55 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_57 = torch.nn.functional.dropout(x_56, 0.0, False, False) + x_56 = None + x_58 = x_57.permute(0, 3, 1, 2) + x_57 = None + reshape_4 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_59 = x_58.mul(reshape_4) + x_58 = reshape_4 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (384,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + reshape_5 = l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_70 = x_69.mul(reshape_5) + x_69 = reshape_5 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (384,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = torch._C._nn.linear( + x_80, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_80 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_82 = torch.nn.functional.dropout(x_81, 0.0, False, False) + x_81 = None + x_83 = x_82.permute(0, 3, 1, 2) + x_82 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_84 = x_83.mul(reshape_6) + x_83 = reshape_6 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = torch._C._nn.linear( + x_91, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_91 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_93 = torch.nn.functional.dropout(x_92, 0.0, False, False) + x_92 = None + x_94 = x_93.permute(0, 3, 1, 2) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_104 = torch.nn.functional.dropout(x_103, 0.0, False, False) + x_103 = None + x_105 = x_104.permute(0, 3, 1, 2) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = torch._C._nn.linear( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_113 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_115 = torch.nn.functional.dropout(x_114, 0.0, False, False) + x_114 = None + x_116 = x_115.permute(0, 3, 1, 2) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_124 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_126.permute(0, 3, 1, 2) + x_126 = None + reshape_10 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_128 = x_127.mul(reshape_10) + x_127 = reshape_10 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = torch._C._nn.linear( + x_135, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_135 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_137 = torch.nn.functional.dropout(x_136, 0.0, False, False) + x_136 = None + x_138 = x_137.permute(0, 3, 1, 2) + x_137 = None + reshape_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_139 = x_138.mul(reshape_11) + x_138 = reshape_11 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = torch._C._nn.linear( + x_146, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_146 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_148 = torch.nn.functional.dropout(x_147, 0.0, False, False) + x_147 = None + x_149 = x_148.permute(0, 3, 1, 2) + x_148 = None + reshape_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + None + ) + x_150 = x_149.mul(reshape_12) + x_149 = reshape_12 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_159.permute(0, 3, 1, 2) + x_159 = None + reshape_13 = l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + None + ) + x_161 = x_160.mul(reshape_13) + x_160 = reshape_13 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_169 = torch._C._nn.linear( + x_168, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_168 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_170 = torch.nn.functional.dropout(x_169, 0.0, False, False) + x_169 = None + x_171 = x_170.permute(0, 3, 1, 2) + x_170 = None + reshape_14 = l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + None + ) + x_172 = x_171.mul(reshape_14) + x_171 = reshape_14 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = (None) + x_175 = x_174.permute(0, 2, 3, 1) + x_174 = None + x_176 = torch.nn.functional.layer_norm( + x_175, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_, + 1e-06, + ) + x_175 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = (None) + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_176 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_178 = torch._C._nn.gelu(x_177) + x_177 = None + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_180 = torch._C._nn.linear( + x_179, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_179 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_181 = torch.nn.functional.dropout(x_180, 0.0, False, False) + x_180 = None + x_182 = x_181.permute(0, 3, 1, 2) + x_181 = None + reshape_15 = l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ = ( + None + ) + x_183 = x_182.mul(reshape_15) + x_182 = reshape_15 = None + x_184 = x_183 + x_173 + x_183 = x_173 = None + x_185 = torch.conv2d( + x_184, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = (None) + x_186 = x_185.permute(0, 2, 3, 1) + x_185 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = (None) + x_188 = torch._C._nn.linear( + x_187, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_187 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_189 = torch._C._nn.gelu(x_188) + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = torch._C._nn.linear( + x_190, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_190 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_192 = torch.nn.functional.dropout(x_191, 0.0, False, False) + x_191 = None + x_193 = x_192.permute(0, 3, 1, 2) + x_192 = None + reshape_16 = l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ = ( + None + ) + x_194 = x_193.mul(reshape_16) + x_193 = reshape_16 = None + x_195 = x_194 + x_184 + x_194 = x_184 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = (None) + x_197 = x_196.permute(0, 2, 3, 1) + x_196 = None + x_198 = torch.nn.functional.layer_norm( + x_197, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_, + 1e-06, + ) + x_197 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = (None) + x_199 = torch._C._nn.linear( + x_198, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_198 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_200 = torch._C._nn.gelu(x_199) + x_199 = None + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_202 = torch._C._nn.linear( + x_201, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_201 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_203 = torch.nn.functional.dropout(x_202, 0.0, False, False) + x_202 = None + x_204 = x_203.permute(0, 3, 1, 2) + x_203 = None + reshape_17 = l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ = ( + None + ) + x_205 = x_204.mul(reshape_17) + x_204 = reshape_17 = None + x_206 = x_205 + x_195 + x_205 = x_195 = None + x_207 = torch.conv2d( + x_206, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = (None) + x_208 = x_207.permute(0, 2, 3, 1) + x_207 = None + x_209 = torch.nn.functional.layer_norm( + x_208, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_, + 1e-06, + ) + x_208 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = (None) + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_209 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_211 = torch._C._nn.gelu(x_210) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_213 = torch._C._nn.linear( + x_212, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_212 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_214 = torch.nn.functional.dropout(x_213, 0.0, False, False) + x_213 = None + x_215 = x_214.permute(0, 3, 1, 2) + x_214 = None + reshape_18 = l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ = ( + None + ) + x_216 = x_215.mul(reshape_18) + x_215 = reshape_18 = None + x_217 = x_216 + x_206 + x_216 = x_206 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = (None) + x_219 = x_218.permute(0, 2, 3, 1) + x_218 = None + x_220 = torch.nn.functional.layer_norm( + x_219, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_, + 1e-06, + ) + x_219 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = (None) + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_220 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_222 = torch._C._nn.gelu(x_221) + x_221 = None + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_224 = torch._C._nn.linear( + x_223, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_223 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = x_225.permute(0, 3, 1, 2) + x_225 = None + reshape_19 = l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ = ( + None + ) + x_227 = x_226.mul(reshape_19) + x_226 = reshape_19 = None + x_228 = x_227 + x_217 + x_227 = x_217 = None + x_229 = torch.conv2d( + x_228, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = (None) + x_230 = x_229.permute(0, 2, 3, 1) + x_229 = None + x_231 = torch.nn.functional.layer_norm( + x_230, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_, + 1e-06, + ) + x_230 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = (None) + x_232 = torch._C._nn.linear( + x_231, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_231 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_233 = torch._C._nn.gelu(x_232) + x_232 = None + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_235 = torch._C._nn.linear( + x_234, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_234 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_236 = torch.nn.functional.dropout(x_235, 0.0, False, False) + x_235 = None + x_237 = x_236.permute(0, 3, 1, 2) + x_236 = None + reshape_20 = l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ = ( + None + ) + x_238 = x_237.mul(reshape_20) + x_237 = reshape_20 = None + x_239 = x_238 + x_228 + x_238 = x_228 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = (None) + x_241 = x_240.permute(0, 2, 3, 1) + x_240 = None + x_242 = torch.nn.functional.layer_norm( + x_241, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_, + 1e-06, + ) + x_241 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = (None) + x_243 = torch._C._nn.linear( + x_242, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_242 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_244 = torch._C._nn.gelu(x_243) + x_243 = None + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_246 = torch._C._nn.linear( + x_245, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_245 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_247 = torch.nn.functional.dropout(x_246, 0.0, False, False) + x_246 = None + x_248 = x_247.permute(0, 3, 1, 2) + x_247 = None + reshape_21 = l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ = ( + None + ) + x_249 = x_248.mul(reshape_21) + x_248 = reshape_21 = None + x_250 = x_249 + x_239 + x_249 = x_239 = None + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = (None) + x_252 = x_251.permute(0, 2, 3, 1) + x_251 = None + x_253 = torch.nn.functional.layer_norm( + x_252, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_, + 1e-06, + ) + x_252 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = (None) + x_254 = torch._C._nn.linear( + x_253, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_253 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_255 = torch._C._nn.gelu(x_254) + x_254 = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_257 = torch._C._nn.linear( + x_256, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_256 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_258 = torch.nn.functional.dropout(x_257, 0.0, False, False) + x_257 = None + x_259 = x_258.permute(0, 3, 1, 2) + x_258 = None + reshape_22 = l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ = ( + None + ) + x_260 = x_259.mul(reshape_22) + x_259 = reshape_22 = None + x_261 = x_260 + x_250 + x_260 = x_250 = None + x_262 = torch.conv2d( + x_261, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = (None) + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = (None) + x_265 = torch._C._nn.linear( + x_264, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_264 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_266 = torch._C._nn.gelu(x_265) + x_265 = None + x_267 = torch.nn.functional.dropout(x_266, 0.0, False, False) + x_266 = None + x_268 = torch._C._nn.linear( + x_267, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_269 = torch.nn.functional.dropout(x_268, 0.0, False, False) + x_268 = None + x_270 = x_269.permute(0, 3, 1, 2) + x_269 = None + reshape_23 = l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ = ( + None + ) + x_271 = x_270.mul(reshape_23) + x_270 = reshape_23 = None + x_272 = x_271 + x_261 + x_271 = x_261 = None + x_273 = torch.conv2d( + x_272, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = (None) + x_274 = x_273.permute(0, 2, 3, 1) + x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_, + 1e-06, + ) + x_274 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = (None) + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_277 = torch._C._nn.gelu(x_276) + x_276 = None + x_278 = torch.nn.functional.dropout(x_277, 0.0, False, False) + x_277 = None + x_279 = torch._C._nn.linear( + x_278, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_280 = torch.nn.functional.dropout(x_279, 0.0, False, False) + x_279 = None + x_281 = x_280.permute(0, 3, 1, 2) + x_280 = None + reshape_24 = l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ = ( + None + ) + x_282 = x_281.mul(reshape_24) + x_281 = reshape_24 = None + x_283 = x_282 + x_272 + x_282 = x_272 = None + x_284 = torch.conv2d( + x_283, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = (None) + x_285 = x_284.permute(0, 2, 3, 1) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_, + 1e-06, + ) + x_285 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = (None) + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_288 = torch._C._nn.gelu(x_287) + x_287 = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + x_290 = torch._C._nn.linear( + x_289, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_291 = torch.nn.functional.dropout(x_290, 0.0, False, False) + x_290 = None + x_292 = x_291.permute(0, 3, 1, 2) + x_291 = None + reshape_25 = l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ = ( + None + ) + x_293 = x_292.mul(reshape_25) + x_292 = reshape_25 = None + x_294 = x_293 + x_283 + x_293 = x_283 = None + x_295 = torch.conv2d( + x_294, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = (None) + x_296 = x_295.permute(0, 2, 3, 1) + x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_, + 1e-06, + ) + x_296 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = (None) + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_299 = torch._C._nn.gelu(x_298) + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_301 = torch._C._nn.linear( + x_300, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_302 = torch.nn.functional.dropout(x_301, 0.0, False, False) + x_301 = None + x_303 = x_302.permute(0, 3, 1, 2) + x_302 = None + reshape_26 = l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ = ( + None + ) + x_304 = x_303.mul(reshape_26) + x_303 = reshape_26 = None + x_305 = x_304 + x_294 + x_304 = x_294 = None + x_306 = torch.conv2d( + x_305, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = (None) + x_307 = x_306.permute(0, 2, 3, 1) + x_306 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_, + 1e-06, + ) + x_307 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = (None) + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_310 = torch._C._nn.gelu(x_309) + x_309 = None + x_311 = torch.nn.functional.dropout(x_310, 0.0, False, False) + x_310 = None + x_312 = torch._C._nn.linear( + x_311, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_313 = torch.nn.functional.dropout(x_312, 0.0, False, False) + x_312 = None + x_314 = x_313.permute(0, 3, 1, 2) + x_313 = None + reshape_27 = l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ = ( + None + ) + x_315 = x_314.mul(reshape_27) + x_314 = reshape_27 = None + x_316 = x_315 + x_305 + x_315 = x_305 = None + x_317 = torch.conv2d( + x_316, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = (None) + x_318 = x_317.permute(0, 2, 3, 1) + x_317 = None + x_319 = torch.nn.functional.layer_norm( + x_318, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_, + 1e-06, + ) + x_318 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = (None) + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_321 = torch._C._nn.gelu(x_320) + x_320 = None + x_322 = torch.nn.functional.dropout(x_321, 0.0, False, False) + x_321 = None + x_323 = torch._C._nn.linear( + x_322, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_324 = torch.nn.functional.dropout(x_323, 0.0, False, False) + x_323 = None + x_325 = x_324.permute(0, 3, 1, 2) + x_324 = None + reshape_28 = l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ = ( + None + ) + x_326 = x_325.mul(reshape_28) + x_325 = reshape_28 = None + x_327 = x_326 + x_316 + x_326 = x_316 = None + x_328 = torch.conv2d( + x_327, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = (None) + x_329 = x_328.permute(0, 2, 3, 1) + x_328 = None + x_330 = torch.nn.functional.layer_norm( + x_329, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_, + 1e-06, + ) + x_329 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = (None) + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_332 = torch._C._nn.gelu(x_331) + x_331 = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_334 = torch._C._nn.linear( + x_333, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_335 = torch.nn.functional.dropout(x_334, 0.0, False, False) + x_334 = None + x_336 = x_335.permute(0, 3, 1, 2) + x_335 = None + reshape_29 = l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ = ( + None + ) + x_337 = x_336.mul(reshape_29) + x_336 = reshape_29 = None + x_338 = x_337 + x_327 + x_337 = x_327 = None + x_339 = torch.conv2d( + x_338, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = (None) + x_340 = x_339.permute(0, 2, 3, 1) + x_339 = None + x_341 = torch.nn.functional.layer_norm( + x_340, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_, + 1e-06, + ) + x_340 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = (None) + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_343 = torch._C._nn.gelu(x_342) + x_342 = None + x_344 = torch.nn.functional.dropout(x_343, 0.0, False, False) + x_343 = None + x_345 = torch._C._nn.linear( + x_344, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_346 = torch.nn.functional.dropout(x_345, 0.0, False, False) + x_345 = None + x_347 = x_346.permute(0, 3, 1, 2) + x_346 = None + reshape_30 = l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ = ( + None + ) + x_348 = x_347.mul(reshape_30) + x_347 = reshape_30 = None + x_349 = x_348 + x_338 + x_348 = x_338 = None + x_350 = torch.conv2d( + x_349, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = (None) + x_351 = x_350.permute(0, 2, 3, 1) + x_350 = None + x_352 = torch.nn.functional.layer_norm( + x_351, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_, + 1e-06, + ) + x_351 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = (None) + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_354 = torch._C._nn.gelu(x_353) + x_353 = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_356 = torch._C._nn.linear( + x_355, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_357 = torch.nn.functional.dropout(x_356, 0.0, False, False) + x_356 = None + x_358 = x_357.permute(0, 3, 1, 2) + x_357 = None + reshape_31 = l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ = ( + None + ) + x_359 = x_358.mul(reshape_31) + x_358 = reshape_31 = None + x_360 = x_359 + x_349 + x_359 = x_349 = None + x_361 = torch.conv2d( + x_360, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = (None) + x_362 = x_361.permute(0, 2, 3, 1) + x_361 = None + x_363 = torch.nn.functional.layer_norm( + x_362, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_, + 1e-06, + ) + x_362 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = (None) + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_365 = torch._C._nn.gelu(x_364) + x_364 = None + x_366 = torch.nn.functional.dropout(x_365, 0.0, False, False) + x_365 = None + x_367 = torch._C._nn.linear( + x_366, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_368 = torch.nn.functional.dropout(x_367, 0.0, False, False) + x_367 = None + x_369 = x_368.permute(0, 3, 1, 2) + x_368 = None + reshape_32 = l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ = ( + None + ) + x_370 = x_369.mul(reshape_32) + x_369 = reshape_32 = None + x_371 = x_370 + x_360 + x_370 = x_360 = None + x_372 = x_371.permute(0, 2, 3, 1) + x_371 = None + x_373 = torch.nn.functional.layer_norm( + x_372, + (768,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_372 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_374 = x_373.permute(0, 3, 1, 2) + x_373 = None + input_4 = torch.conv2d( + x_374, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_374 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_375 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_376 = x_375.permute(0, 2, 3, 1) + x_375 = None + x_377 = torch.nn.functional.layer_norm( + x_376, + (1536,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_376 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_378 = torch._C._nn.linear( + x_377, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_377 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_379 = torch._C._nn.gelu(x_378) + x_378 = None + x_380 = torch.nn.functional.dropout(x_379, 0.0, False, False) + x_379 = None + x_381 = torch._C._nn.linear( + x_380, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_380 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_382 = torch.nn.functional.dropout(x_381, 0.0, False, False) + x_381 = None + x_383 = x_382.permute(0, 3, 1, 2) + x_382 = None + reshape_33 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_384 = x_383.mul(reshape_33) + x_383 = reshape_33 = None + x_385 = x_384 + input_4 + x_384 = input_4 = None + x_386 = torch.conv2d( + x_385, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_387 = x_386.permute(0, 2, 3, 1) + x_386 = None + x_388 = torch.nn.functional.layer_norm( + x_387, + (1536,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_387 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_389 = torch._C._nn.linear( + x_388, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_388 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_390 = torch._C._nn.gelu(x_389) + x_389 = None + x_391 = torch.nn.functional.dropout(x_390, 0.0, False, False) + x_390 = None + x_392 = torch._C._nn.linear( + x_391, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_391 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_393 = torch.nn.functional.dropout(x_392, 0.0, False, False) + x_392 = None + x_394 = x_393.permute(0, 3, 1, 2) + x_393 = None + reshape_34 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_395 = x_394.mul(reshape_34) + x_394 = reshape_34 = None + x_396 = x_395 + x_385 + x_395 = x_385 = None + x_397 = torch.conv2d( + x_396, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_398 = x_397.permute(0, 2, 3, 1) + x_397 = None + x_399 = torch.nn.functional.layer_norm( + x_398, + (1536,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_398 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_400 = torch._C._nn.linear( + x_399, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_399 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_401 = torch._C._nn.gelu(x_400) + x_400 = None + x_402 = torch.nn.functional.dropout(x_401, 0.0, False, False) + x_401 = None + x_403 = torch._C._nn.linear( + x_402, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_402 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_404 = torch.nn.functional.dropout(x_403, 0.0, False, False) + x_403 = None + x_405 = x_404.permute(0, 3, 1, 2) + x_404 = None + reshape_35 = l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_406 = x_405.mul(reshape_35) + x_405 = reshape_35 = None + x_407 = x_406 + x_396 + x_406 = x_396 = None + x_408 = torch.nn.functional.adaptive_avg_pool2d(x_407, 1) + x_407 = None + x_409 = x_408.permute(0, 2, 3, 1) + x_408 = None + x_410 = torch.nn.functional.layer_norm( + x_409, + (1536,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_409 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_411 = x_410.permute(0, 3, 1, 2) + x_410 = None + x_412 = x_411.flatten(1, -1) + x_411 = None + input_5 = torch._C._nn.linear( + x_412, + l_self_modules_head_modules_pre_logits_modules_fc_parameters_weight_, + l_self_modules_head_modules_pre_logits_modules_fc_parameters_bias_, + ) + x_412 = ( + l_self_modules_head_modules_pre_logits_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_pre_logits_modules_fc_parameters_bias_ = None + input_6 = torch._C._nn.gelu(input_5) + input_5 = None + x_413 = torch.nn.functional.dropout(input_6, 0.0, False, False) + input_6 = None + x_414 = torch._C._nn.linear( + x_413, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_413 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_414,) diff --git a/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/weight_meta.py b/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/weight_meta.py new file mode 100644 index 0000000000..207e9350ee --- /dev/null +++ b/samples/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k/weight_meta.py @@ -0,0 +1,3490 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [192, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.224 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [384, 192, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [768, 384, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [1536, 768, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_pre_logits_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_pre_logits_modules_fc_parameters_weight_" + shape = [1536, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_pre_logits_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_pre_logits_modules_fc_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_nano.d1h_in1k/graph_hash.txt b/samples/timm/convnext_nano.d1h_in1k/graph_hash.txt new file mode 100644 index 0000000000..0058e5ef8a --- /dev/null +++ b/samples/timm/convnext_nano.d1h_in1k/graph_hash.txt @@ -0,0 +1 @@ +4b84b13cf6edfc2284f331776426f1742970193e1d202ffc898a19f4ec8fc5a9 \ No newline at end of file diff --git a/samples/timm/convnext_nano.d1h_in1k/graph_net.json b/samples/timm/convnext_nano.d1h_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnext_nano.d1h_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_nano.d1h_in1k/input_meta.py b/samples/timm/convnext_nano.d1h_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_nano.d1h_in1k/input_tensor_constraints.py b/samples/timm/convnext_nano.d1h_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_nano.d1h_in1k/model.py b/samples/timm/convnext_nano.d1h_in1k/model.py new file mode 100644 index 0000000000..33e949b77a --- /dev/null +++ b/samples/timm/convnext_nano.d1h_in1k/model.py @@ -0,0 +1,1279 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (80,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (80,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (80,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_21 = torch.conv2d( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (80,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_2 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (160,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + reshape_2 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_37 = x_36.mul(reshape_2) + x_36 = reshape_2 = None + x_38 = x_37 + input_2 + x_37 = input_2 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (160,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = torch.conv2d( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (160,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_3 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_60 = torch.conv2d( + x_59, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_59 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + reshape_4 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_62 = x_61.mul(reshape_4) + x_61 = reshape_4 = None + x_63 = x_62 + input_3 + x_62 = input_3 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_71 = torch.conv2d( + x_70, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_70 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + reshape_5 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_73 = x_72.mul(reshape_5) + x_72 = reshape_5 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_84 = x_83.mul(reshape_6) + x_83 = reshape_6 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_115 = torch.conv2d( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_122 = x_121.permute(0, 3, 1, 2) + x_121 = None + x_123 = torch.conv2d( + x_122, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_124 = torch._C._nn.gelu(x_123) + x_123 = None + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + reshape_10 = l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + None + ) + x_128 = x_127.mul(reshape_10) + x_127 = reshape_10 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_133 = x_132.permute(0, 3, 1, 2) + x_132 = None + x_134 = torch.conv2d( + x_133, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_133 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_135 = torch._C._nn.gelu(x_134) + x_134 = None + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + reshape_11 = l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + None + ) + x_139 = x_138.mul(reshape_11) + x_138 = reshape_11 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = x_140.permute(0, 2, 3, 1) + x_140 = None + x_142 = torch.nn.functional.layer_norm( + x_141, + (320,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_141 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_143 = x_142.permute(0, 3, 1, 2) + x_142 = None + input_4 = torch.conv2d( + x_143, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_143 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_144 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 640, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (640,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = torch.conv2d( + x_147, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_147 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_149 = torch._C._nn.gelu(x_148) + x_148 = None + x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) + x_149 = None + x_151 = torch.conv2d( + x_150, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_150 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + reshape_12 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_153 = x_152.mul(reshape_12) + x_152 = reshape_12 = None + x_154 = x_153 + input_4 + x_153 = input_4 = None + x_155 = torch.conv2d( + x_154, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 640, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_156 = x_155.permute(0, 2, 3, 1) + x_155 = None + x_157 = torch.nn.functional.layer_norm( + x_156, + (640,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_156 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_158 = x_157.permute(0, 3, 1, 2) + x_157 = None + x_159 = torch.conv2d( + x_158, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_158 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_160 = torch._C._nn.gelu(x_159) + x_159 = None + x_161 = torch.nn.functional.dropout(x_160, 0.0, False, False) + x_160 = None + x_162 = torch.conv2d( + x_161, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_161 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_163 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + reshape_13 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_164 = x_163.mul(reshape_13) + x_163 = reshape_13 = None + x_165 = x_164 + x_154 + x_164 = x_154 = None + x_166 = torch.nn.functional.adaptive_avg_pool2d(x_165, 1) + x_165 = None + x_167 = x_166.permute(0, 2, 3, 1) + x_166 = None + x_168 = torch.nn.functional.layer_norm( + x_167, + (640,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_167 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_169 = x_168.permute(0, 3, 1, 2) + x_168 = None + x_170 = x_169.flatten(1, -1) + x_169 = None + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = torch._C._nn.linear( + x_171, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_171 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_172,) diff --git a/samples/timm/convnext_nano.d1h_in1k/weight_meta.py b/samples/timm/convnext_nano.d1h_in1k/weight_meta.py new file mode 100644 index 0000000000..3985ec06fc --- /dev/null +++ b/samples/timm/convnext_nano.d1h_in1k/weight_meta.py @@ -0,0 +1,1490 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [80, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.228 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [160, 80, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [320, 160, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [640, 320, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [640, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2560, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [640, 2560, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [640, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2560, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [640, 2560, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_nano_ols.d1h_in1k/graph_hash.txt b/samples/timm/convnext_nano_ols.d1h_in1k/graph_hash.txt new file mode 100644 index 0000000000..5d1479c7d2 --- /dev/null +++ b/samples/timm/convnext_nano_ols.d1h_in1k/graph_hash.txt @@ -0,0 +1 @@ +4b04e033486ab9dcce14edbf03a9da60666faabdba55185f29cb9b0e5e8781d5 \ No newline at end of file diff --git a/samples/timm/convnext_nano_ols.d1h_in1k/graph_net.json b/samples/timm/convnext_nano_ols.d1h_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnext_nano_ols.d1h_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_nano_ols.d1h_in1k/input_meta.py b/samples/timm/convnext_nano_ols.d1h_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_nano_ols.d1h_in1k/input_tensor_constraints.py b/samples/timm/convnext_nano_ols.d1h_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_nano_ols.d1h_in1k/model.py b/samples/timm/convnext_nano_ols.d1h_in1k/model.py new file mode 100644 index 0000000000..2a035e4cc0 --- /dev/null +++ b/samples/timm/convnext_nano_ols.d1h_in1k/model.py @@ -0,0 +1,1299 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stem_modules_2_parameters_weight_ = ( + L_self_modules_stem_modules_2_parameters_weight_ + ) + l_self_modules_stem_modules_2_parameters_bias_ = ( + L_self_modules_stem_modules_2_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + input_2 = torch.conv2d( + input_1, + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_1 = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x = input_2.permute(0, 2, 3, 1) + input_2 = None + x_1 = torch.nn.functional.layer_norm( + x, + (80,), + l_self_modules_stem_modules_2_parameters_weight_, + l_self_modules_stem_modules_2_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_2_parameters_weight_ + ) = l_self_modules_stem_modules_2_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (80,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (80,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_21 = torch.conv2d( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (80,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_3 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_3, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (160,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + reshape_2 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_37 = x_36.mul(reshape_2) + x_36 = reshape_2 = None + x_38 = x_37 + input_3 + x_37 = input_3 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (160,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = torch.conv2d( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (160,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_4 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_4, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_60 = torch.conv2d( + x_59, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_59 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + reshape_4 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_62 = x_61.mul(reshape_4) + x_61 = reshape_4 = None + x_63 = x_62 + input_4 + x_62 = input_4 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_71 = torch.conv2d( + x_70, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_70 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + reshape_5 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_73 = x_72.mul(reshape_5) + x_72 = reshape_5 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_84 = x_83.mul(reshape_6) + x_83 = reshape_6 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_115 = torch.conv2d( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_122 = x_121.permute(0, 3, 1, 2) + x_121 = None + x_123 = torch.conv2d( + x_122, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_124 = torch._C._nn.gelu(x_123) + x_123 = None + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + reshape_10 = l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + None + ) + x_128 = x_127.mul(reshape_10) + x_127 = reshape_10 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_133 = x_132.permute(0, 3, 1, 2) + x_132 = None + x_134 = torch.conv2d( + x_133, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_133 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_135 = torch._C._nn.gelu(x_134) + x_134 = None + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + reshape_11 = l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + None + ) + x_139 = x_138.mul(reshape_11) + x_138 = reshape_11 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = x_140.permute(0, 2, 3, 1) + x_140 = None + x_142 = torch.nn.functional.layer_norm( + x_141, + (320,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_141 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_143 = x_142.permute(0, 3, 1, 2) + x_142 = None + input_5 = torch.conv2d( + x_143, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_143 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_144 = torch.conv2d( + input_5, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 640, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (640,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = torch.conv2d( + x_147, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_147 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_149 = torch._C._nn.gelu(x_148) + x_148 = None + x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) + x_149 = None + x_151 = torch.conv2d( + x_150, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_150 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + reshape_12 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_153 = x_152.mul(reshape_12) + x_152 = reshape_12 = None + x_154 = x_153 + input_5 + x_153 = input_5 = None + x_155 = torch.conv2d( + x_154, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 640, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_156 = x_155.permute(0, 2, 3, 1) + x_155 = None + x_157 = torch.nn.functional.layer_norm( + x_156, + (640,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_156 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_158 = x_157.permute(0, 3, 1, 2) + x_157 = None + x_159 = torch.conv2d( + x_158, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_158 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_160 = torch._C._nn.gelu(x_159) + x_159 = None + x_161 = torch.nn.functional.dropout(x_160, 0.0, False, False) + x_160 = None + x_162 = torch.conv2d( + x_161, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_161 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_163 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + reshape_13 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_164 = x_163.mul(reshape_13) + x_163 = reshape_13 = None + x_165 = x_164 + x_154 + x_164 = x_154 = None + x_166 = torch.nn.functional.adaptive_avg_pool2d(x_165, 1) + x_165 = None + x_167 = x_166.permute(0, 2, 3, 1) + x_166 = None + x_168 = torch.nn.functional.layer_norm( + x_167, + (640,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_167 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_169 = x_168.permute(0, 3, 1, 2) + x_168 = None + x_170 = x_169.flatten(1, -1) + x_169 = None + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = torch._C._nn.linear( + x_171, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_171 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_172,) diff --git a/samples/timm/convnext_nano_ols.d1h_in1k/weight_meta.py b/samples/timm/convnext_nano_ols.d1h_in1k/weight_meta.py new file mode 100644 index 0000000000..67e0136621 --- /dev/null +++ b/samples/timm/convnext_nano_ols.d1h_in1k/weight_meta.py @@ -0,0 +1,1510 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [80, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.230 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [80, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_parameters_weight_: + name = "L_self_modules_stem_modules_2_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_parameters_bias_: + name = "L_self_modules_stem_modules_2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [160, 80, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [320, 160, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [640, 320, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [640, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2560, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [640, 2560, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [640, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2560, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [640, 2560, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 640] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_pico.d1_in1k/graph_hash.txt b/samples/timm/convnext_pico.d1_in1k/graph_hash.txt new file mode 100644 index 0000000000..3a8ac56d97 --- /dev/null +++ b/samples/timm/convnext_pico.d1_in1k/graph_hash.txt @@ -0,0 +1 @@ +d4f4098bb40e1e23a759cde6faf233d3399e1878b3baa0fd14097af0c6316598 \ No newline at end of file diff --git a/samples/timm/convnext_pico.d1_in1k/graph_net.json b/samples/timm/convnext_pico.d1_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnext_pico.d1_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_pico.d1_in1k/input_meta.py b/samples/timm/convnext_pico.d1_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_pico.d1_in1k/input_tensor_constraints.py b/samples/timm/convnext_pico.d1_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_pico.d1_in1k/model.py b/samples/timm/convnext_pico.d1_in1k/model.py new file mode 100644 index 0000000000..11dd6cdd9f --- /dev/null +++ b/samples/timm/convnext_pico.d1_in1k/model.py @@ -0,0 +1,1123 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (64,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 64, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 64, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_21 = torch.conv2d( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (64,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_2 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + reshape_2 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_37 = x_36.mul(reshape_2) + x_36 = reshape_2 = None + x_38 = x_37 + input_2 + x_37 = input_2 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = torch.conv2d( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (128,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_3 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_60 = torch.conv2d( + x_59, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_59 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + reshape_4 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_62 = x_61.mul(reshape_4) + x_61 = reshape_4 = None + x_63 = x_62 + input_3 + x_62 = input_3 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_71 = torch.conv2d( + x_70, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_70 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + reshape_5 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_73 = x_72.mul(reshape_5) + x_72 = reshape_5 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_84 = x_83.mul(reshape_6) + x_83 = reshape_6 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_115 = torch.conv2d( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = x_118.permute(0, 2, 3, 1) + x_118 = None + x_120 = torch.nn.functional.layer_norm( + x_119, + (256,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_119 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_121 = x_120.permute(0, 3, 1, 2) + x_120 = None + input_4 = torch.conv2d( + x_121, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_122 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_123 = x_122.permute(0, 2, 3, 1) + x_122 = None + x_124 = torch.nn.functional.layer_norm( + x_123, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_125 = x_124.permute(0, 3, 1, 2) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126) + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_129 = torch.conv2d( + x_128, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_128 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) + x_129 = None + reshape_10 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_131 = x_130.mul(reshape_10) + x_130 = reshape_10 = None + x_132 = x_131 + input_4 + x_131 = input_4 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_134 = x_133.permute(0, 2, 3, 1) + x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_134 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_136 = x_135.permute(0, 3, 1, 2) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137) + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = torch.conv2d( + x_139, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_139 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_141 = torch.nn.functional.dropout(x_140, 0.0, False, False) + x_140 = None + reshape_11 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_142 = x_141.mul(reshape_11) + x_141 = reshape_11 = None + x_143 = x_142 + x_132 + x_142 = x_132 = None + x_144 = torch.nn.functional.adaptive_avg_pool2d(x_143, 1) + x_143 = None + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (512,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = x_147.flatten(1, -1) + x_147 = None + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = torch._C._nn.linear( + x_149, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_149 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_150,) diff --git a/samples/timm/convnext_pico.d1_in1k/weight_meta.py b/samples/timm/convnext_pico.d1_in1k/weight_meta.py new file mode 100644 index 0000000000..5449045a4e --- /dev/null +++ b/samples/timm/convnext_pico.d1_in1k/weight_meta.py @@ -0,0 +1,1310 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [64, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.225 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [64, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [64, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [128, 64, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [256, 128, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [512, 256, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_pico_ols.d1_in1k/graph_hash.txt b/samples/timm/convnext_pico_ols.d1_in1k/graph_hash.txt new file mode 100644 index 0000000000..a3afbfa7c0 --- /dev/null +++ b/samples/timm/convnext_pico_ols.d1_in1k/graph_hash.txt @@ -0,0 +1 @@ +2c0f4ae67241056d833f6493124e5a0295849c51c5c9cf26e5a4d988a38b6b3a \ No newline at end of file diff --git a/samples/timm/convnext_pico_ols.d1_in1k/graph_net.json b/samples/timm/convnext_pico_ols.d1_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnext_pico_ols.d1_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_pico_ols.d1_in1k/input_meta.py b/samples/timm/convnext_pico_ols.d1_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_pico_ols.d1_in1k/input_tensor_constraints.py b/samples/timm/convnext_pico_ols.d1_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_pico_ols.d1_in1k/model.py b/samples/timm/convnext_pico_ols.d1_in1k/model.py new file mode 100644 index 0000000000..47c87c9f24 --- /dev/null +++ b/samples/timm/convnext_pico_ols.d1_in1k/model.py @@ -0,0 +1,1143 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stem_modules_2_parameters_weight_ = ( + L_self_modules_stem_modules_2_parameters_weight_ + ) + l_self_modules_stem_modules_2_parameters_bias_ = ( + L_self_modules_stem_modules_2_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + input_2 = torch.conv2d( + input_1, + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_1 = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x = input_2.permute(0, 2, 3, 1) + input_2 = None + x_1 = torch.nn.functional.layer_norm( + x, + (64,), + l_self_modules_stem_modules_2_parameters_weight_, + l_self_modules_stem_modules_2_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_2_parameters_weight_ + ) = l_self_modules_stem_modules_2_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 64, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 64, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_21 = torch.conv2d( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (64,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_3 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_3, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_36 = torch.nn.functional.dropout(x_35, 0.0, False, False) + x_35 = None + reshape_2 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_37 = x_36.mul(reshape_2) + x_36 = reshape_2 = None + x_38 = x_37 + input_3 + x_37 = input_3 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = torch.conv2d( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (128,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_4 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_4, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_60 = torch.conv2d( + x_59, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_59 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + reshape_4 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_62 = x_61.mul(reshape_4) + x_61 = reshape_4 = None + x_63 = x_62 + input_4 + x_62 = input_4 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_71 = torch.conv2d( + x_70, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_70 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + reshape_5 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_73 = x_72.mul(reshape_5) + x_72 = reshape_5 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_84 = x_83.mul(reshape_6) + x_83 = reshape_6 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_115 = torch.conv2d( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = x_118.permute(0, 2, 3, 1) + x_118 = None + x_120 = torch.nn.functional.layer_norm( + x_119, + (256,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_119 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_121 = x_120.permute(0, 3, 1, 2) + x_120 = None + input_5 = torch.conv2d( + x_121, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_122 = torch.conv2d( + input_5, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_123 = x_122.permute(0, 2, 3, 1) + x_122 = None + x_124 = torch.nn.functional.layer_norm( + x_123, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_125 = x_124.permute(0, 3, 1, 2) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126) + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_129 = torch.conv2d( + x_128, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_128 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) + x_129 = None + reshape_10 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_131 = x_130.mul(reshape_10) + x_130 = reshape_10 = None + x_132 = x_131 + input_5 + x_131 = input_5 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_134 = x_133.permute(0, 2, 3, 1) + x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_134 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_136 = x_135.permute(0, 3, 1, 2) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137) + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = torch.conv2d( + x_139, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_139 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_141 = torch.nn.functional.dropout(x_140, 0.0, False, False) + x_140 = None + reshape_11 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_142 = x_141.mul(reshape_11) + x_141 = reshape_11 = None + x_143 = x_142 + x_132 + x_142 = x_132 = None + x_144 = torch.nn.functional.adaptive_avg_pool2d(x_143, 1) + x_143 = None + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (512,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = x_147.flatten(1, -1) + x_147 = None + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = torch._C._nn.linear( + x_149, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_149 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_150,) diff --git a/samples/timm/convnext_pico_ols.d1_in1k/weight_meta.py b/samples/timm/convnext_pico_ols.d1_in1k/weight_meta.py new file mode 100644 index 0000000000..3dcb9f67dd --- /dev/null +++ b/samples/timm/convnext_pico_ols.d1_in1k/weight_meta.py @@ -0,0 +1,1330 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [32, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.225 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [64, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_parameters_weight_: + name = "L_self_modules_stem_modules_2_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_parameters_bias_: + name = "L_self_modules_stem_modules_2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [64, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [64, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [128, 64, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [256, 128, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [512, 256, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_small.fb_in1k/graph_hash.txt b/samples/timm/convnext_small.fb_in1k/graph_hash.txt new file mode 100644 index 0000000000..1ecc6e1860 --- /dev/null +++ b/samples/timm/convnext_small.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +0b86a61c951a12e51f0f7e01e08306b14026c01e6a6f636521e61c790236501f \ No newline at end of file diff --git a/samples/timm/convnext_small.fb_in1k/graph_net.json b/samples/timm/convnext_small.fb_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnext_small.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_small.fb_in1k/input_meta.py b/samples/timm/convnext_small.fb_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_small.fb_in1k/input_tensor_constraints.py b/samples/timm/convnext_small.fb_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_small.fb_in1k/model.py b/samples/timm/convnext_small.fb_in1k/model.py new file mode 100644 index 0000000000..4d33872797 --- /dev/null +++ b/samples/timm/convnext_small.fb_in1k/model.py @@ -0,0 +1,2707 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (96,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_9 = torch._C._nn.linear( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = x_10.permute(0, 3, 1, 2) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = torch._C._nn.linear( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_19 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = x_21.permute(0, 3, 1, 2) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_31 = torch._C._nn.linear( + x_30, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_30 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = x_32.permute(0, 3, 1, 2) + x_32 = None + reshape_2 = l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_34 = x_33.mul(reshape_2) + x_33 = reshape_2 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (96,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_45 = torch._C._nn.linear( + x_44, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_44 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_46 = torch.nn.functional.dropout(x_45, 0.0, False, False) + x_45 = None + x_47 = x_46.permute(0, 3, 1, 2) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = torch._C._nn.linear( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_55 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_57 = torch.nn.functional.dropout(x_56, 0.0, False, False) + x_56 = None + x_58 = x_57.permute(0, 3, 1, 2) + x_57 = None + reshape_4 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_59 = x_58.mul(reshape_4) + x_58 = reshape_4 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + reshape_5 = l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_70 = x_69.mul(reshape_5) + x_69 = reshape_5 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (192,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = torch._C._nn.linear( + x_80, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_80 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_82 = torch.nn.functional.dropout(x_81, 0.0, False, False) + x_81 = None + x_83 = x_82.permute(0, 3, 1, 2) + x_82 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_84 = x_83.mul(reshape_6) + x_83 = reshape_6 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = torch._C._nn.linear( + x_91, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_91 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_93 = torch.nn.functional.dropout(x_92, 0.0, False, False) + x_92 = None + x_94 = x_93.permute(0, 3, 1, 2) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_104 = torch.nn.functional.dropout(x_103, 0.0, False, False) + x_103 = None + x_105 = x_104.permute(0, 3, 1, 2) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = torch._C._nn.linear( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_113 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_115 = torch.nn.functional.dropout(x_114, 0.0, False, False) + x_114 = None + x_116 = x_115.permute(0, 3, 1, 2) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_124 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_126.permute(0, 3, 1, 2) + x_126 = None + reshape_10 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_128 = x_127.mul(reshape_10) + x_127 = reshape_10 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = torch._C._nn.linear( + x_135, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_135 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_137 = torch.nn.functional.dropout(x_136, 0.0, False, False) + x_136 = None + x_138 = x_137.permute(0, 3, 1, 2) + x_137 = None + reshape_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_139 = x_138.mul(reshape_11) + x_138 = reshape_11 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = torch._C._nn.linear( + x_146, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_146 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_148 = torch.nn.functional.dropout(x_147, 0.0, False, False) + x_147 = None + x_149 = x_148.permute(0, 3, 1, 2) + x_148 = None + reshape_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + None + ) + x_150 = x_149.mul(reshape_12) + x_149 = reshape_12 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_159.permute(0, 3, 1, 2) + x_159 = None + reshape_13 = l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + None + ) + x_161 = x_160.mul(reshape_13) + x_160 = reshape_13 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_169 = torch._C._nn.linear( + x_168, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_168 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_170 = torch.nn.functional.dropout(x_169, 0.0, False, False) + x_169 = None + x_171 = x_170.permute(0, 3, 1, 2) + x_170 = None + reshape_14 = l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + None + ) + x_172 = x_171.mul(reshape_14) + x_171 = reshape_14 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = (None) + x_175 = x_174.permute(0, 2, 3, 1) + x_174 = None + x_176 = torch.nn.functional.layer_norm( + x_175, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_, + 1e-06, + ) + x_175 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = (None) + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_176 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_178 = torch._C._nn.gelu(x_177) + x_177 = None + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_180 = torch._C._nn.linear( + x_179, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_179 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_181 = torch.nn.functional.dropout(x_180, 0.0, False, False) + x_180 = None + x_182 = x_181.permute(0, 3, 1, 2) + x_181 = None + reshape_15 = l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ = ( + None + ) + x_183 = x_182.mul(reshape_15) + x_182 = reshape_15 = None + x_184 = x_183 + x_173 + x_183 = x_173 = None + x_185 = torch.conv2d( + x_184, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = (None) + x_186 = x_185.permute(0, 2, 3, 1) + x_185 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = (None) + x_188 = torch._C._nn.linear( + x_187, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_187 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_189 = torch._C._nn.gelu(x_188) + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = torch._C._nn.linear( + x_190, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_190 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_192 = torch.nn.functional.dropout(x_191, 0.0, False, False) + x_191 = None + x_193 = x_192.permute(0, 3, 1, 2) + x_192 = None + reshape_16 = l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ = ( + None + ) + x_194 = x_193.mul(reshape_16) + x_193 = reshape_16 = None + x_195 = x_194 + x_184 + x_194 = x_184 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = (None) + x_197 = x_196.permute(0, 2, 3, 1) + x_196 = None + x_198 = torch.nn.functional.layer_norm( + x_197, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_, + 1e-06, + ) + x_197 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = (None) + x_199 = torch._C._nn.linear( + x_198, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_198 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_200 = torch._C._nn.gelu(x_199) + x_199 = None + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_202 = torch._C._nn.linear( + x_201, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_201 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_203 = torch.nn.functional.dropout(x_202, 0.0, False, False) + x_202 = None + x_204 = x_203.permute(0, 3, 1, 2) + x_203 = None + reshape_17 = l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ = ( + None + ) + x_205 = x_204.mul(reshape_17) + x_204 = reshape_17 = None + x_206 = x_205 + x_195 + x_205 = x_195 = None + x_207 = torch.conv2d( + x_206, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = (None) + x_208 = x_207.permute(0, 2, 3, 1) + x_207 = None + x_209 = torch.nn.functional.layer_norm( + x_208, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_, + 1e-06, + ) + x_208 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = (None) + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_209 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_211 = torch._C._nn.gelu(x_210) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_213 = torch._C._nn.linear( + x_212, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_212 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_214 = torch.nn.functional.dropout(x_213, 0.0, False, False) + x_213 = None + x_215 = x_214.permute(0, 3, 1, 2) + x_214 = None + reshape_18 = l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ = ( + None + ) + x_216 = x_215.mul(reshape_18) + x_215 = reshape_18 = None + x_217 = x_216 + x_206 + x_216 = x_206 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = (None) + x_219 = x_218.permute(0, 2, 3, 1) + x_218 = None + x_220 = torch.nn.functional.layer_norm( + x_219, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_, + 1e-06, + ) + x_219 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = (None) + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_220 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_222 = torch._C._nn.gelu(x_221) + x_221 = None + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_224 = torch._C._nn.linear( + x_223, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_223 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = x_225.permute(0, 3, 1, 2) + x_225 = None + reshape_19 = l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ = ( + None + ) + x_227 = x_226.mul(reshape_19) + x_226 = reshape_19 = None + x_228 = x_227 + x_217 + x_227 = x_217 = None + x_229 = torch.conv2d( + x_228, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = (None) + x_230 = x_229.permute(0, 2, 3, 1) + x_229 = None + x_231 = torch.nn.functional.layer_norm( + x_230, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_, + 1e-06, + ) + x_230 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = (None) + x_232 = torch._C._nn.linear( + x_231, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_231 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_233 = torch._C._nn.gelu(x_232) + x_232 = None + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_235 = torch._C._nn.linear( + x_234, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_234 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_236 = torch.nn.functional.dropout(x_235, 0.0, False, False) + x_235 = None + x_237 = x_236.permute(0, 3, 1, 2) + x_236 = None + reshape_20 = l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ = ( + None + ) + x_238 = x_237.mul(reshape_20) + x_237 = reshape_20 = None + x_239 = x_238 + x_228 + x_238 = x_228 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = (None) + x_241 = x_240.permute(0, 2, 3, 1) + x_240 = None + x_242 = torch.nn.functional.layer_norm( + x_241, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_, + 1e-06, + ) + x_241 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = (None) + x_243 = torch._C._nn.linear( + x_242, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_242 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_244 = torch._C._nn.gelu(x_243) + x_243 = None + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_246 = torch._C._nn.linear( + x_245, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_245 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_247 = torch.nn.functional.dropout(x_246, 0.0, False, False) + x_246 = None + x_248 = x_247.permute(0, 3, 1, 2) + x_247 = None + reshape_21 = l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ = ( + None + ) + x_249 = x_248.mul(reshape_21) + x_248 = reshape_21 = None + x_250 = x_249 + x_239 + x_249 = x_239 = None + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = (None) + x_252 = x_251.permute(0, 2, 3, 1) + x_251 = None + x_253 = torch.nn.functional.layer_norm( + x_252, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_, + 1e-06, + ) + x_252 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = (None) + x_254 = torch._C._nn.linear( + x_253, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_253 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_255 = torch._C._nn.gelu(x_254) + x_254 = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_257 = torch._C._nn.linear( + x_256, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_256 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_258 = torch.nn.functional.dropout(x_257, 0.0, False, False) + x_257 = None + x_259 = x_258.permute(0, 3, 1, 2) + x_258 = None + reshape_22 = l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ = ( + None + ) + x_260 = x_259.mul(reshape_22) + x_259 = reshape_22 = None + x_261 = x_260 + x_250 + x_260 = x_250 = None + x_262 = torch.conv2d( + x_261, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = (None) + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = (None) + x_265 = torch._C._nn.linear( + x_264, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_264 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_266 = torch._C._nn.gelu(x_265) + x_265 = None + x_267 = torch.nn.functional.dropout(x_266, 0.0, False, False) + x_266 = None + x_268 = torch._C._nn.linear( + x_267, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_269 = torch.nn.functional.dropout(x_268, 0.0, False, False) + x_268 = None + x_270 = x_269.permute(0, 3, 1, 2) + x_269 = None + reshape_23 = l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ = ( + None + ) + x_271 = x_270.mul(reshape_23) + x_270 = reshape_23 = None + x_272 = x_271 + x_261 + x_271 = x_261 = None + x_273 = torch.conv2d( + x_272, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = (None) + x_274 = x_273.permute(0, 2, 3, 1) + x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_, + 1e-06, + ) + x_274 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = (None) + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_277 = torch._C._nn.gelu(x_276) + x_276 = None + x_278 = torch.nn.functional.dropout(x_277, 0.0, False, False) + x_277 = None + x_279 = torch._C._nn.linear( + x_278, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_280 = torch.nn.functional.dropout(x_279, 0.0, False, False) + x_279 = None + x_281 = x_280.permute(0, 3, 1, 2) + x_280 = None + reshape_24 = l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ = ( + None + ) + x_282 = x_281.mul(reshape_24) + x_281 = reshape_24 = None + x_283 = x_282 + x_272 + x_282 = x_272 = None + x_284 = torch.conv2d( + x_283, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = (None) + x_285 = x_284.permute(0, 2, 3, 1) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_, + 1e-06, + ) + x_285 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = (None) + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_288 = torch._C._nn.gelu(x_287) + x_287 = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + x_290 = torch._C._nn.linear( + x_289, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_291 = torch.nn.functional.dropout(x_290, 0.0, False, False) + x_290 = None + x_292 = x_291.permute(0, 3, 1, 2) + x_291 = None + reshape_25 = l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ = ( + None + ) + x_293 = x_292.mul(reshape_25) + x_292 = reshape_25 = None + x_294 = x_293 + x_283 + x_293 = x_283 = None + x_295 = torch.conv2d( + x_294, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = (None) + x_296 = x_295.permute(0, 2, 3, 1) + x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_, + 1e-06, + ) + x_296 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = (None) + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_299 = torch._C._nn.gelu(x_298) + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_301 = torch._C._nn.linear( + x_300, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_302 = torch.nn.functional.dropout(x_301, 0.0, False, False) + x_301 = None + x_303 = x_302.permute(0, 3, 1, 2) + x_302 = None + reshape_26 = l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ = ( + None + ) + x_304 = x_303.mul(reshape_26) + x_303 = reshape_26 = None + x_305 = x_304 + x_294 + x_304 = x_294 = None + x_306 = torch.conv2d( + x_305, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = (None) + x_307 = x_306.permute(0, 2, 3, 1) + x_306 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_, + 1e-06, + ) + x_307 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = (None) + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_310 = torch._C._nn.gelu(x_309) + x_309 = None + x_311 = torch.nn.functional.dropout(x_310, 0.0, False, False) + x_310 = None + x_312 = torch._C._nn.linear( + x_311, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_313 = torch.nn.functional.dropout(x_312, 0.0, False, False) + x_312 = None + x_314 = x_313.permute(0, 3, 1, 2) + x_313 = None + reshape_27 = l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ = ( + None + ) + x_315 = x_314.mul(reshape_27) + x_314 = reshape_27 = None + x_316 = x_315 + x_305 + x_315 = x_305 = None + x_317 = torch.conv2d( + x_316, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = (None) + x_318 = x_317.permute(0, 2, 3, 1) + x_317 = None + x_319 = torch.nn.functional.layer_norm( + x_318, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_, + 1e-06, + ) + x_318 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = (None) + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_321 = torch._C._nn.gelu(x_320) + x_320 = None + x_322 = torch.nn.functional.dropout(x_321, 0.0, False, False) + x_321 = None + x_323 = torch._C._nn.linear( + x_322, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_324 = torch.nn.functional.dropout(x_323, 0.0, False, False) + x_323 = None + x_325 = x_324.permute(0, 3, 1, 2) + x_324 = None + reshape_28 = l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ = ( + None + ) + x_326 = x_325.mul(reshape_28) + x_325 = reshape_28 = None + x_327 = x_326 + x_316 + x_326 = x_316 = None + x_328 = torch.conv2d( + x_327, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = (None) + x_329 = x_328.permute(0, 2, 3, 1) + x_328 = None + x_330 = torch.nn.functional.layer_norm( + x_329, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_, + 1e-06, + ) + x_329 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = (None) + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_332 = torch._C._nn.gelu(x_331) + x_331 = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_334 = torch._C._nn.linear( + x_333, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_335 = torch.nn.functional.dropout(x_334, 0.0, False, False) + x_334 = None + x_336 = x_335.permute(0, 3, 1, 2) + x_335 = None + reshape_29 = l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ = ( + None + ) + x_337 = x_336.mul(reshape_29) + x_336 = reshape_29 = None + x_338 = x_337 + x_327 + x_337 = x_327 = None + x_339 = torch.conv2d( + x_338, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = (None) + x_340 = x_339.permute(0, 2, 3, 1) + x_339 = None + x_341 = torch.nn.functional.layer_norm( + x_340, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_, + 1e-06, + ) + x_340 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = (None) + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_343 = torch._C._nn.gelu(x_342) + x_342 = None + x_344 = torch.nn.functional.dropout(x_343, 0.0, False, False) + x_343 = None + x_345 = torch._C._nn.linear( + x_344, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_346 = torch.nn.functional.dropout(x_345, 0.0, False, False) + x_345 = None + x_347 = x_346.permute(0, 3, 1, 2) + x_346 = None + reshape_30 = l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ = ( + None + ) + x_348 = x_347.mul(reshape_30) + x_347 = reshape_30 = None + x_349 = x_348 + x_338 + x_348 = x_338 = None + x_350 = torch.conv2d( + x_349, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = (None) + x_351 = x_350.permute(0, 2, 3, 1) + x_350 = None + x_352 = torch.nn.functional.layer_norm( + x_351, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_, + 1e-06, + ) + x_351 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = (None) + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_354 = torch._C._nn.gelu(x_353) + x_353 = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_356 = torch._C._nn.linear( + x_355, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_357 = torch.nn.functional.dropout(x_356, 0.0, False, False) + x_356 = None + x_358 = x_357.permute(0, 3, 1, 2) + x_357 = None + reshape_31 = l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ = ( + None + ) + x_359 = x_358.mul(reshape_31) + x_358 = reshape_31 = None + x_360 = x_359 + x_349 + x_359 = x_349 = None + x_361 = torch.conv2d( + x_360, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = (None) + x_362 = x_361.permute(0, 2, 3, 1) + x_361 = None + x_363 = torch.nn.functional.layer_norm( + x_362, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_, + 1e-06, + ) + x_362 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = (None) + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_365 = torch._C._nn.gelu(x_364) + x_364 = None + x_366 = torch.nn.functional.dropout(x_365, 0.0, False, False) + x_365 = None + x_367 = torch._C._nn.linear( + x_366, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_368 = torch.nn.functional.dropout(x_367, 0.0, False, False) + x_367 = None + x_369 = x_368.permute(0, 3, 1, 2) + x_368 = None + reshape_32 = l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ = ( + None + ) + x_370 = x_369.mul(reshape_32) + x_369 = reshape_32 = None + x_371 = x_370 + x_360 + x_370 = x_360 = None + x_372 = x_371.permute(0, 2, 3, 1) + x_371 = None + x_373 = torch.nn.functional.layer_norm( + x_372, + (384,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_372 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_374 = x_373.permute(0, 3, 1, 2) + x_373 = None + input_4 = torch.conv2d( + x_374, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_374 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_375 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_376 = x_375.permute(0, 2, 3, 1) + x_375 = None + x_377 = torch.nn.functional.layer_norm( + x_376, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_376 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_378 = torch._C._nn.linear( + x_377, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_377 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_379 = torch._C._nn.gelu(x_378) + x_378 = None + x_380 = torch.nn.functional.dropout(x_379, 0.0, False, False) + x_379 = None + x_381 = torch._C._nn.linear( + x_380, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_380 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_382 = torch.nn.functional.dropout(x_381, 0.0, False, False) + x_381 = None + x_383 = x_382.permute(0, 3, 1, 2) + x_382 = None + reshape_33 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_384 = x_383.mul(reshape_33) + x_383 = reshape_33 = None + x_385 = x_384 + input_4 + x_384 = input_4 = None + x_386 = torch.conv2d( + x_385, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_387 = x_386.permute(0, 2, 3, 1) + x_386 = None + x_388 = torch.nn.functional.layer_norm( + x_387, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_387 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_389 = torch._C._nn.linear( + x_388, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_388 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_390 = torch._C._nn.gelu(x_389) + x_389 = None + x_391 = torch.nn.functional.dropout(x_390, 0.0, False, False) + x_390 = None + x_392 = torch._C._nn.linear( + x_391, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_391 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_393 = torch.nn.functional.dropout(x_392, 0.0, False, False) + x_392 = None + x_394 = x_393.permute(0, 3, 1, 2) + x_393 = None + reshape_34 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_395 = x_394.mul(reshape_34) + x_394 = reshape_34 = None + x_396 = x_395 + x_385 + x_395 = x_385 = None + x_397 = torch.conv2d( + x_396, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_398 = x_397.permute(0, 2, 3, 1) + x_397 = None + x_399 = torch.nn.functional.layer_norm( + x_398, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_398 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_400 = torch._C._nn.linear( + x_399, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_399 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_401 = torch._C._nn.gelu(x_400) + x_400 = None + x_402 = torch.nn.functional.dropout(x_401, 0.0, False, False) + x_401 = None + x_403 = torch._C._nn.linear( + x_402, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_402 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_404 = torch.nn.functional.dropout(x_403, 0.0, False, False) + x_403 = None + x_405 = x_404.permute(0, 3, 1, 2) + x_404 = None + reshape_35 = l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_406 = x_405.mul(reshape_35) + x_405 = reshape_35 = None + x_407 = x_406 + x_396 + x_406 = x_396 = None + x_408 = torch.nn.functional.adaptive_avg_pool2d(x_407, 1) + x_407 = None + x_409 = x_408.permute(0, 2, 3, 1) + x_408 = None + x_410 = torch.nn.functional.layer_norm( + x_409, + (768,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_409 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_411 = x_410.permute(0, 3, 1, 2) + x_410 = None + x_412 = x_411.flatten(1, -1) + x_411 = None + x_413 = torch.nn.functional.dropout(x_412, 0.0, False, False) + x_412 = None + x_414 = torch._C._nn.linear( + x_413, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_413 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_414,) diff --git a/samples/timm/convnext_small.fb_in1k/weight_meta.py b/samples/timm/convnext_small.fb_in1k/weight_meta.py new file mode 100644 index 0000000000..4444eb5606 --- /dev/null +++ b/samples/timm/convnext_small.fb_in1k/weight_meta.py @@ -0,0 +1,3470 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [96, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.225 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [192, 96, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [384, 192, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [768, 384, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_tiny.fb_in1k/graph_hash.txt b/samples/timm/convnext_tiny.fb_in1k/graph_hash.txt new file mode 100644 index 0000000000..b81f0c3337 --- /dev/null +++ b/samples/timm/convnext_tiny.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +49bda9b465818c7f5416d39636d24cdae7ea5d056153a366e1f60fe453b56e6e \ No newline at end of file diff --git a/samples/timm/convnext_tiny.fb_in1k/graph_net.json b/samples/timm/convnext_tiny.fb_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnext_tiny.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_tiny.fb_in1k/input_meta.py b/samples/timm/convnext_tiny.fb_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_tiny.fb_in1k/input_tensor_constraints.py b/samples/timm/convnext_tiny.fb_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_tiny.fb_in1k/model.py b/samples/timm/convnext_tiny.fb_in1k/model.py new file mode 100644 index 0000000000..1ff70532d8 --- /dev/null +++ b/samples/timm/convnext_tiny.fb_in1k/model.py @@ -0,0 +1,1447 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (96,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_9 = torch._C._nn.linear( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = x_10.permute(0, 3, 1, 2) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = torch._C._nn.linear( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_19 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = x_21.permute(0, 3, 1, 2) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_31 = torch._C._nn.linear( + x_30, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_30 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = x_32.permute(0, 3, 1, 2) + x_32 = None + reshape_2 = l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_34 = x_33.mul(reshape_2) + x_33 = reshape_2 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (96,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_45 = torch._C._nn.linear( + x_44, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_44 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_46 = torch.nn.functional.dropout(x_45, 0.0, False, False) + x_45 = None + x_47 = x_46.permute(0, 3, 1, 2) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = torch._C._nn.linear( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_55 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_57 = torch.nn.functional.dropout(x_56, 0.0, False, False) + x_56 = None + x_58 = x_57.permute(0, 3, 1, 2) + x_57 = None + reshape_4 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_59 = x_58.mul(reshape_4) + x_58 = reshape_4 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + reshape_5 = l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_70 = x_69.mul(reshape_5) + x_69 = reshape_5 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (192,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = torch._C._nn.linear( + x_80, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_80 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_82 = torch.nn.functional.dropout(x_81, 0.0, False, False) + x_81 = None + x_83 = x_82.permute(0, 3, 1, 2) + x_82 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_84 = x_83.mul(reshape_6) + x_83 = reshape_6 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = torch._C._nn.linear( + x_91, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_91 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_93 = torch.nn.functional.dropout(x_92, 0.0, False, False) + x_92 = None + x_94 = x_93.permute(0, 3, 1, 2) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_104 = torch.nn.functional.dropout(x_103, 0.0, False, False) + x_103 = None + x_105 = x_104.permute(0, 3, 1, 2) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = torch._C._nn.linear( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_113 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_115 = torch.nn.functional.dropout(x_114, 0.0, False, False) + x_114 = None + x_116 = x_115.permute(0, 3, 1, 2) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_124 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_126.permute(0, 3, 1, 2) + x_126 = None + reshape_10 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_128 = x_127.mul(reshape_10) + x_127 = reshape_10 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = torch._C._nn.linear( + x_135, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_135 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_137 = torch.nn.functional.dropout(x_136, 0.0, False, False) + x_136 = None + x_138 = x_137.permute(0, 3, 1, 2) + x_137 = None + reshape_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_139 = x_138.mul(reshape_11) + x_138 = reshape_11 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = torch._C._nn.linear( + x_146, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_146 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_148 = torch.nn.functional.dropout(x_147, 0.0, False, False) + x_147 = None + x_149 = x_148.permute(0, 3, 1, 2) + x_148 = None + reshape_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + None + ) + x_150 = x_149.mul(reshape_12) + x_149 = reshape_12 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_159.permute(0, 3, 1, 2) + x_159 = None + reshape_13 = l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + None + ) + x_161 = x_160.mul(reshape_13) + x_160 = reshape_13 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_169 = torch._C._nn.linear( + x_168, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_168 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_170 = torch.nn.functional.dropout(x_169, 0.0, False, False) + x_169 = None + x_171 = x_170.permute(0, 3, 1, 2) + x_170 = None + reshape_14 = l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + None + ) + x_172 = x_171.mul(reshape_14) + x_171 = reshape_14 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = x_173.permute(0, 2, 3, 1) + x_173 = None + x_175 = torch.nn.functional.layer_norm( + x_174, + (384,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_174 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_176 = x_175.permute(0, 3, 1, 2) + x_175 = None + input_4 = torch.conv2d( + x_176, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_176 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_177 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_178 = x_177.permute(0, 2, 3, 1) + x_177 = None + x_179 = torch.nn.functional.layer_norm( + x_178, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_178 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_180 = torch._C._nn.linear( + x_179, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_179 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_181 = torch._C._nn.gelu(x_180) + x_180 = None + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = torch._C._nn.linear( + x_182, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_182 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_184 = torch.nn.functional.dropout(x_183, 0.0, False, False) + x_183 = None + x_185 = x_184.permute(0, 3, 1, 2) + x_184 = None + reshape_15 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_186 = x_185.mul(reshape_15) + x_185 = reshape_15 = None + x_187 = x_186 + input_4 + x_186 = input_4 = None + x_188 = torch.conv2d( + x_187, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_189 = x_188.permute(0, 2, 3, 1) + x_188 = None + x_190 = torch.nn.functional.layer_norm( + x_189, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_189 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_191 = torch._C._nn.linear( + x_190, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_190 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_192 = torch._C._nn.gelu(x_191) + x_191 = None + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = torch._C._nn.linear( + x_193, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_193 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_195 = torch.nn.functional.dropout(x_194, 0.0, False, False) + x_194 = None + x_196 = x_195.permute(0, 3, 1, 2) + x_195 = None + reshape_16 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_197 = x_196.mul(reshape_16) + x_196 = reshape_16 = None + x_198 = x_197 + x_187 + x_197 = x_187 = None + x_199 = torch.conv2d( + x_198, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_200 = x_199.permute(0, 2, 3, 1) + x_199 = None + x_201 = torch.nn.functional.layer_norm( + x_200, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_200 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_202 = torch._C._nn.linear( + x_201, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_201 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_203 = torch._C._nn.gelu(x_202) + x_202 = None + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_205 = torch._C._nn.linear( + x_204, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_204 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_206 = torch.nn.functional.dropout(x_205, 0.0, False, False) + x_205 = None + x_207 = x_206.permute(0, 3, 1, 2) + x_206 = None + reshape_17 = l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_208 = x_207.mul(reshape_17) + x_207 = reshape_17 = None + x_209 = x_208 + x_198 + x_208 = x_198 = None + x_210 = torch.nn.functional.adaptive_avg_pool2d(x_209, 1) + x_209 = None + x_211 = x_210.permute(0, 2, 3, 1) + x_210 = None + x_212 = torch.nn.functional.layer_norm( + x_211, + (768,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_211 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_213 = x_212.permute(0, 3, 1, 2) + x_212 = None + x_214 = x_213.flatten(1, -1) + x_213 = None + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = torch._C._nn.linear( + x_215, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_215 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_216,) diff --git a/samples/timm/convnext_tiny.fb_in1k/weight_meta.py b/samples/timm/convnext_tiny.fb_in1k/weight_meta.py new file mode 100644 index 0000000000..075d12cb41 --- /dev/null +++ b/samples/timm/convnext_tiny.fb_in1k/weight_meta.py @@ -0,0 +1,1850 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [96, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.230 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [192, 96, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [384, 192, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [768, 384, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_tiny_hnf.a2h_in1k/graph_hash.txt b/samples/timm/convnext_tiny_hnf.a2h_in1k/graph_hash.txt new file mode 100644 index 0000000000..e13286982e --- /dev/null +++ b/samples/timm/convnext_tiny_hnf.a2h_in1k/graph_hash.txt @@ -0,0 +1 @@ +0d3941534699169e11117d7d410fb329687499ec49c1026f9adc7c4f15cbaa83 \ No newline at end of file diff --git a/samples/timm/convnext_tiny_hnf.a2h_in1k/graph_net.json b/samples/timm/convnext_tiny_hnf.a2h_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnext_tiny_hnf.a2h_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_tiny_hnf.a2h_in1k/input_meta.py b/samples/timm/convnext_tiny_hnf.a2h_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_tiny_hnf.a2h_in1k/input_tensor_constraints.py b/samples/timm/convnext_tiny_hnf.a2h_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_tiny_hnf.a2h_in1k/model.py b/samples/timm/convnext_tiny_hnf.a2h_in1k/model.py new file mode 100644 index 0000000000..9d9250ee8f --- /dev/null +++ b/samples/timm/convnext_tiny_hnf.a2h_in1k/model.py @@ -0,0 +1,1591 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_norm_pre_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_pre_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_norm_pre_parameters_weight_ = ( + L_self_modules_norm_pre_parameters_weight_ + ) + l_self_modules_norm_pre_parameters_bias_ = ( + L_self_modules_norm_pre_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (96,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = torch.conv2d( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_21 = torch.conv2d( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = x_27.permute(0, 3, 1, 2) + x_27 = None + x_29 = torch.conv2d( + x_28, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_28 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_30 = torch._C._nn.gelu(x_29) + x_29 = None + x_31 = torch.nn.functional.dropout(x_30, 0.0, False, False) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + reshape_2 = l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_34 = x_33.mul(reshape_2) + x_33 = reshape_2 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (96,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = torch.conv2d( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = x_52.permute(0, 3, 1, 2) + x_52 = None + x_54 = torch.conv2d( + x_53, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_53 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_55 = torch._C._nn.gelu(x_54) + x_54 = None + x_56 = torch.nn.functional.dropout(x_55, 0.0, False, False) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_58 = torch.nn.functional.dropout(x_57, 0.0, False, False) + x_57 = None + reshape_4 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_59 = x_58.mul(reshape_4) + x_58 = reshape_4 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = x_63.permute(0, 3, 1, 2) + x_63 = None + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_66 = torch._C._nn.gelu(x_65) + x_65 = None + x_67 = torch.nn.functional.dropout(x_66, 0.0, False, False) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + reshape_5 = l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_70 = x_69.mul(reshape_5) + x_69 = reshape_5 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (192,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_84 = x_83.mul(reshape_6) + x_83 = reshape_6 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_115 = torch.conv2d( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = x_121.permute(0, 3, 1, 2) + x_121 = None + x_123 = torch.conv2d( + x_122, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_124 = torch._C._nn.gelu(x_123) + x_123 = None + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + reshape_10 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_128 = x_127.mul(reshape_10) + x_127 = reshape_10 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = x_132.permute(0, 3, 1, 2) + x_132 = None + x_134 = torch.conv2d( + x_133, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_133 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_135 = torch._C._nn.gelu(x_134) + x_134 = None + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + reshape_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_139 = x_138.mul(reshape_11) + x_138 = reshape_11 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = x_143.permute(0, 3, 1, 2) + x_143 = None + x_145 = torch.conv2d( + x_144, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_144 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_146 = torch._C._nn.gelu(x_145) + x_145 = None + x_147 = torch.nn.functional.dropout(x_146, 0.0, False, False) + x_146 = None + x_148 = torch.conv2d( + x_147, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_147 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + reshape_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + None + ) + x_150 = x_149.mul(reshape_12) + x_149 = reshape_12 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = x_154.permute(0, 3, 1, 2) + x_154 = None + x_156 = torch.conv2d( + x_155, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_155 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_157 = torch._C._nn.gelu(x_156) + x_156 = None + x_158 = torch.nn.functional.dropout(x_157, 0.0, False, False) + x_157 = None + x_159 = torch.conv2d( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + reshape_13 = l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + None + ) + x_161 = x_160.mul(reshape_13) + x_160 = reshape_13 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = x_165.permute(0, 3, 1, 2) + x_165 = None + x_167 = torch.conv2d( + x_166, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_166 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_168 = torch._C._nn.gelu(x_167) + x_167 = None + x_169 = torch.nn.functional.dropout(x_168, 0.0, False, False) + x_168 = None + x_170 = torch.conv2d( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + reshape_14 = l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + None + ) + x_172 = x_171.mul(reshape_14) + x_171 = reshape_14 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = x_173.permute(0, 2, 3, 1) + x_173 = None + x_175 = torch.nn.functional.layer_norm( + x_174, + (384,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_174 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_176 = x_175.permute(0, 3, 1, 2) + x_175 = None + input_4 = torch.conv2d( + x_176, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_176 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_177 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_178 = x_177.permute(0, 2, 3, 1) + x_177 = None + x_179 = torch.nn.functional.layer_norm( + x_178, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_178 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_180 = x_179.permute(0, 3, 1, 2) + x_179 = None + x_181 = torch.conv2d( + x_180, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_180 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_182 = torch._C._nn.gelu(x_181) + x_181 = None + x_183 = torch.nn.functional.dropout(x_182, 0.0, False, False) + x_182 = None + x_184 = torch.conv2d( + x_183, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_183 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_185 = torch.nn.functional.dropout(x_184, 0.0, False, False) + x_184 = None + reshape_15 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_186 = x_185.mul(reshape_15) + x_185 = reshape_15 = None + x_187 = x_186 + input_4 + x_186 = input_4 = None + x_188 = torch.conv2d( + x_187, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_189 = x_188.permute(0, 2, 3, 1) + x_188 = None + x_190 = torch.nn.functional.layer_norm( + x_189, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_189 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_191 = x_190.permute(0, 3, 1, 2) + x_190 = None + x_192 = torch.conv2d( + x_191, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_191 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_193 = torch._C._nn.gelu(x_192) + x_192 = None + x_194 = torch.nn.functional.dropout(x_193, 0.0, False, False) + x_193 = None + x_195 = torch.conv2d( + x_194, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_194 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_196 = torch.nn.functional.dropout(x_195, 0.0, False, False) + x_195 = None + reshape_16 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_197 = x_196.mul(reshape_16) + x_196 = reshape_16 = None + x_198 = x_197 + x_187 + x_197 = x_187 = None + x_199 = torch.conv2d( + x_198, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_200 = x_199.permute(0, 2, 3, 1) + x_199 = None + x_201 = torch.nn.functional.layer_norm( + x_200, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_200 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_202 = x_201.permute(0, 3, 1, 2) + x_201 = None + x_203 = torch.conv2d( + x_202, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_202 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_204 = torch._C._nn.gelu(x_203) + x_203 = None + x_205 = torch.nn.functional.dropout(x_204, 0.0, False, False) + x_204 = None + x_206 = torch.conv2d( + x_205, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_205 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_207 = torch.nn.functional.dropout(x_206, 0.0, False, False) + x_206 = None + reshape_17 = l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_208 = x_207.mul(reshape_17) + x_207 = reshape_17 = None + x_209 = x_208 + x_198 + x_208 = x_198 = None + x_210 = x_209.permute(0, 2, 3, 1) + x_209 = None + x_211 = torch.nn.functional.layer_norm( + x_210, + (768,), + l_self_modules_norm_pre_parameters_weight_, + l_self_modules_norm_pre_parameters_bias_, + 1e-06, + ) + x_210 = ( + l_self_modules_norm_pre_parameters_weight_ + ) = l_self_modules_norm_pre_parameters_bias_ = None + x_212 = x_211.permute(0, 3, 1, 2) + x_211 = None + x_213 = torch.nn.functional.adaptive_avg_pool2d(x_212, 1) + x_212 = None + x_214 = x_213.flatten(1, -1) + x_213 = None + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = torch._C._nn.linear( + x_215, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_215 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_216,) diff --git a/samples/timm/convnext_tiny_hnf.a2h_in1k/weight_meta.py b/samples/timm/convnext_tiny_hnf.a2h_in1k/weight_meta.py new file mode 100644 index 0000000000..c5a0ff17dc --- /dev/null +++ b/samples/timm/convnext_tiny_hnf.a2h_in1k/weight_meta.py @@ -0,0 +1,1850 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [96, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.227 + std = 1.288 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [192, 96, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [384, 192, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [768, 384, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_pre_parameters_weight_: + name = "L_self_modules_norm_pre_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_pre_parameters_bias_: + name = "L_self_modules_norm_pre_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_xlarge.fb_in22k/graph_hash.txt b/samples/timm/convnext_xlarge.fb_in22k/graph_hash.txt new file mode 100644 index 0000000000..3a8a8f315e --- /dev/null +++ b/samples/timm/convnext_xlarge.fb_in22k/graph_hash.txt @@ -0,0 +1 @@ +e09f137b683383f74a7c3ec87bf480a8068e0212c4950307376b464a21869dcb \ No newline at end of file diff --git a/samples/timm/convnext_xlarge.fb_in22k/graph_net.json b/samples/timm/convnext_xlarge.fb_in22k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnext_xlarge.fb_in22k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_xlarge.fb_in22k/input_meta.py b/samples/timm/convnext_xlarge.fb_in22k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_xlarge.fb_in22k/input_tensor_constraints.py b/samples/timm/convnext_xlarge.fb_in22k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_xlarge.fb_in22k/model.py b/samples/timm/convnext_xlarge.fb_in22k/model.py new file mode 100644 index 0000000000..b56881e8da --- /dev/null +++ b/samples/timm/convnext_xlarge.fb_in22k/model.py @@ -0,0 +1,2707 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (256,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (256,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_9 = torch._C._nn.linear( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = x_10.permute(0, 3, 1, 2) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (256,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = torch._C._nn.linear( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_19 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = x_21.permute(0, 3, 1, 2) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (256,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_31 = torch._C._nn.linear( + x_30, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_30 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = x_32.permute(0, 3, 1, 2) + x_32 = None + reshape_2 = l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_34 = x_33.mul(reshape_2) + x_33 = reshape_2 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (256,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (512,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_45 = torch._C._nn.linear( + x_44, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_44 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_46 = torch.nn.functional.dropout(x_45, 0.0, False, False) + x_45 = None + x_47 = x_46.permute(0, 3, 1, 2) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (512,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = torch._C._nn.linear( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_55 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_57 = torch.nn.functional.dropout(x_56, 0.0, False, False) + x_56 = None + x_58 = x_57.permute(0, 3, 1, 2) + x_57 = None + reshape_4 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_59 = x_58.mul(reshape_4) + x_58 = reshape_4 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (512,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + reshape_5 = l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_70 = x_69.mul(reshape_5) + x_69 = reshape_5 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (512,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = torch._C._nn.linear( + x_80, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_80 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_82 = torch.nn.functional.dropout(x_81, 0.0, False, False) + x_81 = None + x_83 = x_82.permute(0, 3, 1, 2) + x_82 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_84 = x_83.mul(reshape_6) + x_83 = reshape_6 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = torch._C._nn.linear( + x_91, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_91 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_93 = torch.nn.functional.dropout(x_92, 0.0, False, False) + x_92 = None + x_94 = x_93.permute(0, 3, 1, 2) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_104 = torch.nn.functional.dropout(x_103, 0.0, False, False) + x_103 = None + x_105 = x_104.permute(0, 3, 1, 2) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = torch._C._nn.linear( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_113 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_115 = torch.nn.functional.dropout(x_114, 0.0, False, False) + x_114 = None + x_116 = x_115.permute(0, 3, 1, 2) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_124 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_126.permute(0, 3, 1, 2) + x_126 = None + reshape_10 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_128 = x_127.mul(reshape_10) + x_127 = reshape_10 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = torch._C._nn.linear( + x_135, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_135 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_137 = torch.nn.functional.dropout(x_136, 0.0, False, False) + x_136 = None + x_138 = x_137.permute(0, 3, 1, 2) + x_137 = None + reshape_11 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_139 = x_138.mul(reshape_11) + x_138 = reshape_11 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = torch._C._nn.linear( + x_146, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_146 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_148 = torch.nn.functional.dropout(x_147, 0.0, False, False) + x_147 = None + x_149 = x_148.permute(0, 3, 1, 2) + x_148 = None + reshape_12 = l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + None + ) + x_150 = x_149.mul(reshape_12) + x_149 = reshape_12 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_159.permute(0, 3, 1, 2) + x_159 = None + reshape_13 = l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + None + ) + x_161 = x_160.mul(reshape_13) + x_160 = reshape_13 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_169 = torch._C._nn.linear( + x_168, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_168 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_170 = torch.nn.functional.dropout(x_169, 0.0, False, False) + x_169 = None + x_171 = x_170.permute(0, 3, 1, 2) + x_170 = None + reshape_14 = l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + None + ) + x_172 = x_171.mul(reshape_14) + x_171 = reshape_14 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = (None) + x_175 = x_174.permute(0, 2, 3, 1) + x_174 = None + x_176 = torch.nn.functional.layer_norm( + x_175, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_, + 1e-06, + ) + x_175 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = (None) + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_176 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_178 = torch._C._nn.gelu(x_177) + x_177 = None + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_180 = torch._C._nn.linear( + x_179, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_179 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_181 = torch.nn.functional.dropout(x_180, 0.0, False, False) + x_180 = None + x_182 = x_181.permute(0, 3, 1, 2) + x_181 = None + reshape_15 = l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ = ( + None + ) + x_183 = x_182.mul(reshape_15) + x_182 = reshape_15 = None + x_184 = x_183 + x_173 + x_183 = x_173 = None + x_185 = torch.conv2d( + x_184, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = (None) + x_186 = x_185.permute(0, 2, 3, 1) + x_185 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = (None) + x_188 = torch._C._nn.linear( + x_187, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_187 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_189 = torch._C._nn.gelu(x_188) + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = torch._C._nn.linear( + x_190, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_190 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_192 = torch.nn.functional.dropout(x_191, 0.0, False, False) + x_191 = None + x_193 = x_192.permute(0, 3, 1, 2) + x_192 = None + reshape_16 = l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ = ( + None + ) + x_194 = x_193.mul(reshape_16) + x_193 = reshape_16 = None + x_195 = x_194 + x_184 + x_194 = x_184 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = (None) + x_197 = x_196.permute(0, 2, 3, 1) + x_196 = None + x_198 = torch.nn.functional.layer_norm( + x_197, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_, + 1e-06, + ) + x_197 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = (None) + x_199 = torch._C._nn.linear( + x_198, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_198 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_200 = torch._C._nn.gelu(x_199) + x_199 = None + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_202 = torch._C._nn.linear( + x_201, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_201 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_203 = torch.nn.functional.dropout(x_202, 0.0, False, False) + x_202 = None + x_204 = x_203.permute(0, 3, 1, 2) + x_203 = None + reshape_17 = l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ = ( + None + ) + x_205 = x_204.mul(reshape_17) + x_204 = reshape_17 = None + x_206 = x_205 + x_195 + x_205 = x_195 = None + x_207 = torch.conv2d( + x_206, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = (None) + x_208 = x_207.permute(0, 2, 3, 1) + x_207 = None + x_209 = torch.nn.functional.layer_norm( + x_208, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_, + 1e-06, + ) + x_208 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = (None) + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_209 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_211 = torch._C._nn.gelu(x_210) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_213 = torch._C._nn.linear( + x_212, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_212 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_214 = torch.nn.functional.dropout(x_213, 0.0, False, False) + x_213 = None + x_215 = x_214.permute(0, 3, 1, 2) + x_214 = None + reshape_18 = l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ = ( + None + ) + x_216 = x_215.mul(reshape_18) + x_215 = reshape_18 = None + x_217 = x_216 + x_206 + x_216 = x_206 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = (None) + x_219 = x_218.permute(0, 2, 3, 1) + x_218 = None + x_220 = torch.nn.functional.layer_norm( + x_219, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_, + 1e-06, + ) + x_219 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = (None) + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_220 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_222 = torch._C._nn.gelu(x_221) + x_221 = None + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_224 = torch._C._nn.linear( + x_223, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_223 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = x_225.permute(0, 3, 1, 2) + x_225 = None + reshape_19 = l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ = ( + None + ) + x_227 = x_226.mul(reshape_19) + x_226 = reshape_19 = None + x_228 = x_227 + x_217 + x_227 = x_217 = None + x_229 = torch.conv2d( + x_228, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = (None) + x_230 = x_229.permute(0, 2, 3, 1) + x_229 = None + x_231 = torch.nn.functional.layer_norm( + x_230, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_, + 1e-06, + ) + x_230 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = (None) + x_232 = torch._C._nn.linear( + x_231, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_231 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_233 = torch._C._nn.gelu(x_232) + x_232 = None + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_235 = torch._C._nn.linear( + x_234, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_234 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_236 = torch.nn.functional.dropout(x_235, 0.0, False, False) + x_235 = None + x_237 = x_236.permute(0, 3, 1, 2) + x_236 = None + reshape_20 = l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ = ( + None + ) + x_238 = x_237.mul(reshape_20) + x_237 = reshape_20 = None + x_239 = x_238 + x_228 + x_238 = x_228 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = (None) + x_241 = x_240.permute(0, 2, 3, 1) + x_240 = None + x_242 = torch.nn.functional.layer_norm( + x_241, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_, + 1e-06, + ) + x_241 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = (None) + x_243 = torch._C._nn.linear( + x_242, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_242 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_244 = torch._C._nn.gelu(x_243) + x_243 = None + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_246 = torch._C._nn.linear( + x_245, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_245 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_247 = torch.nn.functional.dropout(x_246, 0.0, False, False) + x_246 = None + x_248 = x_247.permute(0, 3, 1, 2) + x_247 = None + reshape_21 = l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ = ( + None + ) + x_249 = x_248.mul(reshape_21) + x_248 = reshape_21 = None + x_250 = x_249 + x_239 + x_249 = x_239 = None + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = (None) + x_252 = x_251.permute(0, 2, 3, 1) + x_251 = None + x_253 = torch.nn.functional.layer_norm( + x_252, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_, + 1e-06, + ) + x_252 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = (None) + x_254 = torch._C._nn.linear( + x_253, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_253 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_255 = torch._C._nn.gelu(x_254) + x_254 = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_257 = torch._C._nn.linear( + x_256, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_256 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_258 = torch.nn.functional.dropout(x_257, 0.0, False, False) + x_257 = None + x_259 = x_258.permute(0, 3, 1, 2) + x_258 = None + reshape_22 = l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ = ( + None + ) + x_260 = x_259.mul(reshape_22) + x_259 = reshape_22 = None + x_261 = x_260 + x_250 + x_260 = x_250 = None + x_262 = torch.conv2d( + x_261, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = (None) + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = (None) + x_265 = torch._C._nn.linear( + x_264, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_264 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_266 = torch._C._nn.gelu(x_265) + x_265 = None + x_267 = torch.nn.functional.dropout(x_266, 0.0, False, False) + x_266 = None + x_268 = torch._C._nn.linear( + x_267, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_269 = torch.nn.functional.dropout(x_268, 0.0, False, False) + x_268 = None + x_270 = x_269.permute(0, 3, 1, 2) + x_269 = None + reshape_23 = l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ = ( + None + ) + x_271 = x_270.mul(reshape_23) + x_270 = reshape_23 = None + x_272 = x_271 + x_261 + x_271 = x_261 = None + x_273 = torch.conv2d( + x_272, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = (None) + x_274 = x_273.permute(0, 2, 3, 1) + x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_, + 1e-06, + ) + x_274 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = (None) + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_277 = torch._C._nn.gelu(x_276) + x_276 = None + x_278 = torch.nn.functional.dropout(x_277, 0.0, False, False) + x_277 = None + x_279 = torch._C._nn.linear( + x_278, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_280 = torch.nn.functional.dropout(x_279, 0.0, False, False) + x_279 = None + x_281 = x_280.permute(0, 3, 1, 2) + x_280 = None + reshape_24 = l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ = ( + None + ) + x_282 = x_281.mul(reshape_24) + x_281 = reshape_24 = None + x_283 = x_282 + x_272 + x_282 = x_272 = None + x_284 = torch.conv2d( + x_283, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = (None) + x_285 = x_284.permute(0, 2, 3, 1) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_, + 1e-06, + ) + x_285 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = (None) + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_288 = torch._C._nn.gelu(x_287) + x_287 = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + x_290 = torch._C._nn.linear( + x_289, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_291 = torch.nn.functional.dropout(x_290, 0.0, False, False) + x_290 = None + x_292 = x_291.permute(0, 3, 1, 2) + x_291 = None + reshape_25 = l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ = ( + None + ) + x_293 = x_292.mul(reshape_25) + x_292 = reshape_25 = None + x_294 = x_293 + x_283 + x_293 = x_283 = None + x_295 = torch.conv2d( + x_294, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = (None) + x_296 = x_295.permute(0, 2, 3, 1) + x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_, + 1e-06, + ) + x_296 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = (None) + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_299 = torch._C._nn.gelu(x_298) + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_301 = torch._C._nn.linear( + x_300, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_302 = torch.nn.functional.dropout(x_301, 0.0, False, False) + x_301 = None + x_303 = x_302.permute(0, 3, 1, 2) + x_302 = None + reshape_26 = l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ = ( + None + ) + x_304 = x_303.mul(reshape_26) + x_303 = reshape_26 = None + x_305 = x_304 + x_294 + x_304 = x_294 = None + x_306 = torch.conv2d( + x_305, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = (None) + x_307 = x_306.permute(0, 2, 3, 1) + x_306 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_, + 1e-06, + ) + x_307 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = (None) + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_310 = torch._C._nn.gelu(x_309) + x_309 = None + x_311 = torch.nn.functional.dropout(x_310, 0.0, False, False) + x_310 = None + x_312 = torch._C._nn.linear( + x_311, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_313 = torch.nn.functional.dropout(x_312, 0.0, False, False) + x_312 = None + x_314 = x_313.permute(0, 3, 1, 2) + x_313 = None + reshape_27 = l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ = ( + None + ) + x_315 = x_314.mul(reshape_27) + x_314 = reshape_27 = None + x_316 = x_315 + x_305 + x_315 = x_305 = None + x_317 = torch.conv2d( + x_316, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = (None) + x_318 = x_317.permute(0, 2, 3, 1) + x_317 = None + x_319 = torch.nn.functional.layer_norm( + x_318, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_, + 1e-06, + ) + x_318 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = (None) + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_321 = torch._C._nn.gelu(x_320) + x_320 = None + x_322 = torch.nn.functional.dropout(x_321, 0.0, False, False) + x_321 = None + x_323 = torch._C._nn.linear( + x_322, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_324 = torch.nn.functional.dropout(x_323, 0.0, False, False) + x_323 = None + x_325 = x_324.permute(0, 3, 1, 2) + x_324 = None + reshape_28 = l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ = ( + None + ) + x_326 = x_325.mul(reshape_28) + x_325 = reshape_28 = None + x_327 = x_326 + x_316 + x_326 = x_316 = None + x_328 = torch.conv2d( + x_327, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = (None) + x_329 = x_328.permute(0, 2, 3, 1) + x_328 = None + x_330 = torch.nn.functional.layer_norm( + x_329, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_, + 1e-06, + ) + x_329 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = (None) + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_332 = torch._C._nn.gelu(x_331) + x_331 = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_334 = torch._C._nn.linear( + x_333, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_335 = torch.nn.functional.dropout(x_334, 0.0, False, False) + x_334 = None + x_336 = x_335.permute(0, 3, 1, 2) + x_335 = None + reshape_29 = l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ = ( + None + ) + x_337 = x_336.mul(reshape_29) + x_336 = reshape_29 = None + x_338 = x_337 + x_327 + x_337 = x_327 = None + x_339 = torch.conv2d( + x_338, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = (None) + x_340 = x_339.permute(0, 2, 3, 1) + x_339 = None + x_341 = torch.nn.functional.layer_norm( + x_340, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_, + 1e-06, + ) + x_340 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = (None) + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_343 = torch._C._nn.gelu(x_342) + x_342 = None + x_344 = torch.nn.functional.dropout(x_343, 0.0, False, False) + x_343 = None + x_345 = torch._C._nn.linear( + x_344, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_346 = torch.nn.functional.dropout(x_345, 0.0, False, False) + x_345 = None + x_347 = x_346.permute(0, 3, 1, 2) + x_346 = None + reshape_30 = l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ = ( + None + ) + x_348 = x_347.mul(reshape_30) + x_347 = reshape_30 = None + x_349 = x_348 + x_338 + x_348 = x_338 = None + x_350 = torch.conv2d( + x_349, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = (None) + x_351 = x_350.permute(0, 2, 3, 1) + x_350 = None + x_352 = torch.nn.functional.layer_norm( + x_351, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_, + 1e-06, + ) + x_351 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = (None) + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_354 = torch._C._nn.gelu(x_353) + x_353 = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_356 = torch._C._nn.linear( + x_355, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_357 = torch.nn.functional.dropout(x_356, 0.0, False, False) + x_356 = None + x_358 = x_357.permute(0, 3, 1, 2) + x_357 = None + reshape_31 = l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ = ( + None + ) + x_359 = x_358.mul(reshape_31) + x_358 = reshape_31 = None + x_360 = x_359 + x_349 + x_359 = x_349 = None + x_361 = torch.conv2d( + x_360, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = (None) + x_362 = x_361.permute(0, 2, 3, 1) + x_361 = None + x_363 = torch.nn.functional.layer_norm( + x_362, + (1024,), + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_, + 1e-06, + ) + x_362 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = (None) + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_365 = torch._C._nn.gelu(x_364) + x_364 = None + x_366 = torch.nn.functional.dropout(x_365, 0.0, False, False) + x_365 = None + x_367 = torch._C._nn.linear( + x_366, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_368 = torch.nn.functional.dropout(x_367, 0.0, False, False) + x_367 = None + x_369 = x_368.permute(0, 3, 1, 2) + x_368 = None + reshape_32 = l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ = ( + None + ) + x_370 = x_369.mul(reshape_32) + x_369 = reshape_32 = None + x_371 = x_370 + x_360 + x_370 = x_360 = None + x_372 = x_371.permute(0, 2, 3, 1) + x_371 = None + x_373 = torch.nn.functional.layer_norm( + x_372, + (1024,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_372 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_374 = x_373.permute(0, 3, 1, 2) + x_373 = None + input_4 = torch.conv2d( + x_374, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_374 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_375 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 2048, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_376 = x_375.permute(0, 2, 3, 1) + x_375 = None + x_377 = torch.nn.functional.layer_norm( + x_376, + (2048,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_376 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_378 = torch._C._nn.linear( + x_377, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_377 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_379 = torch._C._nn.gelu(x_378) + x_378 = None + x_380 = torch.nn.functional.dropout(x_379, 0.0, False, False) + x_379 = None + x_381 = torch._C._nn.linear( + x_380, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_380 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_382 = torch.nn.functional.dropout(x_381, 0.0, False, False) + x_381 = None + x_383 = x_382.permute(0, 3, 1, 2) + x_382 = None + reshape_33 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_384 = x_383.mul(reshape_33) + x_383 = reshape_33 = None + x_385 = x_384 + input_4 + x_384 = input_4 = None + x_386 = torch.conv2d( + x_385, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 2048, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_387 = x_386.permute(0, 2, 3, 1) + x_386 = None + x_388 = torch.nn.functional.layer_norm( + x_387, + (2048,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_387 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_389 = torch._C._nn.linear( + x_388, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_388 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_390 = torch._C._nn.gelu(x_389) + x_389 = None + x_391 = torch.nn.functional.dropout(x_390, 0.0, False, False) + x_390 = None + x_392 = torch._C._nn.linear( + x_391, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_391 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_393 = torch.nn.functional.dropout(x_392, 0.0, False, False) + x_392 = None + x_394 = x_393.permute(0, 3, 1, 2) + x_393 = None + reshape_34 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_395 = x_394.mul(reshape_34) + x_394 = reshape_34 = None + x_396 = x_395 + x_385 + x_395 = x_385 = None + x_397 = torch.conv2d( + x_396, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 2048, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_398 = x_397.permute(0, 2, 3, 1) + x_397 = None + x_399 = torch.nn.functional.layer_norm( + x_398, + (2048,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_398 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_400 = torch._C._nn.linear( + x_399, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_399 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_401 = torch._C._nn.gelu(x_400) + x_400 = None + x_402 = torch.nn.functional.dropout(x_401, 0.0, False, False) + x_401 = None + x_403 = torch._C._nn.linear( + x_402, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_402 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_404 = torch.nn.functional.dropout(x_403, 0.0, False, False) + x_403 = None + x_405 = x_404.permute(0, 3, 1, 2) + x_404 = None + reshape_35 = l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_406 = x_405.mul(reshape_35) + x_405 = reshape_35 = None + x_407 = x_406 + x_396 + x_406 = x_396 = None + x_408 = torch.nn.functional.adaptive_avg_pool2d(x_407, 1) + x_407 = None + x_409 = x_408.permute(0, 2, 3, 1) + x_408 = None + x_410 = torch.nn.functional.layer_norm( + x_409, + (2048,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_409 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_411 = x_410.permute(0, 3, 1, 2) + x_410 = None + x_412 = x_411.flatten(1, -1) + x_411 = None + x_413 = torch.nn.functional.dropout(x_412, 0.0, False, False) + x_412 = None + x_414 = torch._C._nn.linear( + x_413, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_413 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_414,) diff --git a/samples/timm/convnext_xlarge.fb_in22k/weight_meta.py b/samples/timm/convnext_xlarge.fb_in22k/weight_meta.py new file mode 100644 index 0000000000..5b3472b412 --- /dev/null +++ b/samples/timm/convnext_xlarge.fb_in22k/weight_meta.py @@ -0,0 +1,3470 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [256, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.227 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [512, 256, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [1024, 512, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [2048, 1024, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [2048, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [8192, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [8192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [2048, 8192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [2048, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [8192, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [8192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [2048, 8192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [2048, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [8192, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [8192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [2048, 8192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [21841, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [21841] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_xxlarge.clip_laion2b_rewind/graph_hash.txt b/samples/timm/convnext_xxlarge.clip_laion2b_rewind/graph_hash.txt new file mode 100644 index 0000000000..6891d7f77e --- /dev/null +++ b/samples/timm/convnext_xxlarge.clip_laion2b_rewind/graph_hash.txt @@ -0,0 +1 @@ +c484ac258f449ea7468ea4628d2c98a514c2dec856c9b004245122a56539afe2 \ No newline at end of file diff --git a/samples/timm/convnext_xxlarge.clip_laion2b_rewind/graph_net.json b/samples/timm/convnext_xxlarge.clip_laion2b_rewind/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnext_xxlarge.clip_laion2b_rewind/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_xxlarge.clip_laion2b_rewind/input_meta.py b/samples/timm/convnext_xxlarge.clip_laion2b_rewind/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_xxlarge.clip_laion2b_rewind/input_tensor_constraints.py b/samples/timm/convnext_xxlarge.clip_laion2b_rewind/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_xxlarge.clip_laion2b_rewind/model.py b/samples/timm/convnext_xxlarge.clip_laion2b_rewind/model.py new file mode 100644 index 0000000000..893cd41e47 --- /dev/null +++ b/samples/timm/convnext_xxlarge.clip_laion2b_rewind/model.py @@ -0,0 +1,2987 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_27_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_28_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_29_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_27_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_27_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_28_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_28_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_29_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_29_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (384,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (384,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-05, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_9 = torch._C._nn.linear( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = x_10.permute(0, 3, 1, 2) + x_10 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_12 = x_11.mul(reshape) + x_11 = reshape = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (384,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-05, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = torch._C._nn.linear( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_19 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = x_21.permute(0, 3, 1, 2) + x_21 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_23 = x_22.mul(reshape_1) + x_22 = reshape_1 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (384,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-05, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_31 = torch._C._nn.linear( + x_30, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_30 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = x_32.permute(0, 3, 1, 2) + x_32 = None + reshape_2 = l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_34 = x_33.mul(reshape_2) + x_33 = reshape_2 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (384,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-05, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (768,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-05, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_45 = torch._C._nn.linear( + x_44, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_44 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_46 = torch.nn.functional.dropout(x_45, 0.0, False, False) + x_45 = None + x_47 = x_46.permute(0, 3, 1, 2) + x_46 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_48 = x_47.mul(reshape_3) + x_47 = reshape_3 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (768,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-05, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = torch._C._nn.linear( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_55 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_57 = torch.nn.functional.dropout(x_56, 0.0, False, False) + x_56 = None + x_58 = x_57.permute(0, 3, 1, 2) + x_57 = None + reshape_4 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_59 = x_58.mul(reshape_4) + x_58 = reshape_4 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (768,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-05, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_68.permute(0, 3, 1, 2) + x_68 = None + reshape_5 = l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_70 = x_69.mul(reshape_5) + x_69 = reshape_5 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_73 = x_72.permute(0, 2, 3, 1) + x_72 = None + x_74 = torch.nn.functional.layer_norm( + x_73, + (768,), + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-05, + ) + x_73 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_75 = torch._C._nn.linear( + x_74, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_74 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_76 = torch._C._nn.gelu(x_75) + x_75 = None + x_77 = torch.nn.functional.dropout(x_76, 0.0, False, False) + x_76 = None + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = x_79.permute(0, 3, 1, 2) + x_79 = None + reshape_6 = l_self_modules_stages_modules_1_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_81 = x_80.mul(reshape_6) + x_80 = reshape_6 = None + x_82 = x_81 + x_71 + x_81 = x_71 = None + x_83 = x_82.permute(0, 2, 3, 1) + x_82 = None + x_84 = torch.nn.functional.layer_norm( + x_83, + (768,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-05, + ) + x_83 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_85 = x_84.permute(0, 3, 1, 2) + x_84 = None + input_3 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_85 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_86 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-05, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = torch._C._nn.linear( + x_91, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_91 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_93 = torch.nn.functional.dropout(x_92, 0.0, False, False) + x_92 = None + x_94 = x_93.permute(0, 3, 1, 2) + x_93 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_95 = x_94.mul(reshape_7) + x_94 = reshape_7 = None + x_96 = x_95 + input_3 + x_95 = input_3 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-05, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_104 = torch.nn.functional.dropout(x_103, 0.0, False, False) + x_103 = None + x_105 = x_104.permute(0, 3, 1, 2) + x_104 = None + reshape_8 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_8) + x_105 = reshape_8 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-05, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = torch._C._nn.linear( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_113 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_115 = torch.nn.functional.dropout(x_114, 0.0, False, False) + x_114 = None + x_116 = x_115.permute(0, 3, 1, 2) + x_115 = None + reshape_9 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_117 = x_116.mul(reshape_9) + x_116 = reshape_9 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-05, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_124 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_126.permute(0, 3, 1, 2) + x_126 = None + reshape_10 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_128 = x_127.mul(reshape_10) + x_127 = reshape_10 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-05, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = torch._C._nn.linear( + x_135, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_135 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_137 = torch.nn.functional.dropout(x_136, 0.0, False, False) + x_136 = None + x_138 = x_137.permute(0, 3, 1, 2) + x_137 = None + reshape_11 = l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_ = ( + None + ) + x_139 = x_138.mul(reshape_11) + x_138 = reshape_11 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-05, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = torch._C._nn.linear( + x_146, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_146 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_148 = torch.nn.functional.dropout(x_147, 0.0, False, False) + x_147 = None + x_149 = x_148.permute(0, 3, 1, 2) + x_148 = None + reshape_12 = l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_ = ( + None + ) + x_150 = x_149.mul(reshape_12) + x_149 = reshape_12 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-05, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_159.permute(0, 3, 1, 2) + x_159 = None + reshape_13 = l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_ = ( + None + ) + x_161 = x_160.mul(reshape_13) + x_160 = reshape_13 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-05, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_169 = torch._C._nn.linear( + x_168, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_168 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_170 = torch.nn.functional.dropout(x_169, 0.0, False, False) + x_169 = None + x_171 = x_170.permute(0, 3, 1, 2) + x_170 = None + reshape_14 = l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_ = ( + None + ) + x_172 = x_171.mul(reshape_14) + x_171 = reshape_14 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_175 = x_174.permute(0, 2, 3, 1) + x_174 = None + x_176 = torch.nn.functional.layer_norm( + x_175, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-05, + ) + x_175 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_176 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_178 = torch._C._nn.gelu(x_177) + x_177 = None + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_180 = torch._C._nn.linear( + x_179, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_179 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_181 = torch.nn.functional.dropout(x_180, 0.0, False, False) + x_180 = None + x_182 = x_181.permute(0, 3, 1, 2) + x_181 = None + reshape_15 = l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_ = ( + None + ) + x_183 = x_182.mul(reshape_15) + x_182 = reshape_15 = None + x_184 = x_183 + x_173 + x_183 = x_173 = None + x_185 = torch.conv2d( + x_184, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = (None) + x_186 = x_185.permute(0, 2, 3, 1) + x_185 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_, + 1e-05, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = (None) + x_188 = torch._C._nn.linear( + x_187, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_187 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_189 = torch._C._nn.gelu(x_188) + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = torch._C._nn.linear( + x_190, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_190 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_192 = torch.nn.functional.dropout(x_191, 0.0, False, False) + x_191 = None + x_193 = x_192.permute(0, 3, 1, 2) + x_192 = None + reshape_16 = l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_ = ( + None + ) + x_194 = x_193.mul(reshape_16) + x_193 = reshape_16 = None + x_195 = x_194 + x_184 + x_194 = x_184 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = (None) + x_197 = x_196.permute(0, 2, 3, 1) + x_196 = None + x_198 = torch.nn.functional.layer_norm( + x_197, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_, + 1e-05, + ) + x_197 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = (None) + x_199 = torch._C._nn.linear( + x_198, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_198 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_200 = torch._C._nn.gelu(x_199) + x_199 = None + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_202 = torch._C._nn.linear( + x_201, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_201 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_203 = torch.nn.functional.dropout(x_202, 0.0, False, False) + x_202 = None + x_204 = x_203.permute(0, 3, 1, 2) + x_203 = None + reshape_17 = l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_ = ( + None + ) + x_205 = x_204.mul(reshape_17) + x_204 = reshape_17 = None + x_206 = x_205 + x_195 + x_205 = x_195 = None + x_207 = torch.conv2d( + x_206, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = (None) + x_208 = x_207.permute(0, 2, 3, 1) + x_207 = None + x_209 = torch.nn.functional.layer_norm( + x_208, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_, + 1e-05, + ) + x_208 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = (None) + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_209 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_211 = torch._C._nn.gelu(x_210) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_213 = torch._C._nn.linear( + x_212, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_212 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_214 = torch.nn.functional.dropout(x_213, 0.0, False, False) + x_213 = None + x_215 = x_214.permute(0, 3, 1, 2) + x_214 = None + reshape_18 = l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_ = ( + None + ) + x_216 = x_215.mul(reshape_18) + x_215 = reshape_18 = None + x_217 = x_216 + x_206 + x_216 = x_206 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = (None) + x_219 = x_218.permute(0, 2, 3, 1) + x_218 = None + x_220 = torch.nn.functional.layer_norm( + x_219, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_, + 1e-05, + ) + x_219 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = (None) + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_220 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_222 = torch._C._nn.gelu(x_221) + x_221 = None + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_224 = torch._C._nn.linear( + x_223, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_223 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = x_225.permute(0, 3, 1, 2) + x_225 = None + reshape_19 = l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_ = ( + None + ) + x_227 = x_226.mul(reshape_19) + x_226 = reshape_19 = None + x_228 = x_227 + x_217 + x_227 = x_217 = None + x_229 = torch.conv2d( + x_228, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = (None) + x_230 = x_229.permute(0, 2, 3, 1) + x_229 = None + x_231 = torch.nn.functional.layer_norm( + x_230, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_, + 1e-05, + ) + x_230 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = (None) + x_232 = torch._C._nn.linear( + x_231, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_231 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_233 = torch._C._nn.gelu(x_232) + x_232 = None + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_235 = torch._C._nn.linear( + x_234, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_234 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_236 = torch.nn.functional.dropout(x_235, 0.0, False, False) + x_235 = None + x_237 = x_236.permute(0, 3, 1, 2) + x_236 = None + reshape_20 = l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_ = ( + None + ) + x_238 = x_237.mul(reshape_20) + x_237 = reshape_20 = None + x_239 = x_238 + x_228 + x_238 = x_228 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = (None) + x_241 = x_240.permute(0, 2, 3, 1) + x_240 = None + x_242 = torch.nn.functional.layer_norm( + x_241, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_, + 1e-05, + ) + x_241 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = (None) + x_243 = torch._C._nn.linear( + x_242, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_242 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_244 = torch._C._nn.gelu(x_243) + x_243 = None + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_246 = torch._C._nn.linear( + x_245, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_245 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_247 = torch.nn.functional.dropout(x_246, 0.0, False, False) + x_246 = None + x_248 = x_247.permute(0, 3, 1, 2) + x_247 = None + reshape_21 = l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_ = ( + None + ) + x_249 = x_248.mul(reshape_21) + x_248 = reshape_21 = None + x_250 = x_249 + x_239 + x_249 = x_239 = None + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = (None) + x_252 = x_251.permute(0, 2, 3, 1) + x_251 = None + x_253 = torch.nn.functional.layer_norm( + x_252, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_, + 1e-05, + ) + x_252 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = (None) + x_254 = torch._C._nn.linear( + x_253, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_253 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_255 = torch._C._nn.gelu(x_254) + x_254 = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_257 = torch._C._nn.linear( + x_256, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_256 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_258 = torch.nn.functional.dropout(x_257, 0.0, False, False) + x_257 = None + x_259 = x_258.permute(0, 3, 1, 2) + x_258 = None + reshape_22 = l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_ = ( + None + ) + x_260 = x_259.mul(reshape_22) + x_259 = reshape_22 = None + x_261 = x_260 + x_250 + x_260 = x_250 = None + x_262 = torch.conv2d( + x_261, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = (None) + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_, + 1e-05, + ) + x_263 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = (None) + x_265 = torch._C._nn.linear( + x_264, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_264 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_266 = torch._C._nn.gelu(x_265) + x_265 = None + x_267 = torch.nn.functional.dropout(x_266, 0.0, False, False) + x_266 = None + x_268 = torch._C._nn.linear( + x_267, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_267 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_269 = torch.nn.functional.dropout(x_268, 0.0, False, False) + x_268 = None + x_270 = x_269.permute(0, 3, 1, 2) + x_269 = None + reshape_23 = l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_ = ( + None + ) + x_271 = x_270.mul(reshape_23) + x_270 = reshape_23 = None + x_272 = x_271 + x_261 + x_271 = x_261 = None + x_273 = torch.conv2d( + x_272, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = (None) + x_274 = x_273.permute(0, 2, 3, 1) + x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_, + 1e-05, + ) + x_274 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = (None) + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_277 = torch._C._nn.gelu(x_276) + x_276 = None + x_278 = torch.nn.functional.dropout(x_277, 0.0, False, False) + x_277 = None + x_279 = torch._C._nn.linear( + x_278, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_278 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_280 = torch.nn.functional.dropout(x_279, 0.0, False, False) + x_279 = None + x_281 = x_280.permute(0, 3, 1, 2) + x_280 = None + reshape_24 = l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_ = ( + None + ) + x_282 = x_281.mul(reshape_24) + x_281 = reshape_24 = None + x_283 = x_282 + x_272 + x_282 = x_272 = None + x_284 = torch.conv2d( + x_283, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = (None) + x_285 = x_284.permute(0, 2, 3, 1) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_, + 1e-05, + ) + x_285 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = (None) + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_288 = torch._C._nn.gelu(x_287) + x_287 = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + x_290 = torch._C._nn.linear( + x_289, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_289 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_291 = torch.nn.functional.dropout(x_290, 0.0, False, False) + x_290 = None + x_292 = x_291.permute(0, 3, 1, 2) + x_291 = None + reshape_25 = l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_ = ( + None + ) + x_293 = x_292.mul(reshape_25) + x_292 = reshape_25 = None + x_294 = x_293 + x_283 + x_293 = x_283 = None + x_295 = torch.conv2d( + x_294, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = (None) + x_296 = x_295.permute(0, 2, 3, 1) + x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_, + 1e-05, + ) + x_296 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = (None) + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_299 = torch._C._nn.gelu(x_298) + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_301 = torch._C._nn.linear( + x_300, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_300 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_302 = torch.nn.functional.dropout(x_301, 0.0, False, False) + x_301 = None + x_303 = x_302.permute(0, 3, 1, 2) + x_302 = None + reshape_26 = l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_ = ( + None + ) + x_304 = x_303.mul(reshape_26) + x_303 = reshape_26 = None + x_305 = x_304 + x_294 + x_304 = x_294 = None + x_306 = torch.conv2d( + x_305, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = (None) + x_307 = x_306.permute(0, 2, 3, 1) + x_306 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_, + 1e-05, + ) + x_307 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = (None) + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_310 = torch._C._nn.gelu(x_309) + x_309 = None + x_311 = torch.nn.functional.dropout(x_310, 0.0, False, False) + x_310 = None + x_312 = torch._C._nn.linear( + x_311, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_311 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_313 = torch.nn.functional.dropout(x_312, 0.0, False, False) + x_312 = None + x_314 = x_313.permute(0, 3, 1, 2) + x_313 = None + reshape_27 = l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_ = ( + None + ) + x_315 = x_314.mul(reshape_27) + x_314 = reshape_27 = None + x_316 = x_315 + x_305 + x_315 = x_305 = None + x_317 = torch.conv2d( + x_316, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = (None) + x_318 = x_317.permute(0, 2, 3, 1) + x_317 = None + x_319 = torch.nn.functional.layer_norm( + x_318, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_, + 1e-05, + ) + x_318 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = (None) + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_321 = torch._C._nn.gelu(x_320) + x_320 = None + x_322 = torch.nn.functional.dropout(x_321, 0.0, False, False) + x_321 = None + x_323 = torch._C._nn.linear( + x_322, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_322 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_324 = torch.nn.functional.dropout(x_323, 0.0, False, False) + x_323 = None + x_325 = x_324.permute(0, 3, 1, 2) + x_324 = None + reshape_28 = l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_ = ( + None + ) + x_326 = x_325.mul(reshape_28) + x_325 = reshape_28 = None + x_327 = x_326 + x_316 + x_326 = x_316 = None + x_328 = torch.conv2d( + x_327, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = (None) + x_329 = x_328.permute(0, 2, 3, 1) + x_328 = None + x_330 = torch.nn.functional.layer_norm( + x_329, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_, + 1e-05, + ) + x_329 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = (None) + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_332 = torch._C._nn.gelu(x_331) + x_331 = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_334 = torch._C._nn.linear( + x_333, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_333 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_335 = torch.nn.functional.dropout(x_334, 0.0, False, False) + x_334 = None + x_336 = x_335.permute(0, 3, 1, 2) + x_335 = None + reshape_29 = l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_ = ( + None + ) + x_337 = x_336.mul(reshape_29) + x_336 = reshape_29 = None + x_338 = x_337 + x_327 + x_337 = x_327 = None + x_339 = torch.conv2d( + x_338, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = (None) + x_340 = x_339.permute(0, 2, 3, 1) + x_339 = None + x_341 = torch.nn.functional.layer_norm( + x_340, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_, + 1e-05, + ) + x_340 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = (None) + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_343 = torch._C._nn.gelu(x_342) + x_342 = None + x_344 = torch.nn.functional.dropout(x_343, 0.0, False, False) + x_343 = None + x_345 = torch._C._nn.linear( + x_344, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_344 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_346 = torch.nn.functional.dropout(x_345, 0.0, False, False) + x_345 = None + x_347 = x_346.permute(0, 3, 1, 2) + x_346 = None + reshape_30 = l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_ = ( + None + ) + x_348 = x_347.mul(reshape_30) + x_347 = reshape_30 = None + x_349 = x_348 + x_338 + x_348 = x_338 = None + x_350 = torch.conv2d( + x_349, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = (None) + x_351 = x_350.permute(0, 2, 3, 1) + x_350 = None + x_352 = torch.nn.functional.layer_norm( + x_351, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_, + 1e-05, + ) + x_351 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = (None) + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_354 = torch._C._nn.gelu(x_353) + x_353 = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_356 = torch._C._nn.linear( + x_355, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_355 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_357 = torch.nn.functional.dropout(x_356, 0.0, False, False) + x_356 = None + x_358 = x_357.permute(0, 3, 1, 2) + x_357 = None + reshape_31 = l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_ = ( + None + ) + x_359 = x_358.mul(reshape_31) + x_358 = reshape_31 = None + x_360 = x_359 + x_349 + x_359 = x_349 = None + x_361 = torch.conv2d( + x_360, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = (None) + x_362 = x_361.permute(0, 2, 3, 1) + x_361 = None + x_363 = torch.nn.functional.layer_norm( + x_362, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_, + 1e-05, + ) + x_362 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = (None) + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_365 = torch._C._nn.gelu(x_364) + x_364 = None + x_366 = torch.nn.functional.dropout(x_365, 0.0, False, False) + x_365 = None + x_367 = torch._C._nn.linear( + x_366, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_366 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_368 = torch.nn.functional.dropout(x_367, 0.0, False, False) + x_367 = None + x_369 = x_368.permute(0, 3, 1, 2) + x_368 = None + reshape_32 = l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_ = ( + None + ) + x_370 = x_369.mul(reshape_32) + x_369 = reshape_32 = None + x_371 = x_370 + x_360 + x_370 = x_360 = None + x_372 = torch.conv2d( + x_371, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = (None) + x_373 = x_372.permute(0, 2, 3, 1) + x_372 = None + x_374 = torch.nn.functional.layer_norm( + x_373, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_, + 1e-05, + ) + x_373 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = (None) + x_375 = torch._C._nn.linear( + x_374, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + x_374 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_376 = torch._C._nn.gelu(x_375) + x_375 = None + x_377 = torch.nn.functional.dropout(x_376, 0.0, False, False) + x_376 = None + x_378 = torch._C._nn.linear( + x_377, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_377 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_379 = torch.nn.functional.dropout(x_378, 0.0, False, False) + x_378 = None + x_380 = x_379.permute(0, 3, 1, 2) + x_379 = None + reshape_33 = l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_ = ( + None + ) + x_381 = x_380.mul(reshape_33) + x_380 = reshape_33 = None + x_382 = x_381 + x_371 + x_381 = x_371 = None + x_383 = torch.conv2d( + x_382, + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_bias_ = (None) + x_384 = x_383.permute(0, 2, 3, 1) + x_383 = None + x_385 = torch.nn.functional.layer_norm( + x_384, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_bias_, + 1e-05, + ) + x_384 = l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_bias_ = (None) + x_386 = torch._C._nn.linear( + x_385, + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_, + ) + x_385 = l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_387 = torch._C._nn.gelu(x_386) + x_386 = None + x_388 = torch.nn.functional.dropout(x_387, 0.0, False, False) + x_387 = None + x_389 = torch._C._nn.linear( + x_388, + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_, + ) + x_388 = l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_390 = torch.nn.functional.dropout(x_389, 0.0, False, False) + x_389 = None + x_391 = x_390.permute(0, 3, 1, 2) + x_390 = None + reshape_34 = l_self_modules_stages_modules_2_modules_blocks_modules_27_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_27_parameters_gamma_ = ( + None + ) + x_392 = x_391.mul(reshape_34) + x_391 = reshape_34 = None + x_393 = x_392 + x_382 + x_392 = x_382 = None + x_394 = torch.conv2d( + x_393, + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_bias_ = (None) + x_395 = x_394.permute(0, 2, 3, 1) + x_394 = None + x_396 = torch.nn.functional.layer_norm( + x_395, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_bias_, + 1e-05, + ) + x_395 = l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_bias_ = (None) + x_397 = torch._C._nn.linear( + x_396, + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_, + ) + x_396 = l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_398 = torch._C._nn.gelu(x_397) + x_397 = None + x_399 = torch.nn.functional.dropout(x_398, 0.0, False, False) + x_398 = None + x_400 = torch._C._nn.linear( + x_399, + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_, + ) + x_399 = l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_401 = torch.nn.functional.dropout(x_400, 0.0, False, False) + x_400 = None + x_402 = x_401.permute(0, 3, 1, 2) + x_401 = None + reshape_35 = l_self_modules_stages_modules_2_modules_blocks_modules_28_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_28_parameters_gamma_ = ( + None + ) + x_403 = x_402.mul(reshape_35) + x_402 = reshape_35 = None + x_404 = x_403 + x_393 + x_403 = x_393 = None + x_405 = torch.conv2d( + x_404, + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_bias_ = (None) + x_406 = x_405.permute(0, 2, 3, 1) + x_405 = None + x_407 = torch.nn.functional.layer_norm( + x_406, + (1536,), + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_bias_, + 1e-05, + ) + x_406 = l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_bias_ = (None) + x_408 = torch._C._nn.linear( + x_407, + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_, + ) + x_407 = l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_409 = torch._C._nn.gelu(x_408) + x_408 = None + x_410 = torch.nn.functional.dropout(x_409, 0.0, False, False) + x_409 = None + x_411 = torch._C._nn.linear( + x_410, + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_, + ) + x_410 = l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_412 = torch.nn.functional.dropout(x_411, 0.0, False, False) + x_411 = None + x_413 = x_412.permute(0, 3, 1, 2) + x_412 = None + reshape_36 = l_self_modules_stages_modules_2_modules_blocks_modules_29_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_29_parameters_gamma_ = ( + None + ) + x_414 = x_413.mul(reshape_36) + x_413 = reshape_36 = None + x_415 = x_414 + x_404 + x_414 = x_404 = None + x_416 = x_415.permute(0, 2, 3, 1) + x_415 = None + x_417 = torch.nn.functional.layer_norm( + x_416, + (1536,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-05, + ) + x_416 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_418 = x_417.permute(0, 3, 1, 2) + x_417 = None + input_4 = torch.conv2d( + x_418, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_418 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_419 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 3072, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_420 = x_419.permute(0, 2, 3, 1) + x_419 = None + x_421 = torch.nn.functional.layer_norm( + x_420, + (3072,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-05, + ) + x_420 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_422 = torch._C._nn.linear( + x_421, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_421 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_423 = torch._C._nn.gelu(x_422) + x_422 = None + x_424 = torch.nn.functional.dropout(x_423, 0.0, False, False) + x_423 = None + x_425 = torch._C._nn.linear( + x_424, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_424 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_426 = torch.nn.functional.dropout(x_425, 0.0, False, False) + x_425 = None + x_427 = x_426.permute(0, 3, 1, 2) + x_426 = None + reshape_37 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_428 = x_427.mul(reshape_37) + x_427 = reshape_37 = None + x_429 = x_428 + input_4 + x_428 = input_4 = None + x_430 = torch.conv2d( + x_429, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 3072, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_431 = x_430.permute(0, 2, 3, 1) + x_430 = None + x_432 = torch.nn.functional.layer_norm( + x_431, + (3072,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-05, + ) + x_431 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_433 = torch._C._nn.linear( + x_432, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_432 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_434 = torch._C._nn.gelu(x_433) + x_433 = None + x_435 = torch.nn.functional.dropout(x_434, 0.0, False, False) + x_434 = None + x_436 = torch._C._nn.linear( + x_435, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_435 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_437 = torch.nn.functional.dropout(x_436, 0.0, False, False) + x_436 = None + x_438 = x_437.permute(0, 3, 1, 2) + x_437 = None + reshape_38 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_439 = x_438.mul(reshape_38) + x_438 = reshape_38 = None + x_440 = x_439 + x_429 + x_439 = x_429 = None + x_441 = torch.conv2d( + x_440, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 3072, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_442 = x_441.permute(0, 2, 3, 1) + x_441 = None + x_443 = torch.nn.functional.layer_norm( + x_442, + (3072,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-05, + ) + x_442 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_444 = torch._C._nn.linear( + x_443, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_443 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_445 = torch._C._nn.gelu(x_444) + x_444 = None + x_446 = torch.nn.functional.dropout(x_445, 0.0, False, False) + x_445 = None + x_447 = torch._C._nn.linear( + x_446, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_446 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_448 = torch.nn.functional.dropout(x_447, 0.0, False, False) + x_447 = None + x_449 = x_448.permute(0, 3, 1, 2) + x_448 = None + reshape_39 = l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_450 = x_449.mul(reshape_39) + x_449 = reshape_39 = None + x_451 = x_450 + x_440 + x_450 = x_440 = None + x_452 = torch.nn.functional.adaptive_avg_pool2d(x_451, 1) + x_451 = None + x_453 = x_452.permute(0, 2, 3, 1) + x_452 = None + x_454 = torch.nn.functional.layer_norm( + x_453, + (3072,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-05, + ) + x_453 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_455 = x_454.permute(0, 3, 1, 2) + x_454 = None + x_456 = x_455.flatten(1, -1) + x_455 = None + x_457 = torch.nn.functional.dropout(x_456, 0.0, False, False) + x_456 = None + x_458 = torch._C._nn.linear( + x_457, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_457 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_458,) diff --git a/samples/timm/convnext_xxlarge.clip_laion2b_rewind/weight_meta.py b/samples/timm/convnext_xxlarge.clip_laion2b_rewind/weight_meta.py new file mode 100644 index 0000000000..4caff7eff8 --- /dev/null +++ b/samples/timm/convnext_xxlarge.clip_laion2b_rewind/weight_meta.py @@ -0,0 +1,3830 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [384, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.230 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [768, 384, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [1536, 768, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_27_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_27_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_28_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_28_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_29_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_29_parameters_gamma_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [3072, 1536, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [3072, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [12288, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [12288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [3072, 12288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [3072, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [12288, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [12288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [3072, 12288] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [3072, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [12288, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [12288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [3072, 12288] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_parameters_gamma_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1024, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/graph_hash.txt b/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/graph_hash.txt new file mode 100644 index 0000000000..3074310a12 --- /dev/null +++ b/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/graph_hash.txt @@ -0,0 +1 @@ +314f8e9539d94cb83851ab0d0b0288d5a96b09df0a76049d52406bbce1da79e5 \ No newline at end of file diff --git a/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/graph_net.json b/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/input_meta.py b/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/input_tensor_constraints.py b/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/model.py b/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/model.py new file mode 100644 index 0000000000..551296ee7b --- /dev/null +++ b/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/model.py @@ -0,0 +1,983 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + v = torch.var(x, dim=(-1,), keepdim=True) + add = v + 1e-06 + v = None + rsqrt = torch.rsqrt(add) + add = None + x_1 = x * rsqrt + x = rsqrt = None + x_2 = x_1 * l_self_modules_stem_modules_1_parameters_weight_ + x_1 = l_self_modules_stem_modules_1_parameters_weight_ = None + x_3 = x_2.permute(0, 3, 1, 2) + x_2 = None + x_4 = torch.conv2d( + x_3, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 32, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_5 = x_4.permute(0, 2, 3, 1) + x_4 = None + v_1 = torch.var(x_5, dim=(-1,), keepdim=True) + add_1 = v_1 + 1e-06 + v_1 = None + rsqrt_1 = torch.rsqrt(add_1) + add_1 = None + x_6 = x_5 * rsqrt_1 + x_5 = rsqrt_1 = None + x_7 = ( + x_6 + * l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = (None) + x_8 = x_7.permute(0, 3, 1, 2) + x_7 = None + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_10 = torch._C._nn.gelu(x_9) + x_9 = None + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = torch.conv2d( + x_11, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_11 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_13 = torch.nn.functional.dropout(x_12, 0.0, False, False) + x_12 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_14 = x_13.mul(reshape) + x_13 = reshape = None + x_15 = x_14 + x_3 + x_14 = x_3 = None + x_16 = torch.conv2d( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 32, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_17 = x_16.permute(0, 2, 3, 1) + x_16 = None + v_2 = torch.var(x_17, dim=(-1,), keepdim=True) + add_3 = v_2 + 1e-06 + v_2 = None + rsqrt_2 = torch.rsqrt(add_3) + add_3 = None + x_18 = x_17 * rsqrt_2 + x_17 = rsqrt_2 = None + x_19 = ( + x_18 + * l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + ) + x_18 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = (None) + x_20 = x_19.permute(0, 3, 1, 2) + x_19 = None + x_21 = torch.conv2d( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_22 = torch._C._nn.gelu(x_21) + x_21 = None + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = torch.conv2d( + x_23, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_25 = torch.nn.functional.dropout(x_24, 0.0, False, False) + x_24 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_26 = x_25.mul(reshape_1) + x_25 = reshape_1 = None + x_27 = x_26 + x_15 + x_26 = x_15 = None + x_28 = x_27.permute(0, 2, 3, 1) + x_27 = None + v_3 = torch.var(x_28, dim=(-1,), keepdim=True) + add_5 = v_3 + 1e-06 + v_3 = None + rsqrt_3 = torch.rsqrt(add_5) + add_5 = None + x_29 = x_28 * rsqrt_3 + x_28 = rsqrt_3 = None + x_30 = ( + x_29 + * l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + ) + x_29 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + input_2 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_32 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 64, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_33 = x_32.permute(0, 2, 3, 1) + x_32 = None + v_4 = torch.var(x_33, dim=(-1,), keepdim=True) + add_6 = v_4 + 1e-06 + v_4 = None + rsqrt_4 = torch.rsqrt(add_6) + add_6 = None + x_34 = x_33 * rsqrt_4 + x_33 = rsqrt_4 = None + x_35 = ( + x_34 + * l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + ) + x_34 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = (None) + x_36 = x_35.permute(0, 3, 1, 2) + x_35 = None + x_37 = torch.conv2d( + x_36, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_36 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_38 = torch._C._nn.gelu(x_37) + x_37 = None + x_39 = torch.nn.functional.dropout(x_38, 0.0, False, False) + x_38 = None + x_40 = torch.conv2d( + x_39, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_39 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_41 = torch.nn.functional.dropout(x_40, 0.0, False, False) + x_40 = None + reshape_2 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_42 = x_41.mul(reshape_2) + x_41 = reshape_2 = None + x_43 = x_42 + input_2 + x_42 = input_2 = None + x_44 = torch.conv2d( + x_43, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 64, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_45 = x_44.permute(0, 2, 3, 1) + x_44 = None + v_5 = torch.var(x_45, dim=(-1,), keepdim=True) + add_8 = v_5 + 1e-06 + v_5 = None + rsqrt_5 = torch.rsqrt(add_8) + add_8 = None + x_46 = x_45 * rsqrt_5 + x_45 = rsqrt_5 = None + x_47 = ( + x_46 + * l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = (None) + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_50 = torch._C._nn.gelu(x_49) + x_49 = None + x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_54 = x_53.mul(reshape_3) + x_53 = reshape_3 = None + x_55 = x_54 + x_43 + x_54 = x_43 = None + x_56 = x_55.permute(0, 2, 3, 1) + x_55 = None + v_6 = torch.var(x_56, dim=(-1,), keepdim=True) + add_10 = v_6 + 1e-06 + v_6 = None + rsqrt_6 = torch.rsqrt(add_10) + add_10 = None + x_57 = x_56 * rsqrt_6 + x_56 = rsqrt_6 = None + x_58 = ( + x_57 + * l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + ) + x_57 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = (None) + x_59 = x_58.permute(0, 3, 1, 2) + x_58 = None + input_3 = torch.conv2d( + x_59, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_59 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_60 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_61 = x_60.permute(0, 2, 3, 1) + x_60 = None + v_7 = torch.var(x_61, dim=(-1,), keepdim=True) + add_11 = v_7 + 1e-06 + v_7 = None + rsqrt_7 = torch.rsqrt(add_11) + add_11 = None + x_62 = x_61 * rsqrt_7 + x_61 = rsqrt_7 = None + x_63 = ( + x_62 + * l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + ) + x_62 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = (None) + x_64 = x_63.permute(0, 3, 1, 2) + x_63 = None + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_66 = torch._C._nn.gelu(x_65) + x_65 = None + x_67 = torch.nn.functional.dropout(x_66, 0.0, False, False) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + reshape_4 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_70 = x_69.mul(reshape_4) + x_69 = reshape_4 = None + x_71 = x_70 + input_3 + x_70 = input_3 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_73 = x_72.permute(0, 2, 3, 1) + x_72 = None + v_8 = torch.var(x_73, dim=(-1,), keepdim=True) + add_13 = v_8 + 1e-06 + v_8 = None + rsqrt_8 = torch.rsqrt(add_13) + add_13 = None + x_74 = x_73 * rsqrt_8 + x_73 = rsqrt_8 = None + x_75 = ( + x_74 + * l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + ) + x_74 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = (None) + x_76 = x_75.permute(0, 3, 1, 2) + x_75 = None + x_77 = torch.conv2d( + x_76, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_78 = torch._C._nn.gelu(x_77) + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch.conv2d( + x_79, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_79 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + reshape_5 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_82 = x_81.mul(reshape_5) + x_81 = reshape_5 = None + x_83 = x_82 + x_71 + x_82 = x_71 = None + x_84 = torch.conv2d( + x_83, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_85 = x_84.permute(0, 2, 3, 1) + x_84 = None + v_9 = torch.var(x_85, dim=(-1,), keepdim=True) + add_15 = v_9 + 1e-06 + v_9 = None + rsqrt_9 = torch.rsqrt(add_15) + add_15 = None + x_86 = x_85 * rsqrt_9 + x_85 = rsqrt_9 = None + x_87 = ( + x_86 + * l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + ) + x_86 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = (None) + x_88 = x_87.permute(0, 3, 1, 2) + x_87 = None + x_89 = torch.conv2d( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = torch.conv2d( + x_91, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_91 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_93 = torch.nn.functional.dropout(x_92, 0.0, False, False) + x_92 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_94 = x_93.mul(reshape_6) + x_93 = reshape_6 = None + x_95 = x_94 + x_83 + x_94 = x_83 = None + x_96 = torch.conv2d( + x_95, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_97 = x_96.permute(0, 2, 3, 1) + x_96 = None + v_10 = torch.var(x_97, dim=(-1,), keepdim=True) + add_17 = v_10 + 1e-06 + v_10 = None + rsqrt_10 = torch.rsqrt(add_17) + add_17 = None + x_98 = x_97 * rsqrt_10 + x_97 = rsqrt_10 = None + x_99 = ( + x_98 + * l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_7) + x_105 = reshape_7 = None + x_107 = x_106 + x_95 + x_106 = x_95 = None + x_108 = x_107.permute(0, 2, 3, 1) + x_107 = None + v_11 = torch.var(x_108, dim=(-1,), keepdim=True) + add_19 = v_11 + 1e-06 + v_11 = None + rsqrt_11 = torch.rsqrt(add_19) + add_19 = None + x_109 = x_108 * rsqrt_11 + x_108 = rsqrt_11 = None + x_110 = ( + x_109 + * l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + ) + x_109 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + input_4 = torch.conv2d( + x_111, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_112 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_113 = x_112.permute(0, 2, 3, 1) + x_112 = None + v_12 = torch.var(x_113, dim=(-1,), keepdim=True) + add_20 = v_12 + 1e-06 + v_12 = None + rsqrt_12 = torch.rsqrt(add_20) + add_20 = None + x_114 = x_113 * rsqrt_12 + x_113 = rsqrt_12 = None + x_115 = ( + x_114 + * l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + ) + x_114 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = (None) + x_116 = x_115.permute(0, 3, 1, 2) + x_115 = None + x_117 = torch.conv2d( + x_116, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_116 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_118 = torch._C._nn.gelu(x_117) + x_117 = None + x_119 = torch.nn.functional.dropout(x_118, 0.0, False, False) + x_118 = None + x_120 = torch.conv2d( + x_119, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_119 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_121 = torch.nn.functional.dropout(x_120, 0.0, False, False) + x_120 = None + reshape_8 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_122 = x_121.mul(reshape_8) + x_121 = reshape_8 = None + x_123 = x_122 + input_4 + x_122 = input_4 = None + x_124 = torch.conv2d( + x_123, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_125 = x_124.permute(0, 2, 3, 1) + x_124 = None + v_13 = torch.var(x_125, dim=(-1,), keepdim=True) + add_22 = v_13 + 1e-06 + v_13 = None + rsqrt_13 = torch.rsqrt(add_22) + add_22 = None + x_126 = x_125 * rsqrt_13 + x_125 = rsqrt_13 = None + x_127 = ( + x_126 + * l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + ) + x_126 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = (None) + x_128 = x_127.permute(0, 3, 1, 2) + x_127 = None + x_129 = torch.conv2d( + x_128, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_128 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_130 = torch._C._nn.gelu(x_129) + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch.conv2d( + x_131, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_131 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + reshape_9 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_134 = x_133.mul(reshape_9) + x_133 = reshape_9 = None + x_135 = x_134 + x_123 + x_134 = x_123 = None + x_136 = torch.nn.functional.adaptive_avg_pool2d(x_135, 1) + x_135 = None + x_137 = x_136.permute(0, 2, 3, 1) + x_136 = None + v_14 = torch.var(x_137, dim=(-1,), keepdim=True) + add_24 = v_14 + 1e-06 + v_14 = None + rsqrt_14 = torch.rsqrt(add_24) + add_24 = None + x_138 = x_137 * rsqrt_14 + x_137 = rsqrt_14 = None + x_139 = x_138 * l_self_modules_head_modules_norm_parameters_weight_ + x_138 = l_self_modules_head_modules_norm_parameters_weight_ = None + x_140 = x_139.permute(0, 3, 1, 2) + x_139 = None + x_141 = x_140.flatten(1, -1) + x_140 = None + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = torch._C._nn.linear( + x_142, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_142 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_143,) diff --git a/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/weight_meta.py b/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/weight_meta.py new file mode 100644 index 0000000000..1cab743f15 --- /dev/null +++ b/samples/timm/convnext_zepto_rms.ra4_e3600_r224_in1k/weight_meta.py @@ -0,0 +1,974 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [32, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.226 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [32, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [128, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [32, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [32, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [128, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [32, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [64, 32, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [64, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [64, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [128, 64, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [256, 128, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/graph_hash.txt b/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/graph_hash.txt new file mode 100644 index 0000000000..b733231042 --- /dev/null +++ b/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/graph_hash.txt @@ -0,0 +1 @@ +8c985b6584dd5c5878afa74dc799f3092db39d0e36501155abab738dbc8bae2d \ No newline at end of file diff --git a/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/graph_net.json b/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/input_meta.py b/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/input_tensor_constraints.py b/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/model.py b/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/model.py new file mode 100644 index 0000000000..4282f60588 --- /dev/null +++ b/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/model.py @@ -0,0 +1,1005 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_3_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_2_parameters_weight_ = ( + L_self_modules_stem_modules_2_parameters_weight_ + ) + l_self_modules_stem_modules_2_parameters_bias_ = ( + L_self_modules_stem_modules_2_parameters_bias_ + ) + l_self_modules_stem_modules_3_parameters_weight_ = ( + L_self_modules_stem_modules_3_parameters_weight_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ + ) + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + input_2 = torch._C._nn.gelu(input_1) + input_1 = None + input_3 = torch.conv2d( + input_2, + l_self_modules_stem_modules_2_parameters_weight_, + l_self_modules_stem_modules_2_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_2 = ( + l_self_modules_stem_modules_2_parameters_weight_ + ) = l_self_modules_stem_modules_2_parameters_bias_ = None + x = input_3.permute(0, 2, 3, 1) + input_3 = None + v = torch.var(x, dim=(-1,), keepdim=True) + add = v + 1e-06 + v = None + rsqrt = torch.rsqrt(add) + add = None + x_1 = x * rsqrt + x = rsqrt = None + x_2 = x_1 * l_self_modules_stem_modules_3_parameters_weight_ + x_1 = l_self_modules_stem_modules_3_parameters_weight_ = None + x_3 = x_2.permute(0, 3, 1, 2) + x_2 = None + x_4 = torch.conv2d( + x_3, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 32, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_5 = x_4.permute(0, 2, 3, 1) + x_4 = None + v_1 = torch.var(x_5, dim=(-1,), keepdim=True) + add_1 = v_1 + 1e-06 + v_1 = None + rsqrt_1 = torch.rsqrt(add_1) + add_1 = None + x_6 = x_5 * rsqrt_1 + x_5 = rsqrt_1 = None + x_7 = ( + x_6 + * l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = (None) + x_8 = x_7.permute(0, 3, 1, 2) + x_7 = None + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_10 = torch._C._nn.gelu(x_9) + x_9 = None + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = torch.conv2d( + x_11, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_11 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_13 = torch.nn.functional.dropout(x_12, 0.0, False, False) + x_12 = None + reshape = l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_14 = x_13.mul(reshape) + x_13 = reshape = None + x_15 = x_14 + x_3 + x_14 = x_3 = None + x_16 = torch.conv2d( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 32, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_17 = x_16.permute(0, 2, 3, 1) + x_16 = None + v_2 = torch.var(x_17, dim=(-1,), keepdim=True) + add_3 = v_2 + 1e-06 + v_2 = None + rsqrt_2 = torch.rsqrt(add_3) + add_3 = None + x_18 = x_17 * rsqrt_2 + x_17 = rsqrt_2 = None + x_19 = ( + x_18 + * l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + ) + x_18 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = (None) + x_20 = x_19.permute(0, 3, 1, 2) + x_19 = None + x_21 = torch.conv2d( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_22 = torch._C._nn.gelu(x_21) + x_21 = None + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = torch.conv2d( + x_23, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_25 = torch.nn.functional.dropout(x_24, 0.0, False, False) + x_24 = None + reshape_1 = l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_26 = x_25.mul(reshape_1) + x_25 = reshape_1 = None + x_27 = x_26 + x_15 + x_26 = x_15 = None + x_28 = x_27.permute(0, 2, 3, 1) + x_27 = None + v_3 = torch.var(x_28, dim=(-1,), keepdim=True) + add_5 = v_3 + 1e-06 + v_3 = None + rsqrt_3 = torch.rsqrt(add_5) + add_5 = None + x_29 = x_28 * rsqrt_3 + x_28 = rsqrt_3 = None + x_30 = ( + x_29 + * l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + ) + x_29 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + input_4 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_32 = torch.conv2d( + input_4, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 64, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_33 = x_32.permute(0, 2, 3, 1) + x_32 = None + v_4 = torch.var(x_33, dim=(-1,), keepdim=True) + add_6 = v_4 + 1e-06 + v_4 = None + rsqrt_4 = torch.rsqrt(add_6) + add_6 = None + x_34 = x_33 * rsqrt_4 + x_33 = rsqrt_4 = None + x_35 = ( + x_34 + * l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + ) + x_34 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = (None) + x_36 = x_35.permute(0, 3, 1, 2) + x_35 = None + x_37 = torch.conv2d( + x_36, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_36 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_38 = torch._C._nn.gelu(x_37) + x_37 = None + x_39 = torch.nn.functional.dropout(x_38, 0.0, False, False) + x_38 = None + x_40 = torch.conv2d( + x_39, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_39 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_41 = torch.nn.functional.dropout(x_40, 0.0, False, False) + x_40 = None + reshape_2 = l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_42 = x_41.mul(reshape_2) + x_41 = reshape_2 = None + x_43 = x_42 + input_4 + x_42 = input_4 = None + x_44 = torch.conv2d( + x_43, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 64, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_45 = x_44.permute(0, 2, 3, 1) + x_44 = None + v_5 = torch.var(x_45, dim=(-1,), keepdim=True) + add_8 = v_5 + 1e-06 + v_5 = None + rsqrt_5 = torch.rsqrt(add_8) + add_8 = None + x_46 = x_45 * rsqrt_5 + x_45 = rsqrt_5 = None + x_47 = ( + x_46 + * l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = (None) + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_50 = torch._C._nn.gelu(x_49) + x_49 = None + x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + reshape_3 = l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_54 = x_53.mul(reshape_3) + x_53 = reshape_3 = None + x_55 = x_54 + x_43 + x_54 = x_43 = None + x_56 = x_55.permute(0, 2, 3, 1) + x_55 = None + v_6 = torch.var(x_56, dim=(-1,), keepdim=True) + add_10 = v_6 + 1e-06 + v_6 = None + rsqrt_6 = torch.rsqrt(add_10) + add_10 = None + x_57 = x_56 * rsqrt_6 + x_56 = rsqrt_6 = None + x_58 = ( + x_57 + * l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + ) + x_57 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = (None) + x_59 = x_58.permute(0, 3, 1, 2) + x_58 = None + input_5 = torch.conv2d( + x_59, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_59 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_60 = torch.conv2d( + input_5, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_61 = x_60.permute(0, 2, 3, 1) + x_60 = None + v_7 = torch.var(x_61, dim=(-1,), keepdim=True) + add_11 = v_7 + 1e-06 + v_7 = None + rsqrt_7 = torch.rsqrt(add_11) + add_11 = None + x_62 = x_61 * rsqrt_7 + x_61 = rsqrt_7 = None + x_63 = ( + x_62 + * l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + ) + x_62 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = (None) + x_64 = x_63.permute(0, 3, 1, 2) + x_63 = None + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_66 = torch._C._nn.gelu(x_65) + x_65 = None + x_67 = torch.nn.functional.dropout(x_66, 0.0, False, False) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + reshape_4 = l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_70 = x_69.mul(reshape_4) + x_69 = reshape_4 = None + x_71 = x_70 + input_5 + x_70 = input_5 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_73 = x_72.permute(0, 2, 3, 1) + x_72 = None + v_8 = torch.var(x_73, dim=(-1,), keepdim=True) + add_13 = v_8 + 1e-06 + v_8 = None + rsqrt_8 = torch.rsqrt(add_13) + add_13 = None + x_74 = x_73 * rsqrt_8 + x_73 = rsqrt_8 = None + x_75 = ( + x_74 + * l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + ) + x_74 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = (None) + x_76 = x_75.permute(0, 3, 1, 2) + x_75 = None + x_77 = torch.conv2d( + x_76, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_78 = torch._C._nn.gelu(x_77) + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch.conv2d( + x_79, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_79 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + reshape_5 = l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_82 = x_81.mul(reshape_5) + x_81 = reshape_5 = None + x_83 = x_82 + x_71 + x_82 = x_71 = None + x_84 = torch.conv2d( + x_83, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_85 = x_84.permute(0, 2, 3, 1) + x_84 = None + v_9 = torch.var(x_85, dim=(-1,), keepdim=True) + add_15 = v_9 + 1e-06 + v_9 = None + rsqrt_9 = torch.rsqrt(add_15) + add_15 = None + x_86 = x_85 * rsqrt_9 + x_85 = rsqrt_9 = None + x_87 = ( + x_86 + * l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + ) + x_86 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = (None) + x_88 = x_87.permute(0, 3, 1, 2) + x_87 = None + x_89 = torch.conv2d( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = torch.conv2d( + x_91, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_91 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_93 = torch.nn.functional.dropout(x_92, 0.0, False, False) + x_92 = None + reshape_6 = l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_ = ( + None + ) + x_94 = x_93.mul(reshape_6) + x_93 = reshape_6 = None + x_95 = x_94 + x_83 + x_94 = x_83 = None + x_96 = torch.conv2d( + x_95, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_97 = x_96.permute(0, 2, 3, 1) + x_96 = None + v_10 = torch.var(x_97, dim=(-1,), keepdim=True) + add_17 = v_10 + 1e-06 + v_10 = None + rsqrt_10 = torch.rsqrt(add_17) + add_17 = None + x_98 = x_97 * rsqrt_10 + x_97 = rsqrt_10 = None + x_99 = ( + x_98 + * l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + reshape_7 = l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_ = ( + None + ) + x_106 = x_105.mul(reshape_7) + x_105 = reshape_7 = None + x_107 = x_106 + x_95 + x_106 = x_95 = None + x_108 = x_107.permute(0, 2, 3, 1) + x_107 = None + v_11 = torch.var(x_108, dim=(-1,), keepdim=True) + add_19 = v_11 + 1e-06 + v_11 = None + rsqrt_11 = torch.rsqrt(add_19) + add_19 = None + x_109 = x_108 * rsqrt_11 + x_108 = rsqrt_11 = None + x_110 = ( + x_109 + * l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + ) + x_109 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + input_6 = torch.conv2d( + x_111, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_112 = torch.conv2d( + input_6, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_113 = x_112.permute(0, 2, 3, 1) + x_112 = None + v_12 = torch.var(x_113, dim=(-1,), keepdim=True) + add_20 = v_12 + 1e-06 + v_12 = None + rsqrt_12 = torch.rsqrt(add_20) + add_20 = None + x_114 = x_113 * rsqrt_12 + x_113 = rsqrt_12 = None + x_115 = ( + x_114 + * l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + ) + x_114 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = (None) + x_116 = x_115.permute(0, 3, 1, 2) + x_115 = None + x_117 = torch.conv2d( + x_116, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_116 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_118 = torch._C._nn.gelu(x_117) + x_117 = None + x_119 = torch.nn.functional.dropout(x_118, 0.0, False, False) + x_118 = None + x_120 = torch.conv2d( + x_119, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_119 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_121 = torch.nn.functional.dropout(x_120, 0.0, False, False) + x_120 = None + reshape_8 = l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_ = ( + None + ) + x_122 = x_121.mul(reshape_8) + x_121 = reshape_8 = None + x_123 = x_122 + input_6 + x_122 = input_6 = None + x_124 = torch.conv2d( + x_123, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_125 = x_124.permute(0, 2, 3, 1) + x_124 = None + v_13 = torch.var(x_125, dim=(-1,), keepdim=True) + add_22 = v_13 + 1e-06 + v_13 = None + rsqrt_13 = torch.rsqrt(add_22) + add_22 = None + x_126 = x_125 * rsqrt_13 + x_125 = rsqrt_13 = None + x_127 = ( + x_126 + * l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + ) + x_126 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = (None) + x_128 = x_127.permute(0, 3, 1, 2) + x_127 = None + x_129 = torch.conv2d( + x_128, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_128 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_130 = torch._C._nn.gelu(x_129) + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch.conv2d( + x_131, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_131 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + reshape_9 = l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_.reshape( + 1, -1, 1, 1 + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_ = ( + None + ) + x_134 = x_133.mul(reshape_9) + x_133 = reshape_9 = None + x_135 = x_134 + x_123 + x_134 = x_123 = None + x_136 = torch.nn.functional.adaptive_avg_pool2d(x_135, 1) + x_135 = None + x_137 = x_136.permute(0, 2, 3, 1) + x_136 = None + v_14 = torch.var(x_137, dim=(-1,), keepdim=True) + add_24 = v_14 + 1e-06 + v_14 = None + rsqrt_14 = torch.rsqrt(add_24) + add_24 = None + x_138 = x_137 * rsqrt_14 + x_137 = rsqrt_14 = None + x_139 = x_138 * l_self_modules_head_modules_norm_parameters_weight_ + x_138 = l_self_modules_head_modules_norm_parameters_weight_ = None + x_140 = x_139.permute(0, 3, 1, 2) + x_139 = None + x_141 = x_140.flatten(1, -1) + x_140 = None + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = torch._C._nn.linear( + x_142, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_142 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_143,) diff --git a/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/weight_meta.py b/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/weight_meta.py new file mode 100644 index 0000000000..ee89ada53e --- /dev/null +++ b/samples/timm/convnext_zepto_rms_ols.ra4_e3600_r224_in1k/weight_meta.py @@ -0,0 +1,994 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [32, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.226 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_parameters_weight_: + name = "L_self_modules_stem_modules_2_parameters_weight_" + shape = [32, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_2_parameters_bias_: + name = "L_self_modules_stem_modules_2_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_3_parameters_weight_: + name = "L_self_modules_stem_modules_3_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [32, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [128, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [32, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_parameters_gamma_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [32, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [128, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [32, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_parameters_gamma_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [64, 32, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [64, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_parameters_gamma_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [64, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_parameters_gamma_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [128, 64, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_parameters_gamma_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [256, 128, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_parameters_gamma_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_atto.fcmae/graph_hash.txt b/samples/timm/convnextv2_atto.fcmae/graph_hash.txt new file mode 100644 index 0000000000..22926c0081 --- /dev/null +++ b/samples/timm/convnextv2_atto.fcmae/graph_hash.txt @@ -0,0 +1 @@ +3cf0175b6d1f5c006dfd17d29027c4b6912846eec375b1a975876378f099af57 \ No newline at end of file diff --git a/samples/timm/convnextv2_atto.fcmae/graph_net.json b/samples/timm/convnextv2_atto.fcmae/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnextv2_atto.fcmae/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_atto.fcmae/input_meta.py b/samples/timm/convnextv2_atto.fcmae/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_atto.fcmae/input_tensor_constraints.py b/samples/timm/convnextv2_atto.fcmae/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_atto.fcmae/model.py b/samples/timm/convnextv2_atto.fcmae/model.py new file mode 100644 index 0000000000..b6f5fd7a97 --- /dev/null +++ b/samples/timm/convnextv2_atto.fcmae/model.py @@ -0,0 +1,1299 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (40,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 40, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (40,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_g = x_9.norm(p=2, dim=(2, 3), keepdim=True) + mean = x_g.mean(dim=1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_9 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_10 = x_9 + addcmul + x_9 = addcmul = None + x_11 = torch.conv2d( + x_10, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_10 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 40, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (40,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_g_1 = x_20.norm(p=2, dim=(2, 3), keepdim=True) + mean_1 = x_g_1.mean(dim=1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_20 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_21 = x_20 + addcmul_1 + x_20 = addcmul_1 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (40,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_2 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (80,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_g_2 = x_34.norm(p=2, dim=(2, 3), keepdim=True) + mean_2 = x_g_2.mean(dim=1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_34 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_35 = x_34 + addcmul_2 + x_34 = addcmul_2 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_37 = torch.nn.functional.dropout(x_36, 0.0, False, False) + x_36 = None + x_38 = x_37 + input_2 + x_37 = input_2 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (80,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_g_3 = x_45.norm(p=2, dim=(2, 3), keepdim=True) + mean_3 = x_g_3.mean(dim=1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_45 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_46 = x_45 + addcmul_3 + x_45 = addcmul_3 = None + x_47 = torch.conv2d( + x_46, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (80,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_3 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_g_4 = x_59.norm(p=2, dim=(2, 3), keepdim=True) + mean_4 = x_g_4.mean(dim=1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_59 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_60 = x_59 + addcmul_4 + x_59 = addcmul_4 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_60 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + x_63 = x_62 + input_3 + x_62 = input_3 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_g_5 = x_70.norm(p=2, dim=(2, 3), keepdim=True) + mean_5 = x_g_5.mean(dim=1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_70 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_71 = x_70 + addcmul_5 + x_70 = addcmul_5 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_g_6 = x_81.norm(p=2, dim=(2, 3), keepdim=True) + mean_6 = x_g_6.mean(dim=1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_81 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_82 = x_81 + addcmul_6 + x_81 = addcmul_6 = None + x_83 = torch.conv2d( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_82 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_g_7 = x_92.norm(p=2, dim=(2, 3), keepdim=True) + mean_7 = x_g_7.mean(dim=1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_92 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_93 = x_92 + addcmul_7 + x_92 = addcmul_7 = None + x_94 = torch.conv2d( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_g_8 = x_103.norm(p=2, dim=(2, 3), keepdim=True) + mean_8 = x_g_8.mean(dim=1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_103 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_104 = x_103 + addcmul_8 + x_103 = addcmul_8 = None + x_105 = torch.conv2d( + x_104, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_104 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_g_9 = x_114.norm(p=2, dim=(2, 3), keepdim=True) + mean_9 = x_g_9.mean(dim=1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_114 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_115 = x_114 + addcmul_9 + x_114 = addcmul_9 = None + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = x_118.permute(0, 2, 3, 1) + x_118 = None + x_120 = torch.nn.functional.layer_norm( + x_119, + (160,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_119 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_121 = x_120.permute(0, 3, 1, 2) + x_120 = None + input_4 = torch.conv2d( + x_121, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_122 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_123 = x_122.permute(0, 2, 3, 1) + x_122 = None + x_124 = torch.nn.functional.layer_norm( + x_123, + (320,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_125 = x_124.permute(0, 3, 1, 2) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126) + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_g_10 = x_128.norm(p=2, dim=(2, 3), keepdim=True) + mean_10 = x_g_10.mean(dim=1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_128 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_129 = x_128 + addcmul_10 + x_128 = addcmul_10 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = x_131 + input_4 + x_131 = input_4 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_134 = x_133.permute(0, 2, 3, 1) + x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (320,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_134 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_136 = x_135.permute(0, 3, 1, 2) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137) + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_g_11 = x_139.norm(p=2, dim=(2, 3), keepdim=True) + mean_11 = x_g_11.mean(dim=1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_139 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_140 = x_139 + addcmul_11 + x_139 = addcmul_11 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_140 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_142 + x_132 + x_142 = x_132 = None + x_144 = torch.nn.functional.adaptive_avg_pool2d(x_143, 1) + x_143 = None + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (320,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = x_147.flatten(1, -1) + x_147 = None + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + return (x_149,) diff --git a/samples/timm/convnextv2_atto.fcmae/weight_meta.py b/samples/timm/convnextv2_atto.fcmae/weight_meta.py new file mode 100644 index 0000000000..68ee77ec02 --- /dev/null +++ b/samples/timm/convnextv2_atto.fcmae/weight_meta.py @@ -0,0 +1,1410 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [40, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.229 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [40, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [160, 40, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [40, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [40, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [160, 40, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [40, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [80, 40, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [160, 80, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [320, 160, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_atto.fcmae_ft_in1k/graph_hash.txt b/samples/timm/convnextv2_atto.fcmae_ft_in1k/graph_hash.txt new file mode 100644 index 0000000000..24c226b77d --- /dev/null +++ b/samples/timm/convnextv2_atto.fcmae_ft_in1k/graph_hash.txt @@ -0,0 +1 @@ +c66466ed362e91b7ab9dec5c38bc4e70d2d78e515479441349935b2bc807f770 \ No newline at end of file diff --git a/samples/timm/convnextv2_atto.fcmae_ft_in1k/graph_net.json b/samples/timm/convnextv2_atto.fcmae_ft_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnextv2_atto.fcmae_ft_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_atto.fcmae_ft_in1k/input_meta.py b/samples/timm/convnextv2_atto.fcmae_ft_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_atto.fcmae_ft_in1k/input_tensor_constraints.py b/samples/timm/convnextv2_atto.fcmae_ft_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_atto.fcmae_ft_in1k/model.py b/samples/timm/convnextv2_atto.fcmae_ft_in1k/model.py new file mode 100644 index 0000000000..26d24b3d12 --- /dev/null +++ b/samples/timm/convnextv2_atto.fcmae_ft_in1k/model.py @@ -0,0 +1,1315 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (40,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 40, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (40,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_g = x_9.norm(p=2, dim=(2, 3), keepdim=True) + mean = x_g.mean(dim=1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_9 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_10 = x_9 + addcmul + x_9 = addcmul = None + x_11 = torch.conv2d( + x_10, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_10 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 40, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (40,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_g_1 = x_20.norm(p=2, dim=(2, 3), keepdim=True) + mean_1 = x_g_1.mean(dim=1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_20 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_21 = x_20 + addcmul_1 + x_20 = addcmul_1 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (40,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_2 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (80,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_g_2 = x_34.norm(p=2, dim=(2, 3), keepdim=True) + mean_2 = x_g_2.mean(dim=1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_34 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_35 = x_34 + addcmul_2 + x_34 = addcmul_2 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_37 = torch.nn.functional.dropout(x_36, 0.0, False, False) + x_36 = None + x_38 = x_37 + input_2 + x_37 = input_2 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (80,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_g_3 = x_45.norm(p=2, dim=(2, 3), keepdim=True) + mean_3 = x_g_3.mean(dim=1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_45 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_46 = x_45 + addcmul_3 + x_45 = addcmul_3 = None + x_47 = torch.conv2d( + x_46, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (80,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_3 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_g_4 = x_59.norm(p=2, dim=(2, 3), keepdim=True) + mean_4 = x_g_4.mean(dim=1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_59 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_60 = x_59 + addcmul_4 + x_59 = addcmul_4 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_60 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + x_63 = x_62 + input_3 + x_62 = input_3 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_g_5 = x_70.norm(p=2, dim=(2, 3), keepdim=True) + mean_5 = x_g_5.mean(dim=1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_70 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_71 = x_70 + addcmul_5 + x_70 = addcmul_5 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_g_6 = x_81.norm(p=2, dim=(2, 3), keepdim=True) + mean_6 = x_g_6.mean(dim=1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_81 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_82 = x_81 + addcmul_6 + x_81 = addcmul_6 = None + x_83 = torch.conv2d( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_82 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_g_7 = x_92.norm(p=2, dim=(2, 3), keepdim=True) + mean_7 = x_g_7.mean(dim=1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_92 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_93 = x_92 + addcmul_7 + x_92 = addcmul_7 = None + x_94 = torch.conv2d( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_g_8 = x_103.norm(p=2, dim=(2, 3), keepdim=True) + mean_8 = x_g_8.mean(dim=1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_103 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_104 = x_103 + addcmul_8 + x_103 = addcmul_8 = None + x_105 = torch.conv2d( + x_104, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_104 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (160,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_g_9 = x_114.norm(p=2, dim=(2, 3), keepdim=True) + mean_9 = x_g_9.mean(dim=1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_114 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_115 = x_114 + addcmul_9 + x_114 = addcmul_9 = None + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = x_118.permute(0, 2, 3, 1) + x_118 = None + x_120 = torch.nn.functional.layer_norm( + x_119, + (160,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_119 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_121 = x_120.permute(0, 3, 1, 2) + x_120 = None + input_4 = torch.conv2d( + x_121, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_122 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_123 = x_122.permute(0, 2, 3, 1) + x_122 = None + x_124 = torch.nn.functional.layer_norm( + x_123, + (320,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_125 = x_124.permute(0, 3, 1, 2) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126) + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_g_10 = x_128.norm(p=2, dim=(2, 3), keepdim=True) + mean_10 = x_g_10.mean(dim=1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_128 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_129 = x_128 + addcmul_10 + x_128 = addcmul_10 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = x_131 + input_4 + x_131 = input_4 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_134 = x_133.permute(0, 2, 3, 1) + x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (320,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_134 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_136 = x_135.permute(0, 3, 1, 2) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137) + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_g_11 = x_139.norm(p=2, dim=(2, 3), keepdim=True) + mean_11 = x_g_11.mean(dim=1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_139 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_140 = x_139 + addcmul_11 + x_139 = addcmul_11 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_140 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_142 + x_132 + x_142 = x_132 = None + x_144 = torch.nn.functional.adaptive_avg_pool2d(x_143, 1) + x_143 = None + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (320,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = x_147.flatten(1, -1) + x_147 = None + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = torch._C._nn.linear( + x_149, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_149 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_150,) diff --git a/samples/timm/convnextv2_atto.fcmae_ft_in1k/weight_meta.py b/samples/timm/convnextv2_atto.fcmae_ft_in1k/weight_meta.py new file mode 100644 index 0000000000..f11ac5c7ba --- /dev/null +++ b/samples/timm/convnextv2_atto.fcmae_ft_in1k/weight_meta.py @@ -0,0 +1,1430 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [40, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.224 + std = 1.284 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [40, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [160, 40, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [40, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [40, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [160, 40, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [40, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [80, 40, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [160, 80, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [320, 160, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 320] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_base.fcmae/graph_hash.txt b/samples/timm/convnextv2_base.fcmae/graph_hash.txt new file mode 100644 index 0000000000..108b252cdc --- /dev/null +++ b/samples/timm/convnextv2_base.fcmae/graph_hash.txt @@ -0,0 +1 @@ +8d3afeab77d8b825e8fe1abe687ffb77e6f066e13c4b04211471aad7ecf62c65 \ No newline at end of file diff --git a/samples/timm/convnextv2_base.fcmae/graph_net.json b/samples/timm/convnextv2_base.fcmae/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnextv2_base.fcmae/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_base.fcmae/input_meta.py b/samples/timm/convnextv2_base.fcmae/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_base.fcmae/input_tensor_constraints.py b/samples/timm/convnextv2_base.fcmae/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_base.fcmae/model.py b/samples/timm/convnextv2_base.fcmae/model.py new file mode 100644 index 0000000000..c9f2785f86 --- /dev/null +++ b/samples/timm/convnextv2_base.fcmae/model.py @@ -0,0 +1,3267 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (128,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_g = x_8.norm(p=2, dim=(1, 2), keepdim=True) + mean = x_g.mean(dim=-1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_8 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_9 = x_8 + addcmul + x_8 = addcmul = None + x_10 = torch._C._nn.linear( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = x_11.permute(0, 3, 1, 2) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_g_1 = x_19.norm(p=2, dim=(1, 2), keepdim=True) + mean_1 = x_g_1.mean(dim=-1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_19 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_20 = x_19 + addcmul_1 + x_19 = addcmul_1 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_g_2 = x_30.norm(p=2, dim=(1, 2), keepdim=True) + mean_2 = x_g_2.mean(dim=-1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_30 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_31 = x_30 + addcmul_2 + x_30 = addcmul_2 = None + x_32 = torch._C._nn.linear( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + x_34 = x_33.permute(0, 3, 1, 2) + x_33 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (128,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_g_3 = x_44.norm(p=2, dim=(1, 2), keepdim=True) + mean_3 = x_g_3.mean(dim=-1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_44 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_45 = x_44 + addcmul_3 + x_44 = addcmul_3 = None + x_46 = torch._C._nn.linear( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_g_4 = x_55.norm(p=2, dim=(1, 2), keepdim=True) + mean_4 = x_g_4.mean(dim=-1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_55 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_56 = x_55 + addcmul_4 + x_55 = addcmul_4 = None + x_57 = torch._C._nn.linear( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_58 = torch.nn.functional.dropout(x_57, 0.0, False, False) + x_57 = None + x_59 = x_58.permute(0, 3, 1, 2) + x_58 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_g_5 = x_66.norm(p=2, dim=(1, 2), keepdim=True) + mean_5 = x_g_5.mean(dim=-1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_66 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_67 = x_66 + addcmul_5 + x_66 = addcmul_5 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = x_69.permute(0, 3, 1, 2) + x_69 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (256,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_g_6 = x_80.norm(p=2, dim=(1, 2), keepdim=True) + mean_6 = x_g_6.mean(dim=-1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_80 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_81 = x_80 + addcmul_6 + x_80 = addcmul_6 = None + x_82 = torch._C._nn.linear( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + x_84 = x_83.permute(0, 3, 1, 2) + x_83 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_g_7 = x_91.norm(p=2, dim=(1, 2), keepdim=True) + mean_7 = x_g_7.mean(dim=-1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_91 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_92 = x_91 + addcmul_7 + x_91 = addcmul_7 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_g_8 = x_102.norm(p=2, dim=(1, 2), keepdim=True) + mean_8 = x_g_8.mean(dim=-1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_102 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_103 = x_102 + addcmul_8 + x_102 = addcmul_8 = None + x_104 = torch._C._nn.linear( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = x_105.permute(0, 3, 1, 2) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_g_9 = x_113.norm(p=2, dim=(1, 2), keepdim=True) + mean_9 = x_g_9.mean(dim=-1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_113 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_114 = x_113 + addcmul_9 + x_113 = addcmul_9 = None + x_115 = torch._C._nn.linear( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + x_117 = x_116.permute(0, 3, 1, 2) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_g_10 = x_124.norm(p=2, dim=(1, 2), keepdim=True) + mean_10 = x_g_10.mean(dim=-1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_124 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_125 = x_124 + addcmul_10 + x_124 = addcmul_10 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = x_127.permute(0, 3, 1, 2) + x_127 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_g_11 = x_135.norm(p=2, dim=(1, 2), keepdim=True) + mean_11 = x_g_11.mean(dim=-1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_135 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_136 = x_135 + addcmul_11 + x_135 = addcmul_11 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_138.permute(0, 3, 1, 2) + x_138 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_g_12 = x_146.norm(p=2, dim=(1, 2), keepdim=True) + mean_12 = x_g_12.mean(dim=-1, keepdim=True) + add_36 = mean_12 + 1e-06 + mean_12 = None + x_n_12 = x_g_12 / add_36 + x_g_12 = add_36 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_12 = x_146 * x_n_12 + x_n_12 = None + addcmul_12 = torch.addcmul(view_24, view_25, mul_12) + view_24 = view_25 = mul_12 = None + x_147 = x_146 + addcmul_12 + x_146 = addcmul_12 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_149.permute(0, 3, 1, 2) + x_149 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_g_13 = x_157.norm(p=2, dim=(1, 2), keepdim=True) + mean_13 = x_g_13.mean(dim=-1, keepdim=True) + add_39 = mean_13 + 1e-06 + mean_13 = None + x_n_13 = x_g_13 / add_39 + x_g_13 = add_39 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_13 = x_157 * x_n_13 + x_n_13 = None + addcmul_13 = torch.addcmul(view_26, view_27, mul_13) + view_26 = view_27 = mul_13 = None + x_158 = x_157 + addcmul_13 + x_157 = addcmul_13 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_160.permute(0, 3, 1, 2) + x_160 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_g_14 = x_168.norm(p=2, dim=(1, 2), keepdim=True) + mean_14 = x_g_14.mean(dim=-1, keepdim=True) + add_42 = mean_14 + 1e-06 + mean_14 = None + x_n_14 = x_g_14 / add_42 + x_g_14 = add_42 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_14 = x_168 * x_n_14 + x_n_14 = None + addcmul_14 = torch.addcmul(view_28, view_29, mul_14) + view_28 = view_29 = mul_14 = None + x_169 = x_168 + addcmul_14 + x_168 = addcmul_14 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_171.permute(0, 3, 1, 2) + x_171 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = (None) + x_175 = x_174.permute(0, 2, 3, 1) + x_174 = None + x_176 = torch.nn.functional.layer_norm( + x_175, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_, + 1e-06, + ) + x_175 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = (None) + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_176 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_178 = torch._C._nn.gelu(x_177) + x_177 = None + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_g_15 = x_179.norm(p=2, dim=(1, 2), keepdim=True) + mean_15 = x_g_15.mean(dim=-1, keepdim=True) + add_45 = mean_15 + 1e-06 + mean_15 = None + x_n_15 = x_g_15 / add_45 + x_g_15 = add_45 = None + view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_15 = x_179 * x_n_15 + x_n_15 = None + addcmul_15 = torch.addcmul(view_30, view_31, mul_15) + view_30 = view_31 = mul_15 = None + x_180 = x_179 + addcmul_15 + x_179 = addcmul_15 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_182.permute(0, 3, 1, 2) + x_182 = None + x_184 = x_183 + x_173 + x_183 = x_173 = None + x_185 = torch.conv2d( + x_184, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = (None) + x_186 = x_185.permute(0, 2, 3, 1) + x_185 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = (None) + x_188 = torch._C._nn.linear( + x_187, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_187 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_189 = torch._C._nn.gelu(x_188) + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_g_16 = x_190.norm(p=2, dim=(1, 2), keepdim=True) + mean_16 = x_g_16.mean(dim=-1, keepdim=True) + add_48 = mean_16 + 1e-06 + mean_16 = None + x_n_16 = x_g_16 / add_48 + x_g_16 = add_48 = None + view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_16 = x_190 * x_n_16 + x_n_16 = None + addcmul_16 = torch.addcmul(view_32, view_33, mul_16) + view_32 = view_33 = mul_16 = None + x_191 = x_190 + addcmul_16 + x_190 = addcmul_16 = None + x_192 = torch._C._nn.linear( + x_191, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_191 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_193.permute(0, 3, 1, 2) + x_193 = None + x_195 = x_194 + x_184 + x_194 = x_184 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = (None) + x_197 = x_196.permute(0, 2, 3, 1) + x_196 = None + x_198 = torch.nn.functional.layer_norm( + x_197, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_, + 1e-06, + ) + x_197 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = (None) + x_199 = torch._C._nn.linear( + x_198, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_198 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_200 = torch._C._nn.gelu(x_199) + x_199 = None + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_g_17 = x_201.norm(p=2, dim=(1, 2), keepdim=True) + mean_17 = x_g_17.mean(dim=-1, keepdim=True) + add_51 = mean_17 + 1e-06 + mean_17 = None + x_n_17 = x_g_17 / add_51 + x_g_17 = add_51 = None + view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_17 = x_201 * x_n_17 + x_n_17 = None + addcmul_17 = torch.addcmul(view_34, view_35, mul_17) + view_34 = view_35 = mul_17 = None + x_202 = x_201 + addcmul_17 + x_201 = addcmul_17 = None + x_203 = torch._C._nn.linear( + x_202, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_202 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_205 = x_204.permute(0, 3, 1, 2) + x_204 = None + x_206 = x_205 + x_195 + x_205 = x_195 = None + x_207 = torch.conv2d( + x_206, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = (None) + x_208 = x_207.permute(0, 2, 3, 1) + x_207 = None + x_209 = torch.nn.functional.layer_norm( + x_208, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_, + 1e-06, + ) + x_208 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = (None) + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_209 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_211 = torch._C._nn.gelu(x_210) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_g_18 = x_212.norm(p=2, dim=(1, 2), keepdim=True) + mean_18 = x_g_18.mean(dim=-1, keepdim=True) + add_54 = mean_18 + 1e-06 + mean_18 = None + x_n_18 = x_g_18 / add_54 + x_g_18 = add_54 = None + view_36 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_37 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_18 = x_212 * x_n_18 + x_n_18 = None + addcmul_18 = torch.addcmul(view_36, view_37, mul_18) + view_36 = view_37 = mul_18 = None + x_213 = x_212 + addcmul_18 + x_212 = addcmul_18 = None + x_214 = torch._C._nn.linear( + x_213, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_213 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = x_215.permute(0, 3, 1, 2) + x_215 = None + x_217 = x_216 + x_206 + x_216 = x_206 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = (None) + x_219 = x_218.permute(0, 2, 3, 1) + x_218 = None + x_220 = torch.nn.functional.layer_norm( + x_219, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_, + 1e-06, + ) + x_219 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = (None) + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_220 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_222 = torch._C._nn.gelu(x_221) + x_221 = None + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_g_19 = x_223.norm(p=2, dim=(1, 2), keepdim=True) + mean_19 = x_g_19.mean(dim=-1, keepdim=True) + add_57 = mean_19 + 1e-06 + mean_19 = None + x_n_19 = x_g_19 / add_57 + x_g_19 = add_57 = None + view_38 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_39 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_19 = x_223 * x_n_19 + x_n_19 = None + addcmul_19 = torch.addcmul(view_38, view_39, mul_19) + view_38 = view_39 = mul_19 = None + x_224 = x_223 + addcmul_19 + x_223 = addcmul_19 = None + x_225 = torch._C._nn.linear( + x_224, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_224 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_226 = torch.nn.functional.dropout(x_225, 0.0, False, False) + x_225 = None + x_227 = x_226.permute(0, 3, 1, 2) + x_226 = None + x_228 = x_227 + x_217 + x_227 = x_217 = None + x_229 = torch.conv2d( + x_228, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = (None) + x_230 = x_229.permute(0, 2, 3, 1) + x_229 = None + x_231 = torch.nn.functional.layer_norm( + x_230, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_, + 1e-06, + ) + x_230 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = (None) + x_232 = torch._C._nn.linear( + x_231, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_231 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_233 = torch._C._nn.gelu(x_232) + x_232 = None + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_g_20 = x_234.norm(p=2, dim=(1, 2), keepdim=True) + mean_20 = x_g_20.mean(dim=-1, keepdim=True) + add_60 = mean_20 + 1e-06 + mean_20 = None + x_n_20 = x_g_20 / add_60 + x_g_20 = add_60 = None + view_40 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_41 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_20 = x_234 * x_n_20 + x_n_20 = None + addcmul_20 = torch.addcmul(view_40, view_41, mul_20) + view_40 = view_41 = mul_20 = None + x_235 = x_234 + addcmul_20 + x_234 = addcmul_20 = None + x_236 = torch._C._nn.linear( + x_235, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_235 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_237 = torch.nn.functional.dropout(x_236, 0.0, False, False) + x_236 = None + x_238 = x_237.permute(0, 3, 1, 2) + x_237 = None + x_239 = x_238 + x_228 + x_238 = x_228 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = (None) + x_241 = x_240.permute(0, 2, 3, 1) + x_240 = None + x_242 = torch.nn.functional.layer_norm( + x_241, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_, + 1e-06, + ) + x_241 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = (None) + x_243 = torch._C._nn.linear( + x_242, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_242 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_244 = torch._C._nn.gelu(x_243) + x_243 = None + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_g_21 = x_245.norm(p=2, dim=(1, 2), keepdim=True) + mean_21 = x_g_21.mean(dim=-1, keepdim=True) + add_63 = mean_21 + 1e-06 + mean_21 = None + x_n_21 = x_g_21 / add_63 + x_g_21 = add_63 = None + view_42 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_43 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_21 = x_245 * x_n_21 + x_n_21 = None + addcmul_21 = torch.addcmul(view_42, view_43, mul_21) + view_42 = view_43 = mul_21 = None + x_246 = x_245 + addcmul_21 + x_245 = addcmul_21 = None + x_247 = torch._C._nn.linear( + x_246, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_246 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_248 = torch.nn.functional.dropout(x_247, 0.0, False, False) + x_247 = None + x_249 = x_248.permute(0, 3, 1, 2) + x_248 = None + x_250 = x_249 + x_239 + x_249 = x_239 = None + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = (None) + x_252 = x_251.permute(0, 2, 3, 1) + x_251 = None + x_253 = torch.nn.functional.layer_norm( + x_252, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_, + 1e-06, + ) + x_252 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = (None) + x_254 = torch._C._nn.linear( + x_253, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_253 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_255 = torch._C._nn.gelu(x_254) + x_254 = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_g_22 = x_256.norm(p=2, dim=(1, 2), keepdim=True) + mean_22 = x_g_22.mean(dim=-1, keepdim=True) + add_66 = mean_22 + 1e-06 + mean_22 = None + x_n_22 = x_g_22 / add_66 + x_g_22 = add_66 = None + view_44 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_45 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_22 = x_256 * x_n_22 + x_n_22 = None + addcmul_22 = torch.addcmul(view_44, view_45, mul_22) + view_44 = view_45 = mul_22 = None + x_257 = x_256 + addcmul_22 + x_256 = addcmul_22 = None + x_258 = torch._C._nn.linear( + x_257, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_257 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = x_259.permute(0, 3, 1, 2) + x_259 = None + x_261 = x_260 + x_250 + x_260 = x_250 = None + x_262 = torch.conv2d( + x_261, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = (None) + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = (None) + x_265 = torch._C._nn.linear( + x_264, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_264 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_266 = torch._C._nn.gelu(x_265) + x_265 = None + x_267 = torch.nn.functional.dropout(x_266, 0.0, False, False) + x_266 = None + x_g_23 = x_267.norm(p=2, dim=(1, 2), keepdim=True) + mean_23 = x_g_23.mean(dim=-1, keepdim=True) + add_69 = mean_23 + 1e-06 + mean_23 = None + x_n_23 = x_g_23 / add_69 + x_g_23 = add_69 = None + view_46 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_47 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_23 = x_267 * x_n_23 + x_n_23 = None + addcmul_23 = torch.addcmul(view_46, view_47, mul_23) + view_46 = view_47 = mul_23 = None + x_268 = x_267 + addcmul_23 + x_267 = addcmul_23 = None + x_269 = torch._C._nn.linear( + x_268, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_268 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_270 = torch.nn.functional.dropout(x_269, 0.0, False, False) + x_269 = None + x_271 = x_270.permute(0, 3, 1, 2) + x_270 = None + x_272 = x_271 + x_261 + x_271 = x_261 = None + x_273 = torch.conv2d( + x_272, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = (None) + x_274 = x_273.permute(0, 2, 3, 1) + x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_, + 1e-06, + ) + x_274 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = (None) + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_277 = torch._C._nn.gelu(x_276) + x_276 = None + x_278 = torch.nn.functional.dropout(x_277, 0.0, False, False) + x_277 = None + x_g_24 = x_278.norm(p=2, dim=(1, 2), keepdim=True) + mean_24 = x_g_24.mean(dim=-1, keepdim=True) + add_72 = mean_24 + 1e-06 + mean_24 = None + x_n_24 = x_g_24 / add_72 + x_g_24 = add_72 = None + view_48 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_49 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_24 = x_278 * x_n_24 + x_n_24 = None + addcmul_24 = torch.addcmul(view_48, view_49, mul_24) + view_48 = view_49 = mul_24 = None + x_279 = x_278 + addcmul_24 + x_278 = addcmul_24 = None + x_280 = torch._C._nn.linear( + x_279, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_279 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_281 = torch.nn.functional.dropout(x_280, 0.0, False, False) + x_280 = None + x_282 = x_281.permute(0, 3, 1, 2) + x_281 = None + x_283 = x_282 + x_272 + x_282 = x_272 = None + x_284 = torch.conv2d( + x_283, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = (None) + x_285 = x_284.permute(0, 2, 3, 1) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_, + 1e-06, + ) + x_285 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = (None) + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_288 = torch._C._nn.gelu(x_287) + x_287 = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + x_g_25 = x_289.norm(p=2, dim=(1, 2), keepdim=True) + mean_25 = x_g_25.mean(dim=-1, keepdim=True) + add_75 = mean_25 + 1e-06 + mean_25 = None + x_n_25 = x_g_25 / add_75 + x_g_25 = add_75 = None + view_50 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_51 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_25 = x_289 * x_n_25 + x_n_25 = None + addcmul_25 = torch.addcmul(view_50, view_51, mul_25) + view_50 = view_51 = mul_25 = None + x_290 = x_289 + addcmul_25 + x_289 = addcmul_25 = None + x_291 = torch._C._nn.linear( + x_290, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_290 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_292 = torch.nn.functional.dropout(x_291, 0.0, False, False) + x_291 = None + x_293 = x_292.permute(0, 3, 1, 2) + x_292 = None + x_294 = x_293 + x_283 + x_293 = x_283 = None + x_295 = torch.conv2d( + x_294, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = (None) + x_296 = x_295.permute(0, 2, 3, 1) + x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_, + 1e-06, + ) + x_296 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = (None) + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_299 = torch._C._nn.gelu(x_298) + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_g_26 = x_300.norm(p=2, dim=(1, 2), keepdim=True) + mean_26 = x_g_26.mean(dim=-1, keepdim=True) + add_78 = mean_26 + 1e-06 + mean_26 = None + x_n_26 = x_g_26 / add_78 + x_g_26 = add_78 = None + view_52 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_53 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_26 = x_300 * x_n_26 + x_n_26 = None + addcmul_26 = torch.addcmul(view_52, view_53, mul_26) + view_52 = view_53 = mul_26 = None + x_301 = x_300 + addcmul_26 + x_300 = addcmul_26 = None + x_302 = torch._C._nn.linear( + x_301, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_301 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_303 = torch.nn.functional.dropout(x_302, 0.0, False, False) + x_302 = None + x_304 = x_303.permute(0, 3, 1, 2) + x_303 = None + x_305 = x_304 + x_294 + x_304 = x_294 = None + x_306 = torch.conv2d( + x_305, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = (None) + x_307 = x_306.permute(0, 2, 3, 1) + x_306 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_, + 1e-06, + ) + x_307 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = (None) + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_310 = torch._C._nn.gelu(x_309) + x_309 = None + x_311 = torch.nn.functional.dropout(x_310, 0.0, False, False) + x_310 = None + x_g_27 = x_311.norm(p=2, dim=(1, 2), keepdim=True) + mean_27 = x_g_27.mean(dim=-1, keepdim=True) + add_81 = mean_27 + 1e-06 + mean_27 = None + x_n_27 = x_g_27 / add_81 + x_g_27 = add_81 = None + view_54 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_55 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_27 = x_311 * x_n_27 + x_n_27 = None + addcmul_27 = torch.addcmul(view_54, view_55, mul_27) + view_54 = view_55 = mul_27 = None + x_312 = x_311 + addcmul_27 + x_311 = addcmul_27 = None + x_313 = torch._C._nn.linear( + x_312, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_312 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_314 = torch.nn.functional.dropout(x_313, 0.0, False, False) + x_313 = None + x_315 = x_314.permute(0, 3, 1, 2) + x_314 = None + x_316 = x_315 + x_305 + x_315 = x_305 = None + x_317 = torch.conv2d( + x_316, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = (None) + x_318 = x_317.permute(0, 2, 3, 1) + x_317 = None + x_319 = torch.nn.functional.layer_norm( + x_318, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_, + 1e-06, + ) + x_318 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = (None) + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_321 = torch._C._nn.gelu(x_320) + x_320 = None + x_322 = torch.nn.functional.dropout(x_321, 0.0, False, False) + x_321 = None + x_g_28 = x_322.norm(p=2, dim=(1, 2), keepdim=True) + mean_28 = x_g_28.mean(dim=-1, keepdim=True) + add_84 = mean_28 + 1e-06 + mean_28 = None + x_n_28 = x_g_28 / add_84 + x_g_28 = add_84 = None + view_56 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_57 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_28 = x_322 * x_n_28 + x_n_28 = None + addcmul_28 = torch.addcmul(view_56, view_57, mul_28) + view_56 = view_57 = mul_28 = None + x_323 = x_322 + addcmul_28 + x_322 = addcmul_28 = None + x_324 = torch._C._nn.linear( + x_323, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_323 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_325 = torch.nn.functional.dropout(x_324, 0.0, False, False) + x_324 = None + x_326 = x_325.permute(0, 3, 1, 2) + x_325 = None + x_327 = x_326 + x_316 + x_326 = x_316 = None + x_328 = torch.conv2d( + x_327, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = (None) + x_329 = x_328.permute(0, 2, 3, 1) + x_328 = None + x_330 = torch.nn.functional.layer_norm( + x_329, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_, + 1e-06, + ) + x_329 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = (None) + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_332 = torch._C._nn.gelu(x_331) + x_331 = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_g_29 = x_333.norm(p=2, dim=(1, 2), keepdim=True) + mean_29 = x_g_29.mean(dim=-1, keepdim=True) + add_87 = mean_29 + 1e-06 + mean_29 = None + x_n_29 = x_g_29 / add_87 + x_g_29 = add_87 = None + view_58 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_59 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_29 = x_333 * x_n_29 + x_n_29 = None + addcmul_29 = torch.addcmul(view_58, view_59, mul_29) + view_58 = view_59 = mul_29 = None + x_334 = x_333 + addcmul_29 + x_333 = addcmul_29 = None + x_335 = torch._C._nn.linear( + x_334, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_334 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_336 = torch.nn.functional.dropout(x_335, 0.0, False, False) + x_335 = None + x_337 = x_336.permute(0, 3, 1, 2) + x_336 = None + x_338 = x_337 + x_327 + x_337 = x_327 = None + x_339 = torch.conv2d( + x_338, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = (None) + x_340 = x_339.permute(0, 2, 3, 1) + x_339 = None + x_341 = torch.nn.functional.layer_norm( + x_340, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_, + 1e-06, + ) + x_340 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = (None) + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_343 = torch._C._nn.gelu(x_342) + x_342 = None + x_344 = torch.nn.functional.dropout(x_343, 0.0, False, False) + x_343 = None + x_g_30 = x_344.norm(p=2, dim=(1, 2), keepdim=True) + mean_30 = x_g_30.mean(dim=-1, keepdim=True) + add_90 = mean_30 + 1e-06 + mean_30 = None + x_n_30 = x_g_30 / add_90 + x_g_30 = add_90 = None + view_60 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_61 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_30 = x_344 * x_n_30 + x_n_30 = None + addcmul_30 = torch.addcmul(view_60, view_61, mul_30) + view_60 = view_61 = mul_30 = None + x_345 = x_344 + addcmul_30 + x_344 = addcmul_30 = None + x_346 = torch._C._nn.linear( + x_345, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_345 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_347 = torch.nn.functional.dropout(x_346, 0.0, False, False) + x_346 = None + x_348 = x_347.permute(0, 3, 1, 2) + x_347 = None + x_349 = x_348 + x_338 + x_348 = x_338 = None + x_350 = torch.conv2d( + x_349, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = (None) + x_351 = x_350.permute(0, 2, 3, 1) + x_350 = None + x_352 = torch.nn.functional.layer_norm( + x_351, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_, + 1e-06, + ) + x_351 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = (None) + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_354 = torch._C._nn.gelu(x_353) + x_353 = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_g_31 = x_355.norm(p=2, dim=(1, 2), keepdim=True) + mean_31 = x_g_31.mean(dim=-1, keepdim=True) + add_93 = mean_31 + 1e-06 + mean_31 = None + x_n_31 = x_g_31 / add_93 + x_g_31 = add_93 = None + view_62 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_63 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_31 = x_355 * x_n_31 + x_n_31 = None + addcmul_31 = torch.addcmul(view_62, view_63, mul_31) + view_62 = view_63 = mul_31 = None + x_356 = x_355 + addcmul_31 + x_355 = addcmul_31 = None + x_357 = torch._C._nn.linear( + x_356, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_356 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_358 = torch.nn.functional.dropout(x_357, 0.0, False, False) + x_357 = None + x_359 = x_358.permute(0, 3, 1, 2) + x_358 = None + x_360 = x_359 + x_349 + x_359 = x_349 = None + x_361 = torch.conv2d( + x_360, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = (None) + x_362 = x_361.permute(0, 2, 3, 1) + x_361 = None + x_363 = torch.nn.functional.layer_norm( + x_362, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_, + 1e-06, + ) + x_362 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = (None) + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_365 = torch._C._nn.gelu(x_364) + x_364 = None + x_366 = torch.nn.functional.dropout(x_365, 0.0, False, False) + x_365 = None + x_g_32 = x_366.norm(p=2, dim=(1, 2), keepdim=True) + mean_32 = x_g_32.mean(dim=-1, keepdim=True) + add_96 = mean_32 + 1e-06 + mean_32 = None + x_n_32 = x_g_32 / add_96 + x_g_32 = add_96 = None + view_64 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_65 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_32 = x_366 * x_n_32 + x_n_32 = None + addcmul_32 = torch.addcmul(view_64, view_65, mul_32) + view_64 = view_65 = mul_32 = None + x_367 = x_366 + addcmul_32 + x_366 = addcmul_32 = None + x_368 = torch._C._nn.linear( + x_367, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_367 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_369 = torch.nn.functional.dropout(x_368, 0.0, False, False) + x_368 = None + x_370 = x_369.permute(0, 3, 1, 2) + x_369 = None + x_371 = x_370 + x_360 + x_370 = x_360 = None + x_372 = x_371.permute(0, 2, 3, 1) + x_371 = None + x_373 = torch.nn.functional.layer_norm( + x_372, + (512,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_372 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_374 = x_373.permute(0, 3, 1, 2) + x_373 = None + input_4 = torch.conv2d( + x_374, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_374 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_375 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_376 = x_375.permute(0, 2, 3, 1) + x_375 = None + x_377 = torch.nn.functional.layer_norm( + x_376, + (1024,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_376 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_378 = torch._C._nn.linear( + x_377, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_377 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_379 = torch._C._nn.gelu(x_378) + x_378 = None + x_380 = torch.nn.functional.dropout(x_379, 0.0, False, False) + x_379 = None + x_g_33 = x_380.norm(p=2, dim=(1, 2), keepdim=True) + mean_33 = x_g_33.mean(dim=-1, keepdim=True) + add_99 = mean_33 + 1e-06 + mean_33 = None + x_n_33 = x_g_33 / add_99 + x_g_33 = add_99 = None + view_66 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_67 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_33 = x_380 * x_n_33 + x_n_33 = None + addcmul_33 = torch.addcmul(view_66, view_67, mul_33) + view_66 = view_67 = mul_33 = None + x_381 = x_380 + addcmul_33 + x_380 = addcmul_33 = None + x_382 = torch._C._nn.linear( + x_381, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_381 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_383 = torch.nn.functional.dropout(x_382, 0.0, False, False) + x_382 = None + x_384 = x_383.permute(0, 3, 1, 2) + x_383 = None + x_385 = x_384 + input_4 + x_384 = input_4 = None + x_386 = torch.conv2d( + x_385, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_387 = x_386.permute(0, 2, 3, 1) + x_386 = None + x_388 = torch.nn.functional.layer_norm( + x_387, + (1024,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_387 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_389 = torch._C._nn.linear( + x_388, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_388 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_390 = torch._C._nn.gelu(x_389) + x_389 = None + x_391 = torch.nn.functional.dropout(x_390, 0.0, False, False) + x_390 = None + x_g_34 = x_391.norm(p=2, dim=(1, 2), keepdim=True) + mean_34 = x_g_34.mean(dim=-1, keepdim=True) + add_102 = mean_34 + 1e-06 + mean_34 = None + x_n_34 = x_g_34 / add_102 + x_g_34 = add_102 = None + view_68 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_69 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_34 = x_391 * x_n_34 + x_n_34 = None + addcmul_34 = torch.addcmul(view_68, view_69, mul_34) + view_68 = view_69 = mul_34 = None + x_392 = x_391 + addcmul_34 + x_391 = addcmul_34 = None + x_393 = torch._C._nn.linear( + x_392, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_392 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_394 = torch.nn.functional.dropout(x_393, 0.0, False, False) + x_393 = None + x_395 = x_394.permute(0, 3, 1, 2) + x_394 = None + x_396 = x_395 + x_385 + x_395 = x_385 = None + x_397 = torch.conv2d( + x_396, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_398 = x_397.permute(0, 2, 3, 1) + x_397 = None + x_399 = torch.nn.functional.layer_norm( + x_398, + (1024,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_398 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_400 = torch._C._nn.linear( + x_399, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_399 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_401 = torch._C._nn.gelu(x_400) + x_400 = None + x_402 = torch.nn.functional.dropout(x_401, 0.0, False, False) + x_401 = None + x_g_35 = x_402.norm(p=2, dim=(1, 2), keepdim=True) + mean_35 = x_g_35.mean(dim=-1, keepdim=True) + add_105 = mean_35 + 1e-06 + mean_35 = None + x_n_35 = x_g_35 / add_105 + x_g_35 = add_105 = None + view_70 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_71 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_35 = x_402 * x_n_35 + x_n_35 = None + addcmul_35 = torch.addcmul(view_70, view_71, mul_35) + view_70 = view_71 = mul_35 = None + x_403 = x_402 + addcmul_35 + x_402 = addcmul_35 = None + x_404 = torch._C._nn.linear( + x_403, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_403 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_405 = torch.nn.functional.dropout(x_404, 0.0, False, False) + x_404 = None + x_406 = x_405.permute(0, 3, 1, 2) + x_405 = None + x_407 = x_406 + x_396 + x_406 = x_396 = None + x_408 = torch.nn.functional.adaptive_avg_pool2d(x_407, 1) + x_407 = None + x_409 = x_408.permute(0, 2, 3, 1) + x_408 = None + x_410 = torch.nn.functional.layer_norm( + x_409, + (1024,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_409 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_411 = x_410.permute(0, 3, 1, 2) + x_410 = None + x_412 = x_411.flatten(1, -1) + x_411 = None + x_413 = torch.nn.functional.dropout(x_412, 0.0, False, False) + x_412 = None + return (x_413,) diff --git a/samples/timm/convnextv2_base.fcmae/weight_meta.py b/samples/timm/convnextv2_base.fcmae/weight_meta.py new file mode 100644 index 0000000000..c4cd9b38d1 --- /dev/null +++ b/samples/timm/convnextv2_base.fcmae/weight_meta.py @@ -0,0 +1,3810 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [128, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.226 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [256, 128, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [512, 256, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [1024, 512, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_base.fcmae_ft_in1k/graph_hash.txt b/samples/timm/convnextv2_base.fcmae_ft_in1k/graph_hash.txt new file mode 100644 index 0000000000..743b4867d6 --- /dev/null +++ b/samples/timm/convnextv2_base.fcmae_ft_in1k/graph_hash.txt @@ -0,0 +1 @@ +8d87ff876e26add5d5c71dcfb9bc70367e838d0b8e8ac6cb2562fd7661d035a5 \ No newline at end of file diff --git a/samples/timm/convnextv2_base.fcmae_ft_in1k/graph_net.json b/samples/timm/convnextv2_base.fcmae_ft_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnextv2_base.fcmae_ft_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_base.fcmae_ft_in1k/input_meta.py b/samples/timm/convnextv2_base.fcmae_ft_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_base.fcmae_ft_in1k/input_tensor_constraints.py b/samples/timm/convnextv2_base.fcmae_ft_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_base.fcmae_ft_in1k/model.py b/samples/timm/convnextv2_base.fcmae_ft_in1k/model.py new file mode 100644 index 0000000000..09c969a054 --- /dev/null +++ b/samples/timm/convnextv2_base.fcmae_ft_in1k/model.py @@ -0,0 +1,3283 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (128,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_g = x_8.norm(p=2, dim=(1, 2), keepdim=True) + mean = x_g.mean(dim=-1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_8 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_9 = x_8 + addcmul + x_8 = addcmul = None + x_10 = torch._C._nn.linear( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = x_11.permute(0, 3, 1, 2) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_g_1 = x_19.norm(p=2, dim=(1, 2), keepdim=True) + mean_1 = x_g_1.mean(dim=-1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_19 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_20 = x_19 + addcmul_1 + x_19 = addcmul_1 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (128,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_g_2 = x_30.norm(p=2, dim=(1, 2), keepdim=True) + mean_2 = x_g_2.mean(dim=-1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_30 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_31 = x_30 + addcmul_2 + x_30 = addcmul_2 = None + x_32 = torch._C._nn.linear( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + x_34 = x_33.permute(0, 3, 1, 2) + x_33 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (128,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_g_3 = x_44.norm(p=2, dim=(1, 2), keepdim=True) + mean_3 = x_g_3.mean(dim=-1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_44 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_45 = x_44 + addcmul_3 + x_44 = addcmul_3 = None + x_46 = torch._C._nn.linear( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_g_4 = x_55.norm(p=2, dim=(1, 2), keepdim=True) + mean_4 = x_g_4.mean(dim=-1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_55 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_56 = x_55 + addcmul_4 + x_55 = addcmul_4 = None + x_57 = torch._C._nn.linear( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_58 = torch.nn.functional.dropout(x_57, 0.0, False, False) + x_57 = None + x_59 = x_58.permute(0, 3, 1, 2) + x_58 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (256,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_g_5 = x_66.norm(p=2, dim=(1, 2), keepdim=True) + mean_5 = x_g_5.mean(dim=-1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_66 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_67 = x_66 + addcmul_5 + x_66 = addcmul_5 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = x_69.permute(0, 3, 1, 2) + x_69 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (256,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_g_6 = x_80.norm(p=2, dim=(1, 2), keepdim=True) + mean_6 = x_g_6.mean(dim=-1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_80 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_81 = x_80 + addcmul_6 + x_80 = addcmul_6 = None + x_82 = torch._C._nn.linear( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + x_84 = x_83.permute(0, 3, 1, 2) + x_83 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_g_7 = x_91.norm(p=2, dim=(1, 2), keepdim=True) + mean_7 = x_g_7.mean(dim=-1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_91 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_92 = x_91 + addcmul_7 + x_91 = addcmul_7 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_g_8 = x_102.norm(p=2, dim=(1, 2), keepdim=True) + mean_8 = x_g_8.mean(dim=-1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_102 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_103 = x_102 + addcmul_8 + x_102 = addcmul_8 = None + x_104 = torch._C._nn.linear( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = x_105.permute(0, 3, 1, 2) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_g_9 = x_113.norm(p=2, dim=(1, 2), keepdim=True) + mean_9 = x_g_9.mean(dim=-1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_113 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_114 = x_113 + addcmul_9 + x_113 = addcmul_9 = None + x_115 = torch._C._nn.linear( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + x_117 = x_116.permute(0, 3, 1, 2) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_g_10 = x_124.norm(p=2, dim=(1, 2), keepdim=True) + mean_10 = x_g_10.mean(dim=-1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_124 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_125 = x_124 + addcmul_10 + x_124 = addcmul_10 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = x_127.permute(0, 3, 1, 2) + x_127 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_g_11 = x_135.norm(p=2, dim=(1, 2), keepdim=True) + mean_11 = x_g_11.mean(dim=-1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_135 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_136 = x_135 + addcmul_11 + x_135 = addcmul_11 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_138.permute(0, 3, 1, 2) + x_138 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_g_12 = x_146.norm(p=2, dim=(1, 2), keepdim=True) + mean_12 = x_g_12.mean(dim=-1, keepdim=True) + add_36 = mean_12 + 1e-06 + mean_12 = None + x_n_12 = x_g_12 / add_36 + x_g_12 = add_36 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_12 = x_146 * x_n_12 + x_n_12 = None + addcmul_12 = torch.addcmul(view_24, view_25, mul_12) + view_24 = view_25 = mul_12 = None + x_147 = x_146 + addcmul_12 + x_146 = addcmul_12 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_149.permute(0, 3, 1, 2) + x_149 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_g_13 = x_157.norm(p=2, dim=(1, 2), keepdim=True) + mean_13 = x_g_13.mean(dim=-1, keepdim=True) + add_39 = mean_13 + 1e-06 + mean_13 = None + x_n_13 = x_g_13 / add_39 + x_g_13 = add_39 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_13 = x_157 * x_n_13 + x_n_13 = None + addcmul_13 = torch.addcmul(view_26, view_27, mul_13) + view_26 = view_27 = mul_13 = None + x_158 = x_157 + addcmul_13 + x_157 = addcmul_13 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_160.permute(0, 3, 1, 2) + x_160 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_g_14 = x_168.norm(p=2, dim=(1, 2), keepdim=True) + mean_14 = x_g_14.mean(dim=-1, keepdim=True) + add_42 = mean_14 + 1e-06 + mean_14 = None + x_n_14 = x_g_14 / add_42 + x_g_14 = add_42 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_14 = x_168 * x_n_14 + x_n_14 = None + addcmul_14 = torch.addcmul(view_28, view_29, mul_14) + view_28 = view_29 = mul_14 = None + x_169 = x_168 + addcmul_14 + x_168 = addcmul_14 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_171.permute(0, 3, 1, 2) + x_171 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = (None) + x_175 = x_174.permute(0, 2, 3, 1) + x_174 = None + x_176 = torch.nn.functional.layer_norm( + x_175, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_, + 1e-06, + ) + x_175 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = (None) + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_176 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_178 = torch._C._nn.gelu(x_177) + x_177 = None + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_g_15 = x_179.norm(p=2, dim=(1, 2), keepdim=True) + mean_15 = x_g_15.mean(dim=-1, keepdim=True) + add_45 = mean_15 + 1e-06 + mean_15 = None + x_n_15 = x_g_15 / add_45 + x_g_15 = add_45 = None + view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_15 = x_179 * x_n_15 + x_n_15 = None + addcmul_15 = torch.addcmul(view_30, view_31, mul_15) + view_30 = view_31 = mul_15 = None + x_180 = x_179 + addcmul_15 + x_179 = addcmul_15 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_182.permute(0, 3, 1, 2) + x_182 = None + x_184 = x_183 + x_173 + x_183 = x_173 = None + x_185 = torch.conv2d( + x_184, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = (None) + x_186 = x_185.permute(0, 2, 3, 1) + x_185 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = (None) + x_188 = torch._C._nn.linear( + x_187, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_187 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_189 = torch._C._nn.gelu(x_188) + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_g_16 = x_190.norm(p=2, dim=(1, 2), keepdim=True) + mean_16 = x_g_16.mean(dim=-1, keepdim=True) + add_48 = mean_16 + 1e-06 + mean_16 = None + x_n_16 = x_g_16 / add_48 + x_g_16 = add_48 = None + view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_16 = x_190 * x_n_16 + x_n_16 = None + addcmul_16 = torch.addcmul(view_32, view_33, mul_16) + view_32 = view_33 = mul_16 = None + x_191 = x_190 + addcmul_16 + x_190 = addcmul_16 = None + x_192 = torch._C._nn.linear( + x_191, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_191 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_193.permute(0, 3, 1, 2) + x_193 = None + x_195 = x_194 + x_184 + x_194 = x_184 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = (None) + x_197 = x_196.permute(0, 2, 3, 1) + x_196 = None + x_198 = torch.nn.functional.layer_norm( + x_197, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_, + 1e-06, + ) + x_197 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = (None) + x_199 = torch._C._nn.linear( + x_198, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_198 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_200 = torch._C._nn.gelu(x_199) + x_199 = None + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_g_17 = x_201.norm(p=2, dim=(1, 2), keepdim=True) + mean_17 = x_g_17.mean(dim=-1, keepdim=True) + add_51 = mean_17 + 1e-06 + mean_17 = None + x_n_17 = x_g_17 / add_51 + x_g_17 = add_51 = None + view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_17 = x_201 * x_n_17 + x_n_17 = None + addcmul_17 = torch.addcmul(view_34, view_35, mul_17) + view_34 = view_35 = mul_17 = None + x_202 = x_201 + addcmul_17 + x_201 = addcmul_17 = None + x_203 = torch._C._nn.linear( + x_202, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_202 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_205 = x_204.permute(0, 3, 1, 2) + x_204 = None + x_206 = x_205 + x_195 + x_205 = x_195 = None + x_207 = torch.conv2d( + x_206, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = (None) + x_208 = x_207.permute(0, 2, 3, 1) + x_207 = None + x_209 = torch.nn.functional.layer_norm( + x_208, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_, + 1e-06, + ) + x_208 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = (None) + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_209 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_211 = torch._C._nn.gelu(x_210) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_g_18 = x_212.norm(p=2, dim=(1, 2), keepdim=True) + mean_18 = x_g_18.mean(dim=-1, keepdim=True) + add_54 = mean_18 + 1e-06 + mean_18 = None + x_n_18 = x_g_18 / add_54 + x_g_18 = add_54 = None + view_36 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_37 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_18 = x_212 * x_n_18 + x_n_18 = None + addcmul_18 = torch.addcmul(view_36, view_37, mul_18) + view_36 = view_37 = mul_18 = None + x_213 = x_212 + addcmul_18 + x_212 = addcmul_18 = None + x_214 = torch._C._nn.linear( + x_213, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_213 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = x_215.permute(0, 3, 1, 2) + x_215 = None + x_217 = x_216 + x_206 + x_216 = x_206 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = (None) + x_219 = x_218.permute(0, 2, 3, 1) + x_218 = None + x_220 = torch.nn.functional.layer_norm( + x_219, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_, + 1e-06, + ) + x_219 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = (None) + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_220 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_222 = torch._C._nn.gelu(x_221) + x_221 = None + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_g_19 = x_223.norm(p=2, dim=(1, 2), keepdim=True) + mean_19 = x_g_19.mean(dim=-1, keepdim=True) + add_57 = mean_19 + 1e-06 + mean_19 = None + x_n_19 = x_g_19 / add_57 + x_g_19 = add_57 = None + view_38 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_39 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_19 = x_223 * x_n_19 + x_n_19 = None + addcmul_19 = torch.addcmul(view_38, view_39, mul_19) + view_38 = view_39 = mul_19 = None + x_224 = x_223 + addcmul_19 + x_223 = addcmul_19 = None + x_225 = torch._C._nn.linear( + x_224, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_224 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_226 = torch.nn.functional.dropout(x_225, 0.0, False, False) + x_225 = None + x_227 = x_226.permute(0, 3, 1, 2) + x_226 = None + x_228 = x_227 + x_217 + x_227 = x_217 = None + x_229 = torch.conv2d( + x_228, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = (None) + x_230 = x_229.permute(0, 2, 3, 1) + x_229 = None + x_231 = torch.nn.functional.layer_norm( + x_230, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_, + 1e-06, + ) + x_230 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = (None) + x_232 = torch._C._nn.linear( + x_231, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_231 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_233 = torch._C._nn.gelu(x_232) + x_232 = None + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_g_20 = x_234.norm(p=2, dim=(1, 2), keepdim=True) + mean_20 = x_g_20.mean(dim=-1, keepdim=True) + add_60 = mean_20 + 1e-06 + mean_20 = None + x_n_20 = x_g_20 / add_60 + x_g_20 = add_60 = None + view_40 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_41 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_20 = x_234 * x_n_20 + x_n_20 = None + addcmul_20 = torch.addcmul(view_40, view_41, mul_20) + view_40 = view_41 = mul_20 = None + x_235 = x_234 + addcmul_20 + x_234 = addcmul_20 = None + x_236 = torch._C._nn.linear( + x_235, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_235 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_237 = torch.nn.functional.dropout(x_236, 0.0, False, False) + x_236 = None + x_238 = x_237.permute(0, 3, 1, 2) + x_237 = None + x_239 = x_238 + x_228 + x_238 = x_228 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = (None) + x_241 = x_240.permute(0, 2, 3, 1) + x_240 = None + x_242 = torch.nn.functional.layer_norm( + x_241, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_, + 1e-06, + ) + x_241 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = (None) + x_243 = torch._C._nn.linear( + x_242, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_242 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_244 = torch._C._nn.gelu(x_243) + x_243 = None + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_g_21 = x_245.norm(p=2, dim=(1, 2), keepdim=True) + mean_21 = x_g_21.mean(dim=-1, keepdim=True) + add_63 = mean_21 + 1e-06 + mean_21 = None + x_n_21 = x_g_21 / add_63 + x_g_21 = add_63 = None + view_42 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_43 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_21 = x_245 * x_n_21 + x_n_21 = None + addcmul_21 = torch.addcmul(view_42, view_43, mul_21) + view_42 = view_43 = mul_21 = None + x_246 = x_245 + addcmul_21 + x_245 = addcmul_21 = None + x_247 = torch._C._nn.linear( + x_246, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_246 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_248 = torch.nn.functional.dropout(x_247, 0.0, False, False) + x_247 = None + x_249 = x_248.permute(0, 3, 1, 2) + x_248 = None + x_250 = x_249 + x_239 + x_249 = x_239 = None + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = (None) + x_252 = x_251.permute(0, 2, 3, 1) + x_251 = None + x_253 = torch.nn.functional.layer_norm( + x_252, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_, + 1e-06, + ) + x_252 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = (None) + x_254 = torch._C._nn.linear( + x_253, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_253 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_255 = torch._C._nn.gelu(x_254) + x_254 = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_g_22 = x_256.norm(p=2, dim=(1, 2), keepdim=True) + mean_22 = x_g_22.mean(dim=-1, keepdim=True) + add_66 = mean_22 + 1e-06 + mean_22 = None + x_n_22 = x_g_22 / add_66 + x_g_22 = add_66 = None + view_44 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_45 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_22 = x_256 * x_n_22 + x_n_22 = None + addcmul_22 = torch.addcmul(view_44, view_45, mul_22) + view_44 = view_45 = mul_22 = None + x_257 = x_256 + addcmul_22 + x_256 = addcmul_22 = None + x_258 = torch._C._nn.linear( + x_257, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_257 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = x_259.permute(0, 3, 1, 2) + x_259 = None + x_261 = x_260 + x_250 + x_260 = x_250 = None + x_262 = torch.conv2d( + x_261, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = (None) + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = (None) + x_265 = torch._C._nn.linear( + x_264, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_264 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_266 = torch._C._nn.gelu(x_265) + x_265 = None + x_267 = torch.nn.functional.dropout(x_266, 0.0, False, False) + x_266 = None + x_g_23 = x_267.norm(p=2, dim=(1, 2), keepdim=True) + mean_23 = x_g_23.mean(dim=-1, keepdim=True) + add_69 = mean_23 + 1e-06 + mean_23 = None + x_n_23 = x_g_23 / add_69 + x_g_23 = add_69 = None + view_46 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_47 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_23 = x_267 * x_n_23 + x_n_23 = None + addcmul_23 = torch.addcmul(view_46, view_47, mul_23) + view_46 = view_47 = mul_23 = None + x_268 = x_267 + addcmul_23 + x_267 = addcmul_23 = None + x_269 = torch._C._nn.linear( + x_268, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_268 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_270 = torch.nn.functional.dropout(x_269, 0.0, False, False) + x_269 = None + x_271 = x_270.permute(0, 3, 1, 2) + x_270 = None + x_272 = x_271 + x_261 + x_271 = x_261 = None + x_273 = torch.conv2d( + x_272, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = (None) + x_274 = x_273.permute(0, 2, 3, 1) + x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_, + 1e-06, + ) + x_274 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = (None) + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_277 = torch._C._nn.gelu(x_276) + x_276 = None + x_278 = torch.nn.functional.dropout(x_277, 0.0, False, False) + x_277 = None + x_g_24 = x_278.norm(p=2, dim=(1, 2), keepdim=True) + mean_24 = x_g_24.mean(dim=-1, keepdim=True) + add_72 = mean_24 + 1e-06 + mean_24 = None + x_n_24 = x_g_24 / add_72 + x_g_24 = add_72 = None + view_48 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_49 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_24 = x_278 * x_n_24 + x_n_24 = None + addcmul_24 = torch.addcmul(view_48, view_49, mul_24) + view_48 = view_49 = mul_24 = None + x_279 = x_278 + addcmul_24 + x_278 = addcmul_24 = None + x_280 = torch._C._nn.linear( + x_279, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_279 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_281 = torch.nn.functional.dropout(x_280, 0.0, False, False) + x_280 = None + x_282 = x_281.permute(0, 3, 1, 2) + x_281 = None + x_283 = x_282 + x_272 + x_282 = x_272 = None + x_284 = torch.conv2d( + x_283, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = (None) + x_285 = x_284.permute(0, 2, 3, 1) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_, + 1e-06, + ) + x_285 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = (None) + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_288 = torch._C._nn.gelu(x_287) + x_287 = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + x_g_25 = x_289.norm(p=2, dim=(1, 2), keepdim=True) + mean_25 = x_g_25.mean(dim=-1, keepdim=True) + add_75 = mean_25 + 1e-06 + mean_25 = None + x_n_25 = x_g_25 / add_75 + x_g_25 = add_75 = None + view_50 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_51 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_25 = x_289 * x_n_25 + x_n_25 = None + addcmul_25 = torch.addcmul(view_50, view_51, mul_25) + view_50 = view_51 = mul_25 = None + x_290 = x_289 + addcmul_25 + x_289 = addcmul_25 = None + x_291 = torch._C._nn.linear( + x_290, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_290 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_292 = torch.nn.functional.dropout(x_291, 0.0, False, False) + x_291 = None + x_293 = x_292.permute(0, 3, 1, 2) + x_292 = None + x_294 = x_293 + x_283 + x_293 = x_283 = None + x_295 = torch.conv2d( + x_294, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = (None) + x_296 = x_295.permute(0, 2, 3, 1) + x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_, + 1e-06, + ) + x_296 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = (None) + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_299 = torch._C._nn.gelu(x_298) + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_g_26 = x_300.norm(p=2, dim=(1, 2), keepdim=True) + mean_26 = x_g_26.mean(dim=-1, keepdim=True) + add_78 = mean_26 + 1e-06 + mean_26 = None + x_n_26 = x_g_26 / add_78 + x_g_26 = add_78 = None + view_52 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_53 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_26 = x_300 * x_n_26 + x_n_26 = None + addcmul_26 = torch.addcmul(view_52, view_53, mul_26) + view_52 = view_53 = mul_26 = None + x_301 = x_300 + addcmul_26 + x_300 = addcmul_26 = None + x_302 = torch._C._nn.linear( + x_301, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_301 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_303 = torch.nn.functional.dropout(x_302, 0.0, False, False) + x_302 = None + x_304 = x_303.permute(0, 3, 1, 2) + x_303 = None + x_305 = x_304 + x_294 + x_304 = x_294 = None + x_306 = torch.conv2d( + x_305, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = (None) + x_307 = x_306.permute(0, 2, 3, 1) + x_306 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_, + 1e-06, + ) + x_307 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = (None) + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_310 = torch._C._nn.gelu(x_309) + x_309 = None + x_311 = torch.nn.functional.dropout(x_310, 0.0, False, False) + x_310 = None + x_g_27 = x_311.norm(p=2, dim=(1, 2), keepdim=True) + mean_27 = x_g_27.mean(dim=-1, keepdim=True) + add_81 = mean_27 + 1e-06 + mean_27 = None + x_n_27 = x_g_27 / add_81 + x_g_27 = add_81 = None + view_54 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_55 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_27 = x_311 * x_n_27 + x_n_27 = None + addcmul_27 = torch.addcmul(view_54, view_55, mul_27) + view_54 = view_55 = mul_27 = None + x_312 = x_311 + addcmul_27 + x_311 = addcmul_27 = None + x_313 = torch._C._nn.linear( + x_312, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_312 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_314 = torch.nn.functional.dropout(x_313, 0.0, False, False) + x_313 = None + x_315 = x_314.permute(0, 3, 1, 2) + x_314 = None + x_316 = x_315 + x_305 + x_315 = x_305 = None + x_317 = torch.conv2d( + x_316, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = (None) + x_318 = x_317.permute(0, 2, 3, 1) + x_317 = None + x_319 = torch.nn.functional.layer_norm( + x_318, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_, + 1e-06, + ) + x_318 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = (None) + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_321 = torch._C._nn.gelu(x_320) + x_320 = None + x_322 = torch.nn.functional.dropout(x_321, 0.0, False, False) + x_321 = None + x_g_28 = x_322.norm(p=2, dim=(1, 2), keepdim=True) + mean_28 = x_g_28.mean(dim=-1, keepdim=True) + add_84 = mean_28 + 1e-06 + mean_28 = None + x_n_28 = x_g_28 / add_84 + x_g_28 = add_84 = None + view_56 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_57 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_28 = x_322 * x_n_28 + x_n_28 = None + addcmul_28 = torch.addcmul(view_56, view_57, mul_28) + view_56 = view_57 = mul_28 = None + x_323 = x_322 + addcmul_28 + x_322 = addcmul_28 = None + x_324 = torch._C._nn.linear( + x_323, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_323 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_325 = torch.nn.functional.dropout(x_324, 0.0, False, False) + x_324 = None + x_326 = x_325.permute(0, 3, 1, 2) + x_325 = None + x_327 = x_326 + x_316 + x_326 = x_316 = None + x_328 = torch.conv2d( + x_327, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = (None) + x_329 = x_328.permute(0, 2, 3, 1) + x_328 = None + x_330 = torch.nn.functional.layer_norm( + x_329, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_, + 1e-06, + ) + x_329 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = (None) + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_332 = torch._C._nn.gelu(x_331) + x_331 = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_g_29 = x_333.norm(p=2, dim=(1, 2), keepdim=True) + mean_29 = x_g_29.mean(dim=-1, keepdim=True) + add_87 = mean_29 + 1e-06 + mean_29 = None + x_n_29 = x_g_29 / add_87 + x_g_29 = add_87 = None + view_58 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_59 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_29 = x_333 * x_n_29 + x_n_29 = None + addcmul_29 = torch.addcmul(view_58, view_59, mul_29) + view_58 = view_59 = mul_29 = None + x_334 = x_333 + addcmul_29 + x_333 = addcmul_29 = None + x_335 = torch._C._nn.linear( + x_334, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_334 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_336 = torch.nn.functional.dropout(x_335, 0.0, False, False) + x_335 = None + x_337 = x_336.permute(0, 3, 1, 2) + x_336 = None + x_338 = x_337 + x_327 + x_337 = x_327 = None + x_339 = torch.conv2d( + x_338, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = (None) + x_340 = x_339.permute(0, 2, 3, 1) + x_339 = None + x_341 = torch.nn.functional.layer_norm( + x_340, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_, + 1e-06, + ) + x_340 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = (None) + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_343 = torch._C._nn.gelu(x_342) + x_342 = None + x_344 = torch.nn.functional.dropout(x_343, 0.0, False, False) + x_343 = None + x_g_30 = x_344.norm(p=2, dim=(1, 2), keepdim=True) + mean_30 = x_g_30.mean(dim=-1, keepdim=True) + add_90 = mean_30 + 1e-06 + mean_30 = None + x_n_30 = x_g_30 / add_90 + x_g_30 = add_90 = None + view_60 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_61 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_30 = x_344 * x_n_30 + x_n_30 = None + addcmul_30 = torch.addcmul(view_60, view_61, mul_30) + view_60 = view_61 = mul_30 = None + x_345 = x_344 + addcmul_30 + x_344 = addcmul_30 = None + x_346 = torch._C._nn.linear( + x_345, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_345 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_347 = torch.nn.functional.dropout(x_346, 0.0, False, False) + x_346 = None + x_348 = x_347.permute(0, 3, 1, 2) + x_347 = None + x_349 = x_348 + x_338 + x_348 = x_338 = None + x_350 = torch.conv2d( + x_349, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = (None) + x_351 = x_350.permute(0, 2, 3, 1) + x_350 = None + x_352 = torch.nn.functional.layer_norm( + x_351, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_, + 1e-06, + ) + x_351 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = (None) + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_354 = torch._C._nn.gelu(x_353) + x_353 = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_g_31 = x_355.norm(p=2, dim=(1, 2), keepdim=True) + mean_31 = x_g_31.mean(dim=-1, keepdim=True) + add_93 = mean_31 + 1e-06 + mean_31 = None + x_n_31 = x_g_31 / add_93 + x_g_31 = add_93 = None + view_62 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_63 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_31 = x_355 * x_n_31 + x_n_31 = None + addcmul_31 = torch.addcmul(view_62, view_63, mul_31) + view_62 = view_63 = mul_31 = None + x_356 = x_355 + addcmul_31 + x_355 = addcmul_31 = None + x_357 = torch._C._nn.linear( + x_356, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_356 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_358 = torch.nn.functional.dropout(x_357, 0.0, False, False) + x_357 = None + x_359 = x_358.permute(0, 3, 1, 2) + x_358 = None + x_360 = x_359 + x_349 + x_359 = x_349 = None + x_361 = torch.conv2d( + x_360, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = (None) + x_362 = x_361.permute(0, 2, 3, 1) + x_361 = None + x_363 = torch.nn.functional.layer_norm( + x_362, + (512,), + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_, + 1e-06, + ) + x_362 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = (None) + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_365 = torch._C._nn.gelu(x_364) + x_364 = None + x_366 = torch.nn.functional.dropout(x_365, 0.0, False, False) + x_365 = None + x_g_32 = x_366.norm(p=2, dim=(1, 2), keepdim=True) + mean_32 = x_g_32.mean(dim=-1, keepdim=True) + add_96 = mean_32 + 1e-06 + mean_32 = None + x_n_32 = x_g_32 / add_96 + x_g_32 = add_96 = None + view_64 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_65 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_32 = x_366 * x_n_32 + x_n_32 = None + addcmul_32 = torch.addcmul(view_64, view_65, mul_32) + view_64 = view_65 = mul_32 = None + x_367 = x_366 + addcmul_32 + x_366 = addcmul_32 = None + x_368 = torch._C._nn.linear( + x_367, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_367 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_369 = torch.nn.functional.dropout(x_368, 0.0, False, False) + x_368 = None + x_370 = x_369.permute(0, 3, 1, 2) + x_369 = None + x_371 = x_370 + x_360 + x_370 = x_360 = None + x_372 = x_371.permute(0, 2, 3, 1) + x_371 = None + x_373 = torch.nn.functional.layer_norm( + x_372, + (512,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_372 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_374 = x_373.permute(0, 3, 1, 2) + x_373 = None + input_4 = torch.conv2d( + x_374, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_374 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_375 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_376 = x_375.permute(0, 2, 3, 1) + x_375 = None + x_377 = torch.nn.functional.layer_norm( + x_376, + (1024,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_376 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_378 = torch._C._nn.linear( + x_377, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_377 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_379 = torch._C._nn.gelu(x_378) + x_378 = None + x_380 = torch.nn.functional.dropout(x_379, 0.0, False, False) + x_379 = None + x_g_33 = x_380.norm(p=2, dim=(1, 2), keepdim=True) + mean_33 = x_g_33.mean(dim=-1, keepdim=True) + add_99 = mean_33 + 1e-06 + mean_33 = None + x_n_33 = x_g_33 / add_99 + x_g_33 = add_99 = None + view_66 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_67 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_33 = x_380 * x_n_33 + x_n_33 = None + addcmul_33 = torch.addcmul(view_66, view_67, mul_33) + view_66 = view_67 = mul_33 = None + x_381 = x_380 + addcmul_33 + x_380 = addcmul_33 = None + x_382 = torch._C._nn.linear( + x_381, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_381 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_383 = torch.nn.functional.dropout(x_382, 0.0, False, False) + x_382 = None + x_384 = x_383.permute(0, 3, 1, 2) + x_383 = None + x_385 = x_384 + input_4 + x_384 = input_4 = None + x_386 = torch.conv2d( + x_385, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_387 = x_386.permute(0, 2, 3, 1) + x_386 = None + x_388 = torch.nn.functional.layer_norm( + x_387, + (1024,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_387 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_389 = torch._C._nn.linear( + x_388, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_388 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_390 = torch._C._nn.gelu(x_389) + x_389 = None + x_391 = torch.nn.functional.dropout(x_390, 0.0, False, False) + x_390 = None + x_g_34 = x_391.norm(p=2, dim=(1, 2), keepdim=True) + mean_34 = x_g_34.mean(dim=-1, keepdim=True) + add_102 = mean_34 + 1e-06 + mean_34 = None + x_n_34 = x_g_34 / add_102 + x_g_34 = add_102 = None + view_68 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_69 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_34 = x_391 * x_n_34 + x_n_34 = None + addcmul_34 = torch.addcmul(view_68, view_69, mul_34) + view_68 = view_69 = mul_34 = None + x_392 = x_391 + addcmul_34 + x_391 = addcmul_34 = None + x_393 = torch._C._nn.linear( + x_392, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_392 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_394 = torch.nn.functional.dropout(x_393, 0.0, False, False) + x_393 = None + x_395 = x_394.permute(0, 3, 1, 2) + x_394 = None + x_396 = x_395 + x_385 + x_395 = x_385 = None + x_397 = torch.conv2d( + x_396, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1024, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_398 = x_397.permute(0, 2, 3, 1) + x_397 = None + x_399 = torch.nn.functional.layer_norm( + x_398, + (1024,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_398 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_400 = torch._C._nn.linear( + x_399, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_399 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_401 = torch._C._nn.gelu(x_400) + x_400 = None + x_402 = torch.nn.functional.dropout(x_401, 0.0, False, False) + x_401 = None + x_g_35 = x_402.norm(p=2, dim=(1, 2), keepdim=True) + mean_35 = x_g_35.mean(dim=-1, keepdim=True) + add_105 = mean_35 + 1e-06 + mean_35 = None + x_n_35 = x_g_35 / add_105 + x_g_35 = add_105 = None + view_70 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_71 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_35 = x_402 * x_n_35 + x_n_35 = None + addcmul_35 = torch.addcmul(view_70, view_71, mul_35) + view_70 = view_71 = mul_35 = None + x_403 = x_402 + addcmul_35 + x_402 = addcmul_35 = None + x_404 = torch._C._nn.linear( + x_403, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_403 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_405 = torch.nn.functional.dropout(x_404, 0.0, False, False) + x_404 = None + x_406 = x_405.permute(0, 3, 1, 2) + x_405 = None + x_407 = x_406 + x_396 + x_406 = x_396 = None + x_408 = torch.nn.functional.adaptive_avg_pool2d(x_407, 1) + x_407 = None + x_409 = x_408.permute(0, 2, 3, 1) + x_408 = None + x_410 = torch.nn.functional.layer_norm( + x_409, + (1024,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_409 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_411 = x_410.permute(0, 3, 1, 2) + x_410 = None + x_412 = x_411.flatten(1, -1) + x_411 = None + x_413 = torch.nn.functional.dropout(x_412, 0.0, False, False) + x_412 = None + x_414 = torch._C._nn.linear( + x_413, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_413 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_414,) diff --git a/samples/timm/convnextv2_base.fcmae_ft_in1k/weight_meta.py b/samples/timm/convnextv2_base.fcmae_ft_in1k/weight_meta.py new file mode 100644 index 0000000000..d29845ffbc --- /dev/null +++ b/samples/timm/convnextv2_base.fcmae_ft_in1k/weight_meta.py @@ -0,0 +1,3830 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [128, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.222 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [256, 128, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [512, 256, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [1024, 512, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [1024, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_femto.fcmae/graph_hash.txt b/samples/timm/convnextv2_femto.fcmae/graph_hash.txt new file mode 100644 index 0000000000..e454cfc2b7 --- /dev/null +++ b/samples/timm/convnextv2_femto.fcmae/graph_hash.txt @@ -0,0 +1 @@ +e5108a1f4678e444be04e203895b2668c3c7af06626cfc8faa2497b31f31b758 \ No newline at end of file diff --git a/samples/timm/convnextv2_femto.fcmae/graph_net.json b/samples/timm/convnextv2_femto.fcmae/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnextv2_femto.fcmae/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_femto.fcmae/input_meta.py b/samples/timm/convnextv2_femto.fcmae/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_femto.fcmae/input_tensor_constraints.py b/samples/timm/convnextv2_femto.fcmae/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_femto.fcmae/model.py b/samples/timm/convnextv2_femto.fcmae/model.py new file mode 100644 index 0000000000..fdb6c19275 --- /dev/null +++ b/samples/timm/convnextv2_femto.fcmae/model.py @@ -0,0 +1,1299 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (48,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 48, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (48,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_g = x_9.norm(p=2, dim=(2, 3), keepdim=True) + mean = x_g.mean(dim=1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_9 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_10 = x_9 + addcmul + x_9 = addcmul = None + x_11 = torch.conv2d( + x_10, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_10 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 48, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (48,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_g_1 = x_20.norm(p=2, dim=(2, 3), keepdim=True) + mean_1 = x_g_1.mean(dim=1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_20 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_21 = x_20 + addcmul_1 + x_20 = addcmul_1 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (48,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_2 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (96,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_g_2 = x_34.norm(p=2, dim=(2, 3), keepdim=True) + mean_2 = x_g_2.mean(dim=1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_34 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_35 = x_34 + addcmul_2 + x_34 = addcmul_2 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_37 = torch.nn.functional.dropout(x_36, 0.0, False, False) + x_36 = None + x_38 = x_37 + input_2 + x_37 = input_2 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (96,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_g_3 = x_45.norm(p=2, dim=(2, 3), keepdim=True) + mean_3 = x_g_3.mean(dim=1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_45 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_46 = x_45 + addcmul_3 + x_45 = addcmul_3 = None + x_47 = torch.conv2d( + x_46, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (96,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_3 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_g_4 = x_59.norm(p=2, dim=(2, 3), keepdim=True) + mean_4 = x_g_4.mean(dim=1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_59 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_60 = x_59 + addcmul_4 + x_59 = addcmul_4 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_60 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + x_63 = x_62 + input_3 + x_62 = input_3 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_g_5 = x_70.norm(p=2, dim=(2, 3), keepdim=True) + mean_5 = x_g_5.mean(dim=1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_70 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_71 = x_70 + addcmul_5 + x_70 = addcmul_5 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_g_6 = x_81.norm(p=2, dim=(2, 3), keepdim=True) + mean_6 = x_g_6.mean(dim=1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_81 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_82 = x_81 + addcmul_6 + x_81 = addcmul_6 = None + x_83 = torch.conv2d( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_82 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_g_7 = x_92.norm(p=2, dim=(2, 3), keepdim=True) + mean_7 = x_g_7.mean(dim=1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_92 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_93 = x_92 + addcmul_7 + x_92 = addcmul_7 = None + x_94 = torch.conv2d( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_g_8 = x_103.norm(p=2, dim=(2, 3), keepdim=True) + mean_8 = x_g_8.mean(dim=1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_103 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_104 = x_103 + addcmul_8 + x_103 = addcmul_8 = None + x_105 = torch.conv2d( + x_104, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_104 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_g_9 = x_114.norm(p=2, dim=(2, 3), keepdim=True) + mean_9 = x_g_9.mean(dim=1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_114 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_115 = x_114 + addcmul_9 + x_114 = addcmul_9 = None + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = x_118.permute(0, 2, 3, 1) + x_118 = None + x_120 = torch.nn.functional.layer_norm( + x_119, + (192,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_119 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_121 = x_120.permute(0, 3, 1, 2) + x_120 = None + input_4 = torch.conv2d( + x_121, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_122 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_123 = x_122.permute(0, 2, 3, 1) + x_122 = None + x_124 = torch.nn.functional.layer_norm( + x_123, + (384,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_125 = x_124.permute(0, 3, 1, 2) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126) + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_g_10 = x_128.norm(p=2, dim=(2, 3), keepdim=True) + mean_10 = x_g_10.mean(dim=1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_128 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_129 = x_128 + addcmul_10 + x_128 = addcmul_10 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = x_131 + input_4 + x_131 = input_4 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_134 = x_133.permute(0, 2, 3, 1) + x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (384,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_134 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_136 = x_135.permute(0, 3, 1, 2) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137) + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_g_11 = x_139.norm(p=2, dim=(2, 3), keepdim=True) + mean_11 = x_g_11.mean(dim=1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_139 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_140 = x_139 + addcmul_11 + x_139 = addcmul_11 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_140 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_142 + x_132 + x_142 = x_132 = None + x_144 = torch.nn.functional.adaptive_avg_pool2d(x_143, 1) + x_143 = None + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (384,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = x_147.flatten(1, -1) + x_147 = None + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + return (x_149,) diff --git a/samples/timm/convnextv2_femto.fcmae/weight_meta.py b/samples/timm/convnextv2_femto.fcmae/weight_meta.py new file mode 100644 index 0000000000..fe65b35608 --- /dev/null +++ b/samples/timm/convnextv2_femto.fcmae/weight_meta.py @@ -0,0 +1,1410 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [48, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.223 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [48, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [192, 48, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [48, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [48, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [192, 48, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [48, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [96, 48, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [192, 96, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [384, 192, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_femto.fcmae_ft_in1k/graph_hash.txt b/samples/timm/convnextv2_femto.fcmae_ft_in1k/graph_hash.txt new file mode 100644 index 0000000000..e7bfa26e23 --- /dev/null +++ b/samples/timm/convnextv2_femto.fcmae_ft_in1k/graph_hash.txt @@ -0,0 +1 @@ +4a937f17830161ff371fbc0820658beeedd7e940c7d349c7a86885d72baab3a6 \ No newline at end of file diff --git a/samples/timm/convnextv2_femto.fcmae_ft_in1k/graph_net.json b/samples/timm/convnextv2_femto.fcmae_ft_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnextv2_femto.fcmae_ft_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_femto.fcmae_ft_in1k/input_meta.py b/samples/timm/convnextv2_femto.fcmae_ft_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_femto.fcmae_ft_in1k/input_tensor_constraints.py b/samples/timm/convnextv2_femto.fcmae_ft_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_femto.fcmae_ft_in1k/model.py b/samples/timm/convnextv2_femto.fcmae_ft_in1k/model.py new file mode 100644 index 0000000000..8cff47d96c --- /dev/null +++ b/samples/timm/convnextv2_femto.fcmae_ft_in1k/model.py @@ -0,0 +1,1315 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (48,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 48, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (48,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_g = x_9.norm(p=2, dim=(2, 3), keepdim=True) + mean = x_g.mean(dim=1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_9 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_10 = x_9 + addcmul + x_9 = addcmul = None + x_11 = torch.conv2d( + x_10, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_10 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 48, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (48,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_g_1 = x_20.norm(p=2, dim=(2, 3), keepdim=True) + mean_1 = x_g_1.mean(dim=1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_20 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_21 = x_20 + addcmul_1 + x_20 = addcmul_1 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (48,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_2 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (96,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_g_2 = x_34.norm(p=2, dim=(2, 3), keepdim=True) + mean_2 = x_g_2.mean(dim=1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_34 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_35 = x_34 + addcmul_2 + x_34 = addcmul_2 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_37 = torch.nn.functional.dropout(x_36, 0.0, False, False) + x_36 = None + x_38 = x_37 + input_2 + x_37 = input_2 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (96,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_g_3 = x_45.norm(p=2, dim=(2, 3), keepdim=True) + mean_3 = x_g_3.mean(dim=1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_45 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_46 = x_45 + addcmul_3 + x_45 = addcmul_3 = None + x_47 = torch.conv2d( + x_46, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (96,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_3 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_g_4 = x_59.norm(p=2, dim=(2, 3), keepdim=True) + mean_4 = x_g_4.mean(dim=1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_59 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_60 = x_59 + addcmul_4 + x_59 = addcmul_4 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_60 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + x_63 = x_62 + input_3 + x_62 = input_3 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_g_5 = x_70.norm(p=2, dim=(2, 3), keepdim=True) + mean_5 = x_g_5.mean(dim=1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_70 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_71 = x_70 + addcmul_5 + x_70 = addcmul_5 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_g_6 = x_81.norm(p=2, dim=(2, 3), keepdim=True) + mean_6 = x_g_6.mean(dim=1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_81 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_82 = x_81 + addcmul_6 + x_81 = addcmul_6 = None + x_83 = torch.conv2d( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_82 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_g_7 = x_92.norm(p=2, dim=(2, 3), keepdim=True) + mean_7 = x_g_7.mean(dim=1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_92 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_93 = x_92 + addcmul_7 + x_92 = addcmul_7 = None + x_94 = torch.conv2d( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_g_8 = x_103.norm(p=2, dim=(2, 3), keepdim=True) + mean_8 = x_g_8.mean(dim=1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_103 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_104 = x_103 + addcmul_8 + x_103 = addcmul_8 = None + x_105 = torch.conv2d( + x_104, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_104 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (192,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_g_9 = x_114.norm(p=2, dim=(2, 3), keepdim=True) + mean_9 = x_g_9.mean(dim=1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_114 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_115 = x_114 + addcmul_9 + x_114 = addcmul_9 = None + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = x_118.permute(0, 2, 3, 1) + x_118 = None + x_120 = torch.nn.functional.layer_norm( + x_119, + (192,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_119 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_121 = x_120.permute(0, 3, 1, 2) + x_120 = None + input_4 = torch.conv2d( + x_121, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_122 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_123 = x_122.permute(0, 2, 3, 1) + x_122 = None + x_124 = torch.nn.functional.layer_norm( + x_123, + (384,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_125 = x_124.permute(0, 3, 1, 2) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126) + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_g_10 = x_128.norm(p=2, dim=(2, 3), keepdim=True) + mean_10 = x_g_10.mean(dim=1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_128 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_129 = x_128 + addcmul_10 + x_128 = addcmul_10 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = x_131 + input_4 + x_131 = input_4 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_134 = x_133.permute(0, 2, 3, 1) + x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (384,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_134 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_136 = x_135.permute(0, 3, 1, 2) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137) + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_g_11 = x_139.norm(p=2, dim=(2, 3), keepdim=True) + mean_11 = x_g_11.mean(dim=1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_139 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_140 = x_139 + addcmul_11 + x_139 = addcmul_11 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_140 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_142 + x_132 + x_142 = x_132 = None + x_144 = torch.nn.functional.adaptive_avg_pool2d(x_143, 1) + x_143 = None + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (384,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = x_147.flatten(1, -1) + x_147 = None + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = torch._C._nn.linear( + x_149, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_149 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_150,) diff --git a/samples/timm/convnextv2_femto.fcmae_ft_in1k/weight_meta.py b/samples/timm/convnextv2_femto.fcmae_ft_in1k/weight_meta.py new file mode 100644 index 0000000000..634c1166e6 --- /dev/null +++ b/samples/timm/convnextv2_femto.fcmae_ft_in1k/weight_meta.py @@ -0,0 +1,1430 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [48, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.223 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [48, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [192, 48, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [48, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [48, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [192, 48, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [48, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [96, 48, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [192, 96, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [384, 192, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_huge.fcmae/graph_hash.txt b/samples/timm/convnextv2_huge.fcmae/graph_hash.txt new file mode 100644 index 0000000000..23cdc31b07 --- /dev/null +++ b/samples/timm/convnextv2_huge.fcmae/graph_hash.txt @@ -0,0 +1 @@ +8fcb4693ac7ae2f805b5c736406cfc3dfd8b7c5eeffaea4e083ab3f65af787d5 \ No newline at end of file diff --git a/samples/timm/convnextv2_huge.fcmae/graph_net.json b/samples/timm/convnextv2_huge.fcmae/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnextv2_huge.fcmae/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_huge.fcmae/input_meta.py b/samples/timm/convnextv2_huge.fcmae/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_huge.fcmae/input_tensor_constraints.py b/samples/timm/convnextv2_huge.fcmae/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_huge.fcmae/model.py b/samples/timm/convnextv2_huge.fcmae/model.py new file mode 100644 index 0000000000..b981869680 --- /dev/null +++ b/samples/timm/convnextv2_huge.fcmae/model.py @@ -0,0 +1,3267 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (352,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 352, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (352,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_g = x_8.norm(p=2, dim=(1, 2), keepdim=True) + mean = x_g.mean(dim=-1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_8 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_9 = x_8 + addcmul + x_8 = addcmul = None + x_10 = torch._C._nn.linear( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = x_11.permute(0, 3, 1, 2) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 352, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (352,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_g_1 = x_19.norm(p=2, dim=(1, 2), keepdim=True) + mean_1 = x_g_1.mean(dim=-1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_19 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_20 = x_19 + addcmul_1 + x_19 = addcmul_1 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 352, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (352,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_g_2 = x_30.norm(p=2, dim=(1, 2), keepdim=True) + mean_2 = x_g_2.mean(dim=-1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_30 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_31 = x_30 + addcmul_2 + x_30 = addcmul_2 = None + x_32 = torch._C._nn.linear( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + x_34 = x_33.permute(0, 3, 1, 2) + x_33 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (352,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 704, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (704,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_g_3 = x_44.norm(p=2, dim=(1, 2), keepdim=True) + mean_3 = x_g_3.mean(dim=-1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_44 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_45 = x_44 + addcmul_3 + x_44 = addcmul_3 = None + x_46 = torch._C._nn.linear( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 704, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (704,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_g_4 = x_55.norm(p=2, dim=(1, 2), keepdim=True) + mean_4 = x_g_4.mean(dim=-1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_55 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_56 = x_55 + addcmul_4 + x_55 = addcmul_4 = None + x_57 = torch._C._nn.linear( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_58 = torch.nn.functional.dropout(x_57, 0.0, False, False) + x_57 = None + x_59 = x_58.permute(0, 3, 1, 2) + x_58 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 704, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (704,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_g_5 = x_66.norm(p=2, dim=(1, 2), keepdim=True) + mean_5 = x_g_5.mean(dim=-1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_66 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_67 = x_66 + addcmul_5 + x_66 = addcmul_5 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = x_69.permute(0, 3, 1, 2) + x_69 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (704,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_g_6 = x_80.norm(p=2, dim=(1, 2), keepdim=True) + mean_6 = x_g_6.mean(dim=-1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_80 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_81 = x_80 + addcmul_6 + x_80 = addcmul_6 = None + x_82 = torch._C._nn.linear( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + x_84 = x_83.permute(0, 3, 1, 2) + x_83 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_g_7 = x_91.norm(p=2, dim=(1, 2), keepdim=True) + mean_7 = x_g_7.mean(dim=-1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_91 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_92 = x_91 + addcmul_7 + x_91 = addcmul_7 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_g_8 = x_102.norm(p=2, dim=(1, 2), keepdim=True) + mean_8 = x_g_8.mean(dim=-1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_102 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_103 = x_102 + addcmul_8 + x_102 = addcmul_8 = None + x_104 = torch._C._nn.linear( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = x_105.permute(0, 3, 1, 2) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_g_9 = x_113.norm(p=2, dim=(1, 2), keepdim=True) + mean_9 = x_g_9.mean(dim=-1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_113 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_114 = x_113 + addcmul_9 + x_113 = addcmul_9 = None + x_115 = torch._C._nn.linear( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + x_117 = x_116.permute(0, 3, 1, 2) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_g_10 = x_124.norm(p=2, dim=(1, 2), keepdim=True) + mean_10 = x_g_10.mean(dim=-1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_124 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_125 = x_124 + addcmul_10 + x_124 = addcmul_10 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = x_127.permute(0, 3, 1, 2) + x_127 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_g_11 = x_135.norm(p=2, dim=(1, 2), keepdim=True) + mean_11 = x_g_11.mean(dim=-1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_135 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_136 = x_135 + addcmul_11 + x_135 = addcmul_11 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_138.permute(0, 3, 1, 2) + x_138 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_g_12 = x_146.norm(p=2, dim=(1, 2), keepdim=True) + mean_12 = x_g_12.mean(dim=-1, keepdim=True) + add_36 = mean_12 + 1e-06 + mean_12 = None + x_n_12 = x_g_12 / add_36 + x_g_12 = add_36 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_12 = x_146 * x_n_12 + x_n_12 = None + addcmul_12 = torch.addcmul(view_24, view_25, mul_12) + view_24 = view_25 = mul_12 = None + x_147 = x_146 + addcmul_12 + x_146 = addcmul_12 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_149.permute(0, 3, 1, 2) + x_149 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_g_13 = x_157.norm(p=2, dim=(1, 2), keepdim=True) + mean_13 = x_g_13.mean(dim=-1, keepdim=True) + add_39 = mean_13 + 1e-06 + mean_13 = None + x_n_13 = x_g_13 / add_39 + x_g_13 = add_39 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_13 = x_157 * x_n_13 + x_n_13 = None + addcmul_13 = torch.addcmul(view_26, view_27, mul_13) + view_26 = view_27 = mul_13 = None + x_158 = x_157 + addcmul_13 + x_157 = addcmul_13 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_160.permute(0, 3, 1, 2) + x_160 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_g_14 = x_168.norm(p=2, dim=(1, 2), keepdim=True) + mean_14 = x_g_14.mean(dim=-1, keepdim=True) + add_42 = mean_14 + 1e-06 + mean_14 = None + x_n_14 = x_g_14 / add_42 + x_g_14 = add_42 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_14 = x_168 * x_n_14 + x_n_14 = None + addcmul_14 = torch.addcmul(view_28, view_29, mul_14) + view_28 = view_29 = mul_14 = None + x_169 = x_168 + addcmul_14 + x_168 = addcmul_14 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_171.permute(0, 3, 1, 2) + x_171 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = (None) + x_175 = x_174.permute(0, 2, 3, 1) + x_174 = None + x_176 = torch.nn.functional.layer_norm( + x_175, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_, + 1e-06, + ) + x_175 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = (None) + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_176 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_178 = torch._C._nn.gelu(x_177) + x_177 = None + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_g_15 = x_179.norm(p=2, dim=(1, 2), keepdim=True) + mean_15 = x_g_15.mean(dim=-1, keepdim=True) + add_45 = mean_15 + 1e-06 + mean_15 = None + x_n_15 = x_g_15 / add_45 + x_g_15 = add_45 = None + view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_15 = x_179 * x_n_15 + x_n_15 = None + addcmul_15 = torch.addcmul(view_30, view_31, mul_15) + view_30 = view_31 = mul_15 = None + x_180 = x_179 + addcmul_15 + x_179 = addcmul_15 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_182.permute(0, 3, 1, 2) + x_182 = None + x_184 = x_183 + x_173 + x_183 = x_173 = None + x_185 = torch.conv2d( + x_184, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = (None) + x_186 = x_185.permute(0, 2, 3, 1) + x_185 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = (None) + x_188 = torch._C._nn.linear( + x_187, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_187 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_189 = torch._C._nn.gelu(x_188) + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_g_16 = x_190.norm(p=2, dim=(1, 2), keepdim=True) + mean_16 = x_g_16.mean(dim=-1, keepdim=True) + add_48 = mean_16 + 1e-06 + mean_16 = None + x_n_16 = x_g_16 / add_48 + x_g_16 = add_48 = None + view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_16 = x_190 * x_n_16 + x_n_16 = None + addcmul_16 = torch.addcmul(view_32, view_33, mul_16) + view_32 = view_33 = mul_16 = None + x_191 = x_190 + addcmul_16 + x_190 = addcmul_16 = None + x_192 = torch._C._nn.linear( + x_191, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_191 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_193.permute(0, 3, 1, 2) + x_193 = None + x_195 = x_194 + x_184 + x_194 = x_184 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = (None) + x_197 = x_196.permute(0, 2, 3, 1) + x_196 = None + x_198 = torch.nn.functional.layer_norm( + x_197, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_, + 1e-06, + ) + x_197 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = (None) + x_199 = torch._C._nn.linear( + x_198, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_198 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_200 = torch._C._nn.gelu(x_199) + x_199 = None + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_g_17 = x_201.norm(p=2, dim=(1, 2), keepdim=True) + mean_17 = x_g_17.mean(dim=-1, keepdim=True) + add_51 = mean_17 + 1e-06 + mean_17 = None + x_n_17 = x_g_17 / add_51 + x_g_17 = add_51 = None + view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_17 = x_201 * x_n_17 + x_n_17 = None + addcmul_17 = torch.addcmul(view_34, view_35, mul_17) + view_34 = view_35 = mul_17 = None + x_202 = x_201 + addcmul_17 + x_201 = addcmul_17 = None + x_203 = torch._C._nn.linear( + x_202, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_202 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_205 = x_204.permute(0, 3, 1, 2) + x_204 = None + x_206 = x_205 + x_195 + x_205 = x_195 = None + x_207 = torch.conv2d( + x_206, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = (None) + x_208 = x_207.permute(0, 2, 3, 1) + x_207 = None + x_209 = torch.nn.functional.layer_norm( + x_208, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_, + 1e-06, + ) + x_208 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = (None) + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_209 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_211 = torch._C._nn.gelu(x_210) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_g_18 = x_212.norm(p=2, dim=(1, 2), keepdim=True) + mean_18 = x_g_18.mean(dim=-1, keepdim=True) + add_54 = mean_18 + 1e-06 + mean_18 = None + x_n_18 = x_g_18 / add_54 + x_g_18 = add_54 = None + view_36 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_37 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_18 = x_212 * x_n_18 + x_n_18 = None + addcmul_18 = torch.addcmul(view_36, view_37, mul_18) + view_36 = view_37 = mul_18 = None + x_213 = x_212 + addcmul_18 + x_212 = addcmul_18 = None + x_214 = torch._C._nn.linear( + x_213, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_213 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = x_215.permute(0, 3, 1, 2) + x_215 = None + x_217 = x_216 + x_206 + x_216 = x_206 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = (None) + x_219 = x_218.permute(0, 2, 3, 1) + x_218 = None + x_220 = torch.nn.functional.layer_norm( + x_219, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_, + 1e-06, + ) + x_219 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = (None) + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_220 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_222 = torch._C._nn.gelu(x_221) + x_221 = None + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_g_19 = x_223.norm(p=2, dim=(1, 2), keepdim=True) + mean_19 = x_g_19.mean(dim=-1, keepdim=True) + add_57 = mean_19 + 1e-06 + mean_19 = None + x_n_19 = x_g_19 / add_57 + x_g_19 = add_57 = None + view_38 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_39 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_19 = x_223 * x_n_19 + x_n_19 = None + addcmul_19 = torch.addcmul(view_38, view_39, mul_19) + view_38 = view_39 = mul_19 = None + x_224 = x_223 + addcmul_19 + x_223 = addcmul_19 = None + x_225 = torch._C._nn.linear( + x_224, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_224 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_226 = torch.nn.functional.dropout(x_225, 0.0, False, False) + x_225 = None + x_227 = x_226.permute(0, 3, 1, 2) + x_226 = None + x_228 = x_227 + x_217 + x_227 = x_217 = None + x_229 = torch.conv2d( + x_228, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = (None) + x_230 = x_229.permute(0, 2, 3, 1) + x_229 = None + x_231 = torch.nn.functional.layer_norm( + x_230, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_, + 1e-06, + ) + x_230 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = (None) + x_232 = torch._C._nn.linear( + x_231, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_231 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_233 = torch._C._nn.gelu(x_232) + x_232 = None + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_g_20 = x_234.norm(p=2, dim=(1, 2), keepdim=True) + mean_20 = x_g_20.mean(dim=-1, keepdim=True) + add_60 = mean_20 + 1e-06 + mean_20 = None + x_n_20 = x_g_20 / add_60 + x_g_20 = add_60 = None + view_40 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_41 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_20 = x_234 * x_n_20 + x_n_20 = None + addcmul_20 = torch.addcmul(view_40, view_41, mul_20) + view_40 = view_41 = mul_20 = None + x_235 = x_234 + addcmul_20 + x_234 = addcmul_20 = None + x_236 = torch._C._nn.linear( + x_235, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_235 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_237 = torch.nn.functional.dropout(x_236, 0.0, False, False) + x_236 = None + x_238 = x_237.permute(0, 3, 1, 2) + x_237 = None + x_239 = x_238 + x_228 + x_238 = x_228 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = (None) + x_241 = x_240.permute(0, 2, 3, 1) + x_240 = None + x_242 = torch.nn.functional.layer_norm( + x_241, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_, + 1e-06, + ) + x_241 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = (None) + x_243 = torch._C._nn.linear( + x_242, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_242 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_244 = torch._C._nn.gelu(x_243) + x_243 = None + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_g_21 = x_245.norm(p=2, dim=(1, 2), keepdim=True) + mean_21 = x_g_21.mean(dim=-1, keepdim=True) + add_63 = mean_21 + 1e-06 + mean_21 = None + x_n_21 = x_g_21 / add_63 + x_g_21 = add_63 = None + view_42 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_43 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_21 = x_245 * x_n_21 + x_n_21 = None + addcmul_21 = torch.addcmul(view_42, view_43, mul_21) + view_42 = view_43 = mul_21 = None + x_246 = x_245 + addcmul_21 + x_245 = addcmul_21 = None + x_247 = torch._C._nn.linear( + x_246, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_246 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_248 = torch.nn.functional.dropout(x_247, 0.0, False, False) + x_247 = None + x_249 = x_248.permute(0, 3, 1, 2) + x_248 = None + x_250 = x_249 + x_239 + x_249 = x_239 = None + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = (None) + x_252 = x_251.permute(0, 2, 3, 1) + x_251 = None + x_253 = torch.nn.functional.layer_norm( + x_252, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_, + 1e-06, + ) + x_252 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = (None) + x_254 = torch._C._nn.linear( + x_253, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_253 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_255 = torch._C._nn.gelu(x_254) + x_254 = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_g_22 = x_256.norm(p=2, dim=(1, 2), keepdim=True) + mean_22 = x_g_22.mean(dim=-1, keepdim=True) + add_66 = mean_22 + 1e-06 + mean_22 = None + x_n_22 = x_g_22 / add_66 + x_g_22 = add_66 = None + view_44 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_45 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_22 = x_256 * x_n_22 + x_n_22 = None + addcmul_22 = torch.addcmul(view_44, view_45, mul_22) + view_44 = view_45 = mul_22 = None + x_257 = x_256 + addcmul_22 + x_256 = addcmul_22 = None + x_258 = torch._C._nn.linear( + x_257, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_257 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = x_259.permute(0, 3, 1, 2) + x_259 = None + x_261 = x_260 + x_250 + x_260 = x_250 = None + x_262 = torch.conv2d( + x_261, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = (None) + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = (None) + x_265 = torch._C._nn.linear( + x_264, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_264 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_266 = torch._C._nn.gelu(x_265) + x_265 = None + x_267 = torch.nn.functional.dropout(x_266, 0.0, False, False) + x_266 = None + x_g_23 = x_267.norm(p=2, dim=(1, 2), keepdim=True) + mean_23 = x_g_23.mean(dim=-1, keepdim=True) + add_69 = mean_23 + 1e-06 + mean_23 = None + x_n_23 = x_g_23 / add_69 + x_g_23 = add_69 = None + view_46 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_47 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_23 = x_267 * x_n_23 + x_n_23 = None + addcmul_23 = torch.addcmul(view_46, view_47, mul_23) + view_46 = view_47 = mul_23 = None + x_268 = x_267 + addcmul_23 + x_267 = addcmul_23 = None + x_269 = torch._C._nn.linear( + x_268, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_268 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_270 = torch.nn.functional.dropout(x_269, 0.0, False, False) + x_269 = None + x_271 = x_270.permute(0, 3, 1, 2) + x_270 = None + x_272 = x_271 + x_261 + x_271 = x_261 = None + x_273 = torch.conv2d( + x_272, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = (None) + x_274 = x_273.permute(0, 2, 3, 1) + x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_, + 1e-06, + ) + x_274 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = (None) + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_277 = torch._C._nn.gelu(x_276) + x_276 = None + x_278 = torch.nn.functional.dropout(x_277, 0.0, False, False) + x_277 = None + x_g_24 = x_278.norm(p=2, dim=(1, 2), keepdim=True) + mean_24 = x_g_24.mean(dim=-1, keepdim=True) + add_72 = mean_24 + 1e-06 + mean_24 = None + x_n_24 = x_g_24 / add_72 + x_g_24 = add_72 = None + view_48 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_49 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_24 = x_278 * x_n_24 + x_n_24 = None + addcmul_24 = torch.addcmul(view_48, view_49, mul_24) + view_48 = view_49 = mul_24 = None + x_279 = x_278 + addcmul_24 + x_278 = addcmul_24 = None + x_280 = torch._C._nn.linear( + x_279, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_279 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_281 = torch.nn.functional.dropout(x_280, 0.0, False, False) + x_280 = None + x_282 = x_281.permute(0, 3, 1, 2) + x_281 = None + x_283 = x_282 + x_272 + x_282 = x_272 = None + x_284 = torch.conv2d( + x_283, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = (None) + x_285 = x_284.permute(0, 2, 3, 1) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_, + 1e-06, + ) + x_285 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = (None) + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_288 = torch._C._nn.gelu(x_287) + x_287 = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + x_g_25 = x_289.norm(p=2, dim=(1, 2), keepdim=True) + mean_25 = x_g_25.mean(dim=-1, keepdim=True) + add_75 = mean_25 + 1e-06 + mean_25 = None + x_n_25 = x_g_25 / add_75 + x_g_25 = add_75 = None + view_50 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_51 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_25 = x_289 * x_n_25 + x_n_25 = None + addcmul_25 = torch.addcmul(view_50, view_51, mul_25) + view_50 = view_51 = mul_25 = None + x_290 = x_289 + addcmul_25 + x_289 = addcmul_25 = None + x_291 = torch._C._nn.linear( + x_290, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_290 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_292 = torch.nn.functional.dropout(x_291, 0.0, False, False) + x_291 = None + x_293 = x_292.permute(0, 3, 1, 2) + x_292 = None + x_294 = x_293 + x_283 + x_293 = x_283 = None + x_295 = torch.conv2d( + x_294, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = (None) + x_296 = x_295.permute(0, 2, 3, 1) + x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_, + 1e-06, + ) + x_296 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = (None) + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_299 = torch._C._nn.gelu(x_298) + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_g_26 = x_300.norm(p=2, dim=(1, 2), keepdim=True) + mean_26 = x_g_26.mean(dim=-1, keepdim=True) + add_78 = mean_26 + 1e-06 + mean_26 = None + x_n_26 = x_g_26 / add_78 + x_g_26 = add_78 = None + view_52 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_53 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_26 = x_300 * x_n_26 + x_n_26 = None + addcmul_26 = torch.addcmul(view_52, view_53, mul_26) + view_52 = view_53 = mul_26 = None + x_301 = x_300 + addcmul_26 + x_300 = addcmul_26 = None + x_302 = torch._C._nn.linear( + x_301, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_301 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_303 = torch.nn.functional.dropout(x_302, 0.0, False, False) + x_302 = None + x_304 = x_303.permute(0, 3, 1, 2) + x_303 = None + x_305 = x_304 + x_294 + x_304 = x_294 = None + x_306 = torch.conv2d( + x_305, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = (None) + x_307 = x_306.permute(0, 2, 3, 1) + x_306 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_, + 1e-06, + ) + x_307 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = (None) + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_310 = torch._C._nn.gelu(x_309) + x_309 = None + x_311 = torch.nn.functional.dropout(x_310, 0.0, False, False) + x_310 = None + x_g_27 = x_311.norm(p=2, dim=(1, 2), keepdim=True) + mean_27 = x_g_27.mean(dim=-1, keepdim=True) + add_81 = mean_27 + 1e-06 + mean_27 = None + x_n_27 = x_g_27 / add_81 + x_g_27 = add_81 = None + view_54 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_55 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_27 = x_311 * x_n_27 + x_n_27 = None + addcmul_27 = torch.addcmul(view_54, view_55, mul_27) + view_54 = view_55 = mul_27 = None + x_312 = x_311 + addcmul_27 + x_311 = addcmul_27 = None + x_313 = torch._C._nn.linear( + x_312, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_312 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_314 = torch.nn.functional.dropout(x_313, 0.0, False, False) + x_313 = None + x_315 = x_314.permute(0, 3, 1, 2) + x_314 = None + x_316 = x_315 + x_305 + x_315 = x_305 = None + x_317 = torch.conv2d( + x_316, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = (None) + x_318 = x_317.permute(0, 2, 3, 1) + x_317 = None + x_319 = torch.nn.functional.layer_norm( + x_318, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_, + 1e-06, + ) + x_318 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = (None) + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_321 = torch._C._nn.gelu(x_320) + x_320 = None + x_322 = torch.nn.functional.dropout(x_321, 0.0, False, False) + x_321 = None + x_g_28 = x_322.norm(p=2, dim=(1, 2), keepdim=True) + mean_28 = x_g_28.mean(dim=-1, keepdim=True) + add_84 = mean_28 + 1e-06 + mean_28 = None + x_n_28 = x_g_28 / add_84 + x_g_28 = add_84 = None + view_56 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_57 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_28 = x_322 * x_n_28 + x_n_28 = None + addcmul_28 = torch.addcmul(view_56, view_57, mul_28) + view_56 = view_57 = mul_28 = None + x_323 = x_322 + addcmul_28 + x_322 = addcmul_28 = None + x_324 = torch._C._nn.linear( + x_323, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_323 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_325 = torch.nn.functional.dropout(x_324, 0.0, False, False) + x_324 = None + x_326 = x_325.permute(0, 3, 1, 2) + x_325 = None + x_327 = x_326 + x_316 + x_326 = x_316 = None + x_328 = torch.conv2d( + x_327, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = (None) + x_329 = x_328.permute(0, 2, 3, 1) + x_328 = None + x_330 = torch.nn.functional.layer_norm( + x_329, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_, + 1e-06, + ) + x_329 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = (None) + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_332 = torch._C._nn.gelu(x_331) + x_331 = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_g_29 = x_333.norm(p=2, dim=(1, 2), keepdim=True) + mean_29 = x_g_29.mean(dim=-1, keepdim=True) + add_87 = mean_29 + 1e-06 + mean_29 = None + x_n_29 = x_g_29 / add_87 + x_g_29 = add_87 = None + view_58 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_59 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_29 = x_333 * x_n_29 + x_n_29 = None + addcmul_29 = torch.addcmul(view_58, view_59, mul_29) + view_58 = view_59 = mul_29 = None + x_334 = x_333 + addcmul_29 + x_333 = addcmul_29 = None + x_335 = torch._C._nn.linear( + x_334, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_334 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_336 = torch.nn.functional.dropout(x_335, 0.0, False, False) + x_335 = None + x_337 = x_336.permute(0, 3, 1, 2) + x_336 = None + x_338 = x_337 + x_327 + x_337 = x_327 = None + x_339 = torch.conv2d( + x_338, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = (None) + x_340 = x_339.permute(0, 2, 3, 1) + x_339 = None + x_341 = torch.nn.functional.layer_norm( + x_340, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_, + 1e-06, + ) + x_340 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = (None) + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_343 = torch._C._nn.gelu(x_342) + x_342 = None + x_344 = torch.nn.functional.dropout(x_343, 0.0, False, False) + x_343 = None + x_g_30 = x_344.norm(p=2, dim=(1, 2), keepdim=True) + mean_30 = x_g_30.mean(dim=-1, keepdim=True) + add_90 = mean_30 + 1e-06 + mean_30 = None + x_n_30 = x_g_30 / add_90 + x_g_30 = add_90 = None + view_60 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_61 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_30 = x_344 * x_n_30 + x_n_30 = None + addcmul_30 = torch.addcmul(view_60, view_61, mul_30) + view_60 = view_61 = mul_30 = None + x_345 = x_344 + addcmul_30 + x_344 = addcmul_30 = None + x_346 = torch._C._nn.linear( + x_345, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_345 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_347 = torch.nn.functional.dropout(x_346, 0.0, False, False) + x_346 = None + x_348 = x_347.permute(0, 3, 1, 2) + x_347 = None + x_349 = x_348 + x_338 + x_348 = x_338 = None + x_350 = torch.conv2d( + x_349, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = (None) + x_351 = x_350.permute(0, 2, 3, 1) + x_350 = None + x_352 = torch.nn.functional.layer_norm( + x_351, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_, + 1e-06, + ) + x_351 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = (None) + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_354 = torch._C._nn.gelu(x_353) + x_353 = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_g_31 = x_355.norm(p=2, dim=(1, 2), keepdim=True) + mean_31 = x_g_31.mean(dim=-1, keepdim=True) + add_93 = mean_31 + 1e-06 + mean_31 = None + x_n_31 = x_g_31 / add_93 + x_g_31 = add_93 = None + view_62 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_63 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_31 = x_355 * x_n_31 + x_n_31 = None + addcmul_31 = torch.addcmul(view_62, view_63, mul_31) + view_62 = view_63 = mul_31 = None + x_356 = x_355 + addcmul_31 + x_355 = addcmul_31 = None + x_357 = torch._C._nn.linear( + x_356, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_356 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_358 = torch.nn.functional.dropout(x_357, 0.0, False, False) + x_357 = None + x_359 = x_358.permute(0, 3, 1, 2) + x_358 = None + x_360 = x_359 + x_349 + x_359 = x_349 = None + x_361 = torch.conv2d( + x_360, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = (None) + x_362 = x_361.permute(0, 2, 3, 1) + x_361 = None + x_363 = torch.nn.functional.layer_norm( + x_362, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_, + 1e-06, + ) + x_362 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = (None) + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_365 = torch._C._nn.gelu(x_364) + x_364 = None + x_366 = torch.nn.functional.dropout(x_365, 0.0, False, False) + x_365 = None + x_g_32 = x_366.norm(p=2, dim=(1, 2), keepdim=True) + mean_32 = x_g_32.mean(dim=-1, keepdim=True) + add_96 = mean_32 + 1e-06 + mean_32 = None + x_n_32 = x_g_32 / add_96 + x_g_32 = add_96 = None + view_64 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_65 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_32 = x_366 * x_n_32 + x_n_32 = None + addcmul_32 = torch.addcmul(view_64, view_65, mul_32) + view_64 = view_65 = mul_32 = None + x_367 = x_366 + addcmul_32 + x_366 = addcmul_32 = None + x_368 = torch._C._nn.linear( + x_367, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_367 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_369 = torch.nn.functional.dropout(x_368, 0.0, False, False) + x_368 = None + x_370 = x_369.permute(0, 3, 1, 2) + x_369 = None + x_371 = x_370 + x_360 + x_370 = x_360 = None + x_372 = x_371.permute(0, 2, 3, 1) + x_371 = None + x_373 = torch.nn.functional.layer_norm( + x_372, + (1408,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_372 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_374 = x_373.permute(0, 3, 1, 2) + x_373 = None + input_4 = torch.conv2d( + x_374, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_374 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_375 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 2816, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_376 = x_375.permute(0, 2, 3, 1) + x_375 = None + x_377 = torch.nn.functional.layer_norm( + x_376, + (2816,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_376 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_378 = torch._C._nn.linear( + x_377, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_377 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_379 = torch._C._nn.gelu(x_378) + x_378 = None + x_380 = torch.nn.functional.dropout(x_379, 0.0, False, False) + x_379 = None + x_g_33 = x_380.norm(p=2, dim=(1, 2), keepdim=True) + mean_33 = x_g_33.mean(dim=-1, keepdim=True) + add_99 = mean_33 + 1e-06 + mean_33 = None + x_n_33 = x_g_33 / add_99 + x_g_33 = add_99 = None + view_66 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_67 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_33 = x_380 * x_n_33 + x_n_33 = None + addcmul_33 = torch.addcmul(view_66, view_67, mul_33) + view_66 = view_67 = mul_33 = None + x_381 = x_380 + addcmul_33 + x_380 = addcmul_33 = None + x_382 = torch._C._nn.linear( + x_381, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_381 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_383 = torch.nn.functional.dropout(x_382, 0.0, False, False) + x_382 = None + x_384 = x_383.permute(0, 3, 1, 2) + x_383 = None + x_385 = x_384 + input_4 + x_384 = input_4 = None + x_386 = torch.conv2d( + x_385, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 2816, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_387 = x_386.permute(0, 2, 3, 1) + x_386 = None + x_388 = torch.nn.functional.layer_norm( + x_387, + (2816,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_387 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_389 = torch._C._nn.linear( + x_388, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_388 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_390 = torch._C._nn.gelu(x_389) + x_389 = None + x_391 = torch.nn.functional.dropout(x_390, 0.0, False, False) + x_390 = None + x_g_34 = x_391.norm(p=2, dim=(1, 2), keepdim=True) + mean_34 = x_g_34.mean(dim=-1, keepdim=True) + add_102 = mean_34 + 1e-06 + mean_34 = None + x_n_34 = x_g_34 / add_102 + x_g_34 = add_102 = None + view_68 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_69 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_34 = x_391 * x_n_34 + x_n_34 = None + addcmul_34 = torch.addcmul(view_68, view_69, mul_34) + view_68 = view_69 = mul_34 = None + x_392 = x_391 + addcmul_34 + x_391 = addcmul_34 = None + x_393 = torch._C._nn.linear( + x_392, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_392 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_394 = torch.nn.functional.dropout(x_393, 0.0, False, False) + x_393 = None + x_395 = x_394.permute(0, 3, 1, 2) + x_394 = None + x_396 = x_395 + x_385 + x_395 = x_385 = None + x_397 = torch.conv2d( + x_396, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 2816, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_398 = x_397.permute(0, 2, 3, 1) + x_397 = None + x_399 = torch.nn.functional.layer_norm( + x_398, + (2816,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_398 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_400 = torch._C._nn.linear( + x_399, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_399 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_401 = torch._C._nn.gelu(x_400) + x_400 = None + x_402 = torch.nn.functional.dropout(x_401, 0.0, False, False) + x_401 = None + x_g_35 = x_402.norm(p=2, dim=(1, 2), keepdim=True) + mean_35 = x_g_35.mean(dim=-1, keepdim=True) + add_105 = mean_35 + 1e-06 + mean_35 = None + x_n_35 = x_g_35 / add_105 + x_g_35 = add_105 = None + view_70 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_71 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_35 = x_402 * x_n_35 + x_n_35 = None + addcmul_35 = torch.addcmul(view_70, view_71, mul_35) + view_70 = view_71 = mul_35 = None + x_403 = x_402 + addcmul_35 + x_402 = addcmul_35 = None + x_404 = torch._C._nn.linear( + x_403, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_403 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_405 = torch.nn.functional.dropout(x_404, 0.0, False, False) + x_404 = None + x_406 = x_405.permute(0, 3, 1, 2) + x_405 = None + x_407 = x_406 + x_396 + x_406 = x_396 = None + x_408 = torch.nn.functional.adaptive_avg_pool2d(x_407, 1) + x_407 = None + x_409 = x_408.permute(0, 2, 3, 1) + x_408 = None + x_410 = torch.nn.functional.layer_norm( + x_409, + (2816,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_409 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_411 = x_410.permute(0, 3, 1, 2) + x_410 = None + x_412 = x_411.flatten(1, -1) + x_411 = None + x_413 = torch.nn.functional.dropout(x_412, 0.0, False, False) + x_412 = None + return (x_413,) diff --git a/samples/timm/convnextv2_huge.fcmae/weight_meta.py b/samples/timm/convnextv2_huge.fcmae/weight_meta.py new file mode 100644 index 0000000000..ae4f9db92d --- /dev/null +++ b/samples/timm/convnextv2_huge.fcmae/weight_meta.py @@ -0,0 +1,3810 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [352, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.225 + std = 1.284 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [352, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1408, 352] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [352, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [352, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1408, 352] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [352, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [352, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1408, 352] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [352, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [704, 352, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [704, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2816, 704] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [704, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [704, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2816, 704] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [704, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [704, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2816, 704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [704, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [1408, 704, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [2816, 1408, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [2816, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [11264, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [2816, 11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [2816, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [11264, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [2816, 11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [2816, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [11264, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [2816, 11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_huge.fcmae_ft_in1k/graph_hash.txt b/samples/timm/convnextv2_huge.fcmae_ft_in1k/graph_hash.txt new file mode 100644 index 0000000000..f32efd31ed --- /dev/null +++ b/samples/timm/convnextv2_huge.fcmae_ft_in1k/graph_hash.txt @@ -0,0 +1 @@ +f89d8e3e398c7a58eb6fce1d1432881adf546a3af3016fdb63062b361da41a1e \ No newline at end of file diff --git a/samples/timm/convnextv2_huge.fcmae_ft_in1k/graph_net.json b/samples/timm/convnextv2_huge.fcmae_ft_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnextv2_huge.fcmae_ft_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_huge.fcmae_ft_in1k/input_meta.py b/samples/timm/convnextv2_huge.fcmae_ft_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_huge.fcmae_ft_in1k/input_tensor_constraints.py b/samples/timm/convnextv2_huge.fcmae_ft_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_huge.fcmae_ft_in1k/model.py b/samples/timm/convnextv2_huge.fcmae_ft_in1k/model.py new file mode 100644 index 0000000000..f795480352 --- /dev/null +++ b/samples/timm/convnextv2_huge.fcmae_ft_in1k/model.py @@ -0,0 +1,3283 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (352,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 352, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (352,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_g = x_8.norm(p=2, dim=(1, 2), keepdim=True) + mean = x_g.mean(dim=-1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_8 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_9 = x_8 + addcmul + x_8 = addcmul = None + x_10 = torch._C._nn.linear( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = x_11.permute(0, 3, 1, 2) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 352, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (352,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_g_1 = x_19.norm(p=2, dim=(1, 2), keepdim=True) + mean_1 = x_g_1.mean(dim=-1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_19 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_20 = x_19 + addcmul_1 + x_19 = addcmul_1 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 352, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (352,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_g_2 = x_30.norm(p=2, dim=(1, 2), keepdim=True) + mean_2 = x_g_2.mean(dim=-1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_30 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_31 = x_30 + addcmul_2 + x_30 = addcmul_2 = None + x_32 = torch._C._nn.linear( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + x_34 = x_33.permute(0, 3, 1, 2) + x_33 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (352,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 704, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (704,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_g_3 = x_44.norm(p=2, dim=(1, 2), keepdim=True) + mean_3 = x_g_3.mean(dim=-1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_44 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_45 = x_44 + addcmul_3 + x_44 = addcmul_3 = None + x_46 = torch._C._nn.linear( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 704, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (704,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_g_4 = x_55.norm(p=2, dim=(1, 2), keepdim=True) + mean_4 = x_g_4.mean(dim=-1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_55 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_56 = x_55 + addcmul_4 + x_55 = addcmul_4 = None + x_57 = torch._C._nn.linear( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_58 = torch.nn.functional.dropout(x_57, 0.0, False, False) + x_57 = None + x_59 = x_58.permute(0, 3, 1, 2) + x_58 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 704, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (704,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_g_5 = x_66.norm(p=2, dim=(1, 2), keepdim=True) + mean_5 = x_g_5.mean(dim=-1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_66 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_67 = x_66 + addcmul_5 + x_66 = addcmul_5 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = x_69.permute(0, 3, 1, 2) + x_69 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (704,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_g_6 = x_80.norm(p=2, dim=(1, 2), keepdim=True) + mean_6 = x_g_6.mean(dim=-1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_80 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_81 = x_80 + addcmul_6 + x_80 = addcmul_6 = None + x_82 = torch._C._nn.linear( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + x_84 = x_83.permute(0, 3, 1, 2) + x_83 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_g_7 = x_91.norm(p=2, dim=(1, 2), keepdim=True) + mean_7 = x_g_7.mean(dim=-1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_91 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_92 = x_91 + addcmul_7 + x_91 = addcmul_7 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_g_8 = x_102.norm(p=2, dim=(1, 2), keepdim=True) + mean_8 = x_g_8.mean(dim=-1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_102 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_103 = x_102 + addcmul_8 + x_102 = addcmul_8 = None + x_104 = torch._C._nn.linear( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = x_105.permute(0, 3, 1, 2) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_g_9 = x_113.norm(p=2, dim=(1, 2), keepdim=True) + mean_9 = x_g_9.mean(dim=-1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_113 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_114 = x_113 + addcmul_9 + x_113 = addcmul_9 = None + x_115 = torch._C._nn.linear( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + x_117 = x_116.permute(0, 3, 1, 2) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_g_10 = x_124.norm(p=2, dim=(1, 2), keepdim=True) + mean_10 = x_g_10.mean(dim=-1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_124 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_125 = x_124 + addcmul_10 + x_124 = addcmul_10 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = x_127.permute(0, 3, 1, 2) + x_127 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_g_11 = x_135.norm(p=2, dim=(1, 2), keepdim=True) + mean_11 = x_g_11.mean(dim=-1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_135 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_136 = x_135 + addcmul_11 + x_135 = addcmul_11 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_138.permute(0, 3, 1, 2) + x_138 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_g_12 = x_146.norm(p=2, dim=(1, 2), keepdim=True) + mean_12 = x_g_12.mean(dim=-1, keepdim=True) + add_36 = mean_12 + 1e-06 + mean_12 = None + x_n_12 = x_g_12 / add_36 + x_g_12 = add_36 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_12 = x_146 * x_n_12 + x_n_12 = None + addcmul_12 = torch.addcmul(view_24, view_25, mul_12) + view_24 = view_25 = mul_12 = None + x_147 = x_146 + addcmul_12 + x_146 = addcmul_12 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_149.permute(0, 3, 1, 2) + x_149 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_g_13 = x_157.norm(p=2, dim=(1, 2), keepdim=True) + mean_13 = x_g_13.mean(dim=-1, keepdim=True) + add_39 = mean_13 + 1e-06 + mean_13 = None + x_n_13 = x_g_13 / add_39 + x_g_13 = add_39 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_13 = x_157 * x_n_13 + x_n_13 = None + addcmul_13 = torch.addcmul(view_26, view_27, mul_13) + view_26 = view_27 = mul_13 = None + x_158 = x_157 + addcmul_13 + x_157 = addcmul_13 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_160.permute(0, 3, 1, 2) + x_160 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_g_14 = x_168.norm(p=2, dim=(1, 2), keepdim=True) + mean_14 = x_g_14.mean(dim=-1, keepdim=True) + add_42 = mean_14 + 1e-06 + mean_14 = None + x_n_14 = x_g_14 / add_42 + x_g_14 = add_42 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_14 = x_168 * x_n_14 + x_n_14 = None + addcmul_14 = torch.addcmul(view_28, view_29, mul_14) + view_28 = view_29 = mul_14 = None + x_169 = x_168 + addcmul_14 + x_168 = addcmul_14 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_171.permute(0, 3, 1, 2) + x_171 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = (None) + x_175 = x_174.permute(0, 2, 3, 1) + x_174 = None + x_176 = torch.nn.functional.layer_norm( + x_175, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_, + 1e-06, + ) + x_175 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = (None) + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_176 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_178 = torch._C._nn.gelu(x_177) + x_177 = None + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_g_15 = x_179.norm(p=2, dim=(1, 2), keepdim=True) + mean_15 = x_g_15.mean(dim=-1, keepdim=True) + add_45 = mean_15 + 1e-06 + mean_15 = None + x_n_15 = x_g_15 / add_45 + x_g_15 = add_45 = None + view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_15 = x_179 * x_n_15 + x_n_15 = None + addcmul_15 = torch.addcmul(view_30, view_31, mul_15) + view_30 = view_31 = mul_15 = None + x_180 = x_179 + addcmul_15 + x_179 = addcmul_15 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_182.permute(0, 3, 1, 2) + x_182 = None + x_184 = x_183 + x_173 + x_183 = x_173 = None + x_185 = torch.conv2d( + x_184, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = (None) + x_186 = x_185.permute(0, 2, 3, 1) + x_185 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = (None) + x_188 = torch._C._nn.linear( + x_187, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_187 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_189 = torch._C._nn.gelu(x_188) + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_g_16 = x_190.norm(p=2, dim=(1, 2), keepdim=True) + mean_16 = x_g_16.mean(dim=-1, keepdim=True) + add_48 = mean_16 + 1e-06 + mean_16 = None + x_n_16 = x_g_16 / add_48 + x_g_16 = add_48 = None + view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_16 = x_190 * x_n_16 + x_n_16 = None + addcmul_16 = torch.addcmul(view_32, view_33, mul_16) + view_32 = view_33 = mul_16 = None + x_191 = x_190 + addcmul_16 + x_190 = addcmul_16 = None + x_192 = torch._C._nn.linear( + x_191, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_191 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_193.permute(0, 3, 1, 2) + x_193 = None + x_195 = x_194 + x_184 + x_194 = x_184 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = (None) + x_197 = x_196.permute(0, 2, 3, 1) + x_196 = None + x_198 = torch.nn.functional.layer_norm( + x_197, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_, + 1e-06, + ) + x_197 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = (None) + x_199 = torch._C._nn.linear( + x_198, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_198 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_200 = torch._C._nn.gelu(x_199) + x_199 = None + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_g_17 = x_201.norm(p=2, dim=(1, 2), keepdim=True) + mean_17 = x_g_17.mean(dim=-1, keepdim=True) + add_51 = mean_17 + 1e-06 + mean_17 = None + x_n_17 = x_g_17 / add_51 + x_g_17 = add_51 = None + view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_17 = x_201 * x_n_17 + x_n_17 = None + addcmul_17 = torch.addcmul(view_34, view_35, mul_17) + view_34 = view_35 = mul_17 = None + x_202 = x_201 + addcmul_17 + x_201 = addcmul_17 = None + x_203 = torch._C._nn.linear( + x_202, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_202 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_205 = x_204.permute(0, 3, 1, 2) + x_204 = None + x_206 = x_205 + x_195 + x_205 = x_195 = None + x_207 = torch.conv2d( + x_206, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = (None) + x_208 = x_207.permute(0, 2, 3, 1) + x_207 = None + x_209 = torch.nn.functional.layer_norm( + x_208, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_, + 1e-06, + ) + x_208 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = (None) + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_209 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_211 = torch._C._nn.gelu(x_210) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_g_18 = x_212.norm(p=2, dim=(1, 2), keepdim=True) + mean_18 = x_g_18.mean(dim=-1, keepdim=True) + add_54 = mean_18 + 1e-06 + mean_18 = None + x_n_18 = x_g_18 / add_54 + x_g_18 = add_54 = None + view_36 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_37 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_18 = x_212 * x_n_18 + x_n_18 = None + addcmul_18 = torch.addcmul(view_36, view_37, mul_18) + view_36 = view_37 = mul_18 = None + x_213 = x_212 + addcmul_18 + x_212 = addcmul_18 = None + x_214 = torch._C._nn.linear( + x_213, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_213 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = x_215.permute(0, 3, 1, 2) + x_215 = None + x_217 = x_216 + x_206 + x_216 = x_206 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = (None) + x_219 = x_218.permute(0, 2, 3, 1) + x_218 = None + x_220 = torch.nn.functional.layer_norm( + x_219, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_, + 1e-06, + ) + x_219 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = (None) + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_220 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_222 = torch._C._nn.gelu(x_221) + x_221 = None + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_g_19 = x_223.norm(p=2, dim=(1, 2), keepdim=True) + mean_19 = x_g_19.mean(dim=-1, keepdim=True) + add_57 = mean_19 + 1e-06 + mean_19 = None + x_n_19 = x_g_19 / add_57 + x_g_19 = add_57 = None + view_38 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_39 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_19 = x_223 * x_n_19 + x_n_19 = None + addcmul_19 = torch.addcmul(view_38, view_39, mul_19) + view_38 = view_39 = mul_19 = None + x_224 = x_223 + addcmul_19 + x_223 = addcmul_19 = None + x_225 = torch._C._nn.linear( + x_224, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_224 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_226 = torch.nn.functional.dropout(x_225, 0.0, False, False) + x_225 = None + x_227 = x_226.permute(0, 3, 1, 2) + x_226 = None + x_228 = x_227 + x_217 + x_227 = x_217 = None + x_229 = torch.conv2d( + x_228, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = (None) + x_230 = x_229.permute(0, 2, 3, 1) + x_229 = None + x_231 = torch.nn.functional.layer_norm( + x_230, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_, + 1e-06, + ) + x_230 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = (None) + x_232 = torch._C._nn.linear( + x_231, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_231 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_233 = torch._C._nn.gelu(x_232) + x_232 = None + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_g_20 = x_234.norm(p=2, dim=(1, 2), keepdim=True) + mean_20 = x_g_20.mean(dim=-1, keepdim=True) + add_60 = mean_20 + 1e-06 + mean_20 = None + x_n_20 = x_g_20 / add_60 + x_g_20 = add_60 = None + view_40 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_41 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_20 = x_234 * x_n_20 + x_n_20 = None + addcmul_20 = torch.addcmul(view_40, view_41, mul_20) + view_40 = view_41 = mul_20 = None + x_235 = x_234 + addcmul_20 + x_234 = addcmul_20 = None + x_236 = torch._C._nn.linear( + x_235, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_235 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_237 = torch.nn.functional.dropout(x_236, 0.0, False, False) + x_236 = None + x_238 = x_237.permute(0, 3, 1, 2) + x_237 = None + x_239 = x_238 + x_228 + x_238 = x_228 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = (None) + x_241 = x_240.permute(0, 2, 3, 1) + x_240 = None + x_242 = torch.nn.functional.layer_norm( + x_241, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_, + 1e-06, + ) + x_241 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = (None) + x_243 = torch._C._nn.linear( + x_242, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_242 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_244 = torch._C._nn.gelu(x_243) + x_243 = None + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_g_21 = x_245.norm(p=2, dim=(1, 2), keepdim=True) + mean_21 = x_g_21.mean(dim=-1, keepdim=True) + add_63 = mean_21 + 1e-06 + mean_21 = None + x_n_21 = x_g_21 / add_63 + x_g_21 = add_63 = None + view_42 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_43 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_21 = x_245 * x_n_21 + x_n_21 = None + addcmul_21 = torch.addcmul(view_42, view_43, mul_21) + view_42 = view_43 = mul_21 = None + x_246 = x_245 + addcmul_21 + x_245 = addcmul_21 = None + x_247 = torch._C._nn.linear( + x_246, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_246 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_248 = torch.nn.functional.dropout(x_247, 0.0, False, False) + x_247 = None + x_249 = x_248.permute(0, 3, 1, 2) + x_248 = None + x_250 = x_249 + x_239 + x_249 = x_239 = None + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = (None) + x_252 = x_251.permute(0, 2, 3, 1) + x_251 = None + x_253 = torch.nn.functional.layer_norm( + x_252, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_, + 1e-06, + ) + x_252 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = (None) + x_254 = torch._C._nn.linear( + x_253, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_253 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_255 = torch._C._nn.gelu(x_254) + x_254 = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_g_22 = x_256.norm(p=2, dim=(1, 2), keepdim=True) + mean_22 = x_g_22.mean(dim=-1, keepdim=True) + add_66 = mean_22 + 1e-06 + mean_22 = None + x_n_22 = x_g_22 / add_66 + x_g_22 = add_66 = None + view_44 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_45 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_22 = x_256 * x_n_22 + x_n_22 = None + addcmul_22 = torch.addcmul(view_44, view_45, mul_22) + view_44 = view_45 = mul_22 = None + x_257 = x_256 + addcmul_22 + x_256 = addcmul_22 = None + x_258 = torch._C._nn.linear( + x_257, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_257 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = x_259.permute(0, 3, 1, 2) + x_259 = None + x_261 = x_260 + x_250 + x_260 = x_250 = None + x_262 = torch.conv2d( + x_261, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = (None) + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = (None) + x_265 = torch._C._nn.linear( + x_264, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_264 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_266 = torch._C._nn.gelu(x_265) + x_265 = None + x_267 = torch.nn.functional.dropout(x_266, 0.0, False, False) + x_266 = None + x_g_23 = x_267.norm(p=2, dim=(1, 2), keepdim=True) + mean_23 = x_g_23.mean(dim=-1, keepdim=True) + add_69 = mean_23 + 1e-06 + mean_23 = None + x_n_23 = x_g_23 / add_69 + x_g_23 = add_69 = None + view_46 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_47 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_23 = x_267 * x_n_23 + x_n_23 = None + addcmul_23 = torch.addcmul(view_46, view_47, mul_23) + view_46 = view_47 = mul_23 = None + x_268 = x_267 + addcmul_23 + x_267 = addcmul_23 = None + x_269 = torch._C._nn.linear( + x_268, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_268 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_270 = torch.nn.functional.dropout(x_269, 0.0, False, False) + x_269 = None + x_271 = x_270.permute(0, 3, 1, 2) + x_270 = None + x_272 = x_271 + x_261 + x_271 = x_261 = None + x_273 = torch.conv2d( + x_272, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = (None) + x_274 = x_273.permute(0, 2, 3, 1) + x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_, + 1e-06, + ) + x_274 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = (None) + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_277 = torch._C._nn.gelu(x_276) + x_276 = None + x_278 = torch.nn.functional.dropout(x_277, 0.0, False, False) + x_277 = None + x_g_24 = x_278.norm(p=2, dim=(1, 2), keepdim=True) + mean_24 = x_g_24.mean(dim=-1, keepdim=True) + add_72 = mean_24 + 1e-06 + mean_24 = None + x_n_24 = x_g_24 / add_72 + x_g_24 = add_72 = None + view_48 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_49 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_24 = x_278 * x_n_24 + x_n_24 = None + addcmul_24 = torch.addcmul(view_48, view_49, mul_24) + view_48 = view_49 = mul_24 = None + x_279 = x_278 + addcmul_24 + x_278 = addcmul_24 = None + x_280 = torch._C._nn.linear( + x_279, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_279 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_281 = torch.nn.functional.dropout(x_280, 0.0, False, False) + x_280 = None + x_282 = x_281.permute(0, 3, 1, 2) + x_281 = None + x_283 = x_282 + x_272 + x_282 = x_272 = None + x_284 = torch.conv2d( + x_283, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = (None) + x_285 = x_284.permute(0, 2, 3, 1) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_, + 1e-06, + ) + x_285 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = (None) + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_288 = torch._C._nn.gelu(x_287) + x_287 = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + x_g_25 = x_289.norm(p=2, dim=(1, 2), keepdim=True) + mean_25 = x_g_25.mean(dim=-1, keepdim=True) + add_75 = mean_25 + 1e-06 + mean_25 = None + x_n_25 = x_g_25 / add_75 + x_g_25 = add_75 = None + view_50 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_51 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_25 = x_289 * x_n_25 + x_n_25 = None + addcmul_25 = torch.addcmul(view_50, view_51, mul_25) + view_50 = view_51 = mul_25 = None + x_290 = x_289 + addcmul_25 + x_289 = addcmul_25 = None + x_291 = torch._C._nn.linear( + x_290, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_290 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_292 = torch.nn.functional.dropout(x_291, 0.0, False, False) + x_291 = None + x_293 = x_292.permute(0, 3, 1, 2) + x_292 = None + x_294 = x_293 + x_283 + x_293 = x_283 = None + x_295 = torch.conv2d( + x_294, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = (None) + x_296 = x_295.permute(0, 2, 3, 1) + x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_, + 1e-06, + ) + x_296 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = (None) + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_299 = torch._C._nn.gelu(x_298) + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_g_26 = x_300.norm(p=2, dim=(1, 2), keepdim=True) + mean_26 = x_g_26.mean(dim=-1, keepdim=True) + add_78 = mean_26 + 1e-06 + mean_26 = None + x_n_26 = x_g_26 / add_78 + x_g_26 = add_78 = None + view_52 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_53 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_26 = x_300 * x_n_26 + x_n_26 = None + addcmul_26 = torch.addcmul(view_52, view_53, mul_26) + view_52 = view_53 = mul_26 = None + x_301 = x_300 + addcmul_26 + x_300 = addcmul_26 = None + x_302 = torch._C._nn.linear( + x_301, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_301 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_303 = torch.nn.functional.dropout(x_302, 0.0, False, False) + x_302 = None + x_304 = x_303.permute(0, 3, 1, 2) + x_303 = None + x_305 = x_304 + x_294 + x_304 = x_294 = None + x_306 = torch.conv2d( + x_305, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = (None) + x_307 = x_306.permute(0, 2, 3, 1) + x_306 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_, + 1e-06, + ) + x_307 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = (None) + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_310 = torch._C._nn.gelu(x_309) + x_309 = None + x_311 = torch.nn.functional.dropout(x_310, 0.0, False, False) + x_310 = None + x_g_27 = x_311.norm(p=2, dim=(1, 2), keepdim=True) + mean_27 = x_g_27.mean(dim=-1, keepdim=True) + add_81 = mean_27 + 1e-06 + mean_27 = None + x_n_27 = x_g_27 / add_81 + x_g_27 = add_81 = None + view_54 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_55 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_27 = x_311 * x_n_27 + x_n_27 = None + addcmul_27 = torch.addcmul(view_54, view_55, mul_27) + view_54 = view_55 = mul_27 = None + x_312 = x_311 + addcmul_27 + x_311 = addcmul_27 = None + x_313 = torch._C._nn.linear( + x_312, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_312 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_314 = torch.nn.functional.dropout(x_313, 0.0, False, False) + x_313 = None + x_315 = x_314.permute(0, 3, 1, 2) + x_314 = None + x_316 = x_315 + x_305 + x_315 = x_305 = None + x_317 = torch.conv2d( + x_316, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = (None) + x_318 = x_317.permute(0, 2, 3, 1) + x_317 = None + x_319 = torch.nn.functional.layer_norm( + x_318, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_, + 1e-06, + ) + x_318 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = (None) + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_321 = torch._C._nn.gelu(x_320) + x_320 = None + x_322 = torch.nn.functional.dropout(x_321, 0.0, False, False) + x_321 = None + x_g_28 = x_322.norm(p=2, dim=(1, 2), keepdim=True) + mean_28 = x_g_28.mean(dim=-1, keepdim=True) + add_84 = mean_28 + 1e-06 + mean_28 = None + x_n_28 = x_g_28 / add_84 + x_g_28 = add_84 = None + view_56 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_57 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_28 = x_322 * x_n_28 + x_n_28 = None + addcmul_28 = torch.addcmul(view_56, view_57, mul_28) + view_56 = view_57 = mul_28 = None + x_323 = x_322 + addcmul_28 + x_322 = addcmul_28 = None + x_324 = torch._C._nn.linear( + x_323, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_323 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_325 = torch.nn.functional.dropout(x_324, 0.0, False, False) + x_324 = None + x_326 = x_325.permute(0, 3, 1, 2) + x_325 = None + x_327 = x_326 + x_316 + x_326 = x_316 = None + x_328 = torch.conv2d( + x_327, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = (None) + x_329 = x_328.permute(0, 2, 3, 1) + x_328 = None + x_330 = torch.nn.functional.layer_norm( + x_329, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_, + 1e-06, + ) + x_329 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = (None) + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_332 = torch._C._nn.gelu(x_331) + x_331 = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_g_29 = x_333.norm(p=2, dim=(1, 2), keepdim=True) + mean_29 = x_g_29.mean(dim=-1, keepdim=True) + add_87 = mean_29 + 1e-06 + mean_29 = None + x_n_29 = x_g_29 / add_87 + x_g_29 = add_87 = None + view_58 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_59 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_29 = x_333 * x_n_29 + x_n_29 = None + addcmul_29 = torch.addcmul(view_58, view_59, mul_29) + view_58 = view_59 = mul_29 = None + x_334 = x_333 + addcmul_29 + x_333 = addcmul_29 = None + x_335 = torch._C._nn.linear( + x_334, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_334 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_336 = torch.nn.functional.dropout(x_335, 0.0, False, False) + x_335 = None + x_337 = x_336.permute(0, 3, 1, 2) + x_336 = None + x_338 = x_337 + x_327 + x_337 = x_327 = None + x_339 = torch.conv2d( + x_338, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = (None) + x_340 = x_339.permute(0, 2, 3, 1) + x_339 = None + x_341 = torch.nn.functional.layer_norm( + x_340, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_, + 1e-06, + ) + x_340 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = (None) + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_343 = torch._C._nn.gelu(x_342) + x_342 = None + x_344 = torch.nn.functional.dropout(x_343, 0.0, False, False) + x_343 = None + x_g_30 = x_344.norm(p=2, dim=(1, 2), keepdim=True) + mean_30 = x_g_30.mean(dim=-1, keepdim=True) + add_90 = mean_30 + 1e-06 + mean_30 = None + x_n_30 = x_g_30 / add_90 + x_g_30 = add_90 = None + view_60 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_61 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_30 = x_344 * x_n_30 + x_n_30 = None + addcmul_30 = torch.addcmul(view_60, view_61, mul_30) + view_60 = view_61 = mul_30 = None + x_345 = x_344 + addcmul_30 + x_344 = addcmul_30 = None + x_346 = torch._C._nn.linear( + x_345, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_345 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_347 = torch.nn.functional.dropout(x_346, 0.0, False, False) + x_346 = None + x_348 = x_347.permute(0, 3, 1, 2) + x_347 = None + x_349 = x_348 + x_338 + x_348 = x_338 = None + x_350 = torch.conv2d( + x_349, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = (None) + x_351 = x_350.permute(0, 2, 3, 1) + x_350 = None + x_352 = torch.nn.functional.layer_norm( + x_351, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_, + 1e-06, + ) + x_351 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = (None) + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_354 = torch._C._nn.gelu(x_353) + x_353 = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_g_31 = x_355.norm(p=2, dim=(1, 2), keepdim=True) + mean_31 = x_g_31.mean(dim=-1, keepdim=True) + add_93 = mean_31 + 1e-06 + mean_31 = None + x_n_31 = x_g_31 / add_93 + x_g_31 = add_93 = None + view_62 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_63 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_31 = x_355 * x_n_31 + x_n_31 = None + addcmul_31 = torch.addcmul(view_62, view_63, mul_31) + view_62 = view_63 = mul_31 = None + x_356 = x_355 + addcmul_31 + x_355 = addcmul_31 = None + x_357 = torch._C._nn.linear( + x_356, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_356 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_358 = torch.nn.functional.dropout(x_357, 0.0, False, False) + x_357 = None + x_359 = x_358.permute(0, 3, 1, 2) + x_358 = None + x_360 = x_359 + x_349 + x_359 = x_349 = None + x_361 = torch.conv2d( + x_360, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1408, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = (None) + x_362 = x_361.permute(0, 2, 3, 1) + x_361 = None + x_363 = torch.nn.functional.layer_norm( + x_362, + (1408,), + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_, + 1e-06, + ) + x_362 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = (None) + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_365 = torch._C._nn.gelu(x_364) + x_364 = None + x_366 = torch.nn.functional.dropout(x_365, 0.0, False, False) + x_365 = None + x_g_32 = x_366.norm(p=2, dim=(1, 2), keepdim=True) + mean_32 = x_g_32.mean(dim=-1, keepdim=True) + add_96 = mean_32 + 1e-06 + mean_32 = None + x_n_32 = x_g_32 / add_96 + x_g_32 = add_96 = None + view_64 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_65 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_32 = x_366 * x_n_32 + x_n_32 = None + addcmul_32 = torch.addcmul(view_64, view_65, mul_32) + view_64 = view_65 = mul_32 = None + x_367 = x_366 + addcmul_32 + x_366 = addcmul_32 = None + x_368 = torch._C._nn.linear( + x_367, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_367 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_369 = torch.nn.functional.dropout(x_368, 0.0, False, False) + x_368 = None + x_370 = x_369.permute(0, 3, 1, 2) + x_369 = None + x_371 = x_370 + x_360 + x_370 = x_360 = None + x_372 = x_371.permute(0, 2, 3, 1) + x_371 = None + x_373 = torch.nn.functional.layer_norm( + x_372, + (1408,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_372 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_374 = x_373.permute(0, 3, 1, 2) + x_373 = None + input_4 = torch.conv2d( + x_374, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_374 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_375 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 2816, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_376 = x_375.permute(0, 2, 3, 1) + x_375 = None + x_377 = torch.nn.functional.layer_norm( + x_376, + (2816,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_376 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_378 = torch._C._nn.linear( + x_377, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_377 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_379 = torch._C._nn.gelu(x_378) + x_378 = None + x_380 = torch.nn.functional.dropout(x_379, 0.0, False, False) + x_379 = None + x_g_33 = x_380.norm(p=2, dim=(1, 2), keepdim=True) + mean_33 = x_g_33.mean(dim=-1, keepdim=True) + add_99 = mean_33 + 1e-06 + mean_33 = None + x_n_33 = x_g_33 / add_99 + x_g_33 = add_99 = None + view_66 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_67 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_33 = x_380 * x_n_33 + x_n_33 = None + addcmul_33 = torch.addcmul(view_66, view_67, mul_33) + view_66 = view_67 = mul_33 = None + x_381 = x_380 + addcmul_33 + x_380 = addcmul_33 = None + x_382 = torch._C._nn.linear( + x_381, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_381 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_383 = torch.nn.functional.dropout(x_382, 0.0, False, False) + x_382 = None + x_384 = x_383.permute(0, 3, 1, 2) + x_383 = None + x_385 = x_384 + input_4 + x_384 = input_4 = None + x_386 = torch.conv2d( + x_385, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 2816, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_387 = x_386.permute(0, 2, 3, 1) + x_386 = None + x_388 = torch.nn.functional.layer_norm( + x_387, + (2816,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_387 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_389 = torch._C._nn.linear( + x_388, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_388 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_390 = torch._C._nn.gelu(x_389) + x_389 = None + x_391 = torch.nn.functional.dropout(x_390, 0.0, False, False) + x_390 = None + x_g_34 = x_391.norm(p=2, dim=(1, 2), keepdim=True) + mean_34 = x_g_34.mean(dim=-1, keepdim=True) + add_102 = mean_34 + 1e-06 + mean_34 = None + x_n_34 = x_g_34 / add_102 + x_g_34 = add_102 = None + view_68 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_69 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_34 = x_391 * x_n_34 + x_n_34 = None + addcmul_34 = torch.addcmul(view_68, view_69, mul_34) + view_68 = view_69 = mul_34 = None + x_392 = x_391 + addcmul_34 + x_391 = addcmul_34 = None + x_393 = torch._C._nn.linear( + x_392, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_392 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_394 = torch.nn.functional.dropout(x_393, 0.0, False, False) + x_393 = None + x_395 = x_394.permute(0, 3, 1, 2) + x_394 = None + x_396 = x_395 + x_385 + x_395 = x_385 = None + x_397 = torch.conv2d( + x_396, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 2816, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_398 = x_397.permute(0, 2, 3, 1) + x_397 = None + x_399 = torch.nn.functional.layer_norm( + x_398, + (2816,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_398 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_400 = torch._C._nn.linear( + x_399, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_399 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_401 = torch._C._nn.gelu(x_400) + x_400 = None + x_402 = torch.nn.functional.dropout(x_401, 0.0, False, False) + x_401 = None + x_g_35 = x_402.norm(p=2, dim=(1, 2), keepdim=True) + mean_35 = x_g_35.mean(dim=-1, keepdim=True) + add_105 = mean_35 + 1e-06 + mean_35 = None + x_n_35 = x_g_35 / add_105 + x_g_35 = add_105 = None + view_70 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_71 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_35 = x_402 * x_n_35 + x_n_35 = None + addcmul_35 = torch.addcmul(view_70, view_71, mul_35) + view_70 = view_71 = mul_35 = None + x_403 = x_402 + addcmul_35 + x_402 = addcmul_35 = None + x_404 = torch._C._nn.linear( + x_403, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_403 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_405 = torch.nn.functional.dropout(x_404, 0.0, False, False) + x_404 = None + x_406 = x_405.permute(0, 3, 1, 2) + x_405 = None + x_407 = x_406 + x_396 + x_406 = x_396 = None + x_408 = torch.nn.functional.adaptive_avg_pool2d(x_407, 1) + x_407 = None + x_409 = x_408.permute(0, 2, 3, 1) + x_408 = None + x_410 = torch.nn.functional.layer_norm( + x_409, + (2816,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_409 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_411 = x_410.permute(0, 3, 1, 2) + x_410 = None + x_412 = x_411.flatten(1, -1) + x_411 = None + x_413 = torch.nn.functional.dropout(x_412, 0.0, False, False) + x_412 = None + x_414 = torch._C._nn.linear( + x_413, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_413 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_414,) diff --git a/samples/timm/convnextv2_huge.fcmae_ft_in1k/weight_meta.py b/samples/timm/convnextv2_huge.fcmae_ft_in1k/weight_meta.py new file mode 100644 index 0000000000..809be42e6f --- /dev/null +++ b/samples/timm/convnextv2_huge.fcmae_ft_in1k/weight_meta.py @@ -0,0 +1,3830 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [352, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.222 + std = 1.284 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [352, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1408, 352] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [352, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [352, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1408, 352] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [352, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [352, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1408, 352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [352, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [704, 352, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [704, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2816, 704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [704, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [704, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2816, 704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [704, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [704, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2816, 704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [704, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [1408, 704, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_" + shape = [1408, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [5632, 1408] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_" + shape = [5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [1408, 5632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [2816, 1408, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [2816, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [11264, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [2816, 11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [2816, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [11264, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [2816, 11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [2816, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [11264, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [11264] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [2816, 11264] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [2816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 2816] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_large.fcmae/graph_hash.txt b/samples/timm/convnextv2_large.fcmae/graph_hash.txt new file mode 100644 index 0000000000..e31486c4f9 --- /dev/null +++ b/samples/timm/convnextv2_large.fcmae/graph_hash.txt @@ -0,0 +1 @@ +5edc6a6e18554e542bf2d9176814b8693f8f0f277af6cc827ca95f027f217354 \ No newline at end of file diff --git a/samples/timm/convnextv2_large.fcmae/graph_net.json b/samples/timm/convnextv2_large.fcmae/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnextv2_large.fcmae/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_large.fcmae/input_meta.py b/samples/timm/convnextv2_large.fcmae/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_large.fcmae/input_tensor_constraints.py b/samples/timm/convnextv2_large.fcmae/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_large.fcmae/model.py b/samples/timm/convnextv2_large.fcmae/model.py new file mode 100644 index 0000000000..dae73df9e0 --- /dev/null +++ b/samples/timm/convnextv2_large.fcmae/model.py @@ -0,0 +1,3267 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (192,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (192,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_g = x_8.norm(p=2, dim=(1, 2), keepdim=True) + mean = x_g.mean(dim=-1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_8 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_9 = x_8 + addcmul + x_8 = addcmul = None + x_10 = torch._C._nn.linear( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = x_11.permute(0, 3, 1, 2) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (192,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_g_1 = x_19.norm(p=2, dim=(1, 2), keepdim=True) + mean_1 = x_g_1.mean(dim=-1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_19 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_20 = x_19 + addcmul_1 + x_19 = addcmul_1 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (192,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_g_2 = x_30.norm(p=2, dim=(1, 2), keepdim=True) + mean_2 = x_g_2.mean(dim=-1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_30 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_31 = x_30 + addcmul_2 + x_30 = addcmul_2 = None + x_32 = torch._C._nn.linear( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + x_34 = x_33.permute(0, 3, 1, 2) + x_33 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (192,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (384,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_g_3 = x_44.norm(p=2, dim=(1, 2), keepdim=True) + mean_3 = x_g_3.mean(dim=-1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_44 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_45 = x_44 + addcmul_3 + x_44 = addcmul_3 = None + x_46 = torch._C._nn.linear( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (384,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_g_4 = x_55.norm(p=2, dim=(1, 2), keepdim=True) + mean_4 = x_g_4.mean(dim=-1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_55 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_56 = x_55 + addcmul_4 + x_55 = addcmul_4 = None + x_57 = torch._C._nn.linear( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_58 = torch.nn.functional.dropout(x_57, 0.0, False, False) + x_57 = None + x_59 = x_58.permute(0, 3, 1, 2) + x_58 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (384,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_g_5 = x_66.norm(p=2, dim=(1, 2), keepdim=True) + mean_5 = x_g_5.mean(dim=-1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_66 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_67 = x_66 + addcmul_5 + x_66 = addcmul_5 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = x_69.permute(0, 3, 1, 2) + x_69 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (384,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_g_6 = x_80.norm(p=2, dim=(1, 2), keepdim=True) + mean_6 = x_g_6.mean(dim=-1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_80 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_81 = x_80 + addcmul_6 + x_80 = addcmul_6 = None + x_82 = torch._C._nn.linear( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + x_84 = x_83.permute(0, 3, 1, 2) + x_83 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_g_7 = x_91.norm(p=2, dim=(1, 2), keepdim=True) + mean_7 = x_g_7.mean(dim=-1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_91 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_92 = x_91 + addcmul_7 + x_91 = addcmul_7 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_g_8 = x_102.norm(p=2, dim=(1, 2), keepdim=True) + mean_8 = x_g_8.mean(dim=-1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_102 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_103 = x_102 + addcmul_8 + x_102 = addcmul_8 = None + x_104 = torch._C._nn.linear( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = x_105.permute(0, 3, 1, 2) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_g_9 = x_113.norm(p=2, dim=(1, 2), keepdim=True) + mean_9 = x_g_9.mean(dim=-1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_113 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_114 = x_113 + addcmul_9 + x_113 = addcmul_9 = None + x_115 = torch._C._nn.linear( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + x_117 = x_116.permute(0, 3, 1, 2) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_g_10 = x_124.norm(p=2, dim=(1, 2), keepdim=True) + mean_10 = x_g_10.mean(dim=-1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_124 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_125 = x_124 + addcmul_10 + x_124 = addcmul_10 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = x_127.permute(0, 3, 1, 2) + x_127 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_g_11 = x_135.norm(p=2, dim=(1, 2), keepdim=True) + mean_11 = x_g_11.mean(dim=-1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_135 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_136 = x_135 + addcmul_11 + x_135 = addcmul_11 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_138.permute(0, 3, 1, 2) + x_138 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_g_12 = x_146.norm(p=2, dim=(1, 2), keepdim=True) + mean_12 = x_g_12.mean(dim=-1, keepdim=True) + add_36 = mean_12 + 1e-06 + mean_12 = None + x_n_12 = x_g_12 / add_36 + x_g_12 = add_36 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_12 = x_146 * x_n_12 + x_n_12 = None + addcmul_12 = torch.addcmul(view_24, view_25, mul_12) + view_24 = view_25 = mul_12 = None + x_147 = x_146 + addcmul_12 + x_146 = addcmul_12 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_149.permute(0, 3, 1, 2) + x_149 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_g_13 = x_157.norm(p=2, dim=(1, 2), keepdim=True) + mean_13 = x_g_13.mean(dim=-1, keepdim=True) + add_39 = mean_13 + 1e-06 + mean_13 = None + x_n_13 = x_g_13 / add_39 + x_g_13 = add_39 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_13 = x_157 * x_n_13 + x_n_13 = None + addcmul_13 = torch.addcmul(view_26, view_27, mul_13) + view_26 = view_27 = mul_13 = None + x_158 = x_157 + addcmul_13 + x_157 = addcmul_13 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_160.permute(0, 3, 1, 2) + x_160 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_g_14 = x_168.norm(p=2, dim=(1, 2), keepdim=True) + mean_14 = x_g_14.mean(dim=-1, keepdim=True) + add_42 = mean_14 + 1e-06 + mean_14 = None + x_n_14 = x_g_14 / add_42 + x_g_14 = add_42 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_14 = x_168 * x_n_14 + x_n_14 = None + addcmul_14 = torch.addcmul(view_28, view_29, mul_14) + view_28 = view_29 = mul_14 = None + x_169 = x_168 + addcmul_14 + x_168 = addcmul_14 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_171.permute(0, 3, 1, 2) + x_171 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = (None) + x_175 = x_174.permute(0, 2, 3, 1) + x_174 = None + x_176 = torch.nn.functional.layer_norm( + x_175, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_, + 1e-06, + ) + x_175 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = (None) + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_176 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_178 = torch._C._nn.gelu(x_177) + x_177 = None + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_g_15 = x_179.norm(p=2, dim=(1, 2), keepdim=True) + mean_15 = x_g_15.mean(dim=-1, keepdim=True) + add_45 = mean_15 + 1e-06 + mean_15 = None + x_n_15 = x_g_15 / add_45 + x_g_15 = add_45 = None + view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_15 = x_179 * x_n_15 + x_n_15 = None + addcmul_15 = torch.addcmul(view_30, view_31, mul_15) + view_30 = view_31 = mul_15 = None + x_180 = x_179 + addcmul_15 + x_179 = addcmul_15 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_182.permute(0, 3, 1, 2) + x_182 = None + x_184 = x_183 + x_173 + x_183 = x_173 = None + x_185 = torch.conv2d( + x_184, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = (None) + x_186 = x_185.permute(0, 2, 3, 1) + x_185 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = (None) + x_188 = torch._C._nn.linear( + x_187, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_187 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_189 = torch._C._nn.gelu(x_188) + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_g_16 = x_190.norm(p=2, dim=(1, 2), keepdim=True) + mean_16 = x_g_16.mean(dim=-1, keepdim=True) + add_48 = mean_16 + 1e-06 + mean_16 = None + x_n_16 = x_g_16 / add_48 + x_g_16 = add_48 = None + view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_16 = x_190 * x_n_16 + x_n_16 = None + addcmul_16 = torch.addcmul(view_32, view_33, mul_16) + view_32 = view_33 = mul_16 = None + x_191 = x_190 + addcmul_16 + x_190 = addcmul_16 = None + x_192 = torch._C._nn.linear( + x_191, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_191 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_193.permute(0, 3, 1, 2) + x_193 = None + x_195 = x_194 + x_184 + x_194 = x_184 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = (None) + x_197 = x_196.permute(0, 2, 3, 1) + x_196 = None + x_198 = torch.nn.functional.layer_norm( + x_197, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_, + 1e-06, + ) + x_197 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = (None) + x_199 = torch._C._nn.linear( + x_198, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_198 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_200 = torch._C._nn.gelu(x_199) + x_199 = None + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_g_17 = x_201.norm(p=2, dim=(1, 2), keepdim=True) + mean_17 = x_g_17.mean(dim=-1, keepdim=True) + add_51 = mean_17 + 1e-06 + mean_17 = None + x_n_17 = x_g_17 / add_51 + x_g_17 = add_51 = None + view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_17 = x_201 * x_n_17 + x_n_17 = None + addcmul_17 = torch.addcmul(view_34, view_35, mul_17) + view_34 = view_35 = mul_17 = None + x_202 = x_201 + addcmul_17 + x_201 = addcmul_17 = None + x_203 = torch._C._nn.linear( + x_202, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_202 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_205 = x_204.permute(0, 3, 1, 2) + x_204 = None + x_206 = x_205 + x_195 + x_205 = x_195 = None + x_207 = torch.conv2d( + x_206, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = (None) + x_208 = x_207.permute(0, 2, 3, 1) + x_207 = None + x_209 = torch.nn.functional.layer_norm( + x_208, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_, + 1e-06, + ) + x_208 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = (None) + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_209 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_211 = torch._C._nn.gelu(x_210) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_g_18 = x_212.norm(p=2, dim=(1, 2), keepdim=True) + mean_18 = x_g_18.mean(dim=-1, keepdim=True) + add_54 = mean_18 + 1e-06 + mean_18 = None + x_n_18 = x_g_18 / add_54 + x_g_18 = add_54 = None + view_36 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_37 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_18 = x_212 * x_n_18 + x_n_18 = None + addcmul_18 = torch.addcmul(view_36, view_37, mul_18) + view_36 = view_37 = mul_18 = None + x_213 = x_212 + addcmul_18 + x_212 = addcmul_18 = None + x_214 = torch._C._nn.linear( + x_213, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_213 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = x_215.permute(0, 3, 1, 2) + x_215 = None + x_217 = x_216 + x_206 + x_216 = x_206 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = (None) + x_219 = x_218.permute(0, 2, 3, 1) + x_218 = None + x_220 = torch.nn.functional.layer_norm( + x_219, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_, + 1e-06, + ) + x_219 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = (None) + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_220 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_222 = torch._C._nn.gelu(x_221) + x_221 = None + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_g_19 = x_223.norm(p=2, dim=(1, 2), keepdim=True) + mean_19 = x_g_19.mean(dim=-1, keepdim=True) + add_57 = mean_19 + 1e-06 + mean_19 = None + x_n_19 = x_g_19 / add_57 + x_g_19 = add_57 = None + view_38 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_39 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_19 = x_223 * x_n_19 + x_n_19 = None + addcmul_19 = torch.addcmul(view_38, view_39, mul_19) + view_38 = view_39 = mul_19 = None + x_224 = x_223 + addcmul_19 + x_223 = addcmul_19 = None + x_225 = torch._C._nn.linear( + x_224, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_224 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_226 = torch.nn.functional.dropout(x_225, 0.0, False, False) + x_225 = None + x_227 = x_226.permute(0, 3, 1, 2) + x_226 = None + x_228 = x_227 + x_217 + x_227 = x_217 = None + x_229 = torch.conv2d( + x_228, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = (None) + x_230 = x_229.permute(0, 2, 3, 1) + x_229 = None + x_231 = torch.nn.functional.layer_norm( + x_230, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_, + 1e-06, + ) + x_230 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = (None) + x_232 = torch._C._nn.linear( + x_231, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_231 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_233 = torch._C._nn.gelu(x_232) + x_232 = None + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_g_20 = x_234.norm(p=2, dim=(1, 2), keepdim=True) + mean_20 = x_g_20.mean(dim=-1, keepdim=True) + add_60 = mean_20 + 1e-06 + mean_20 = None + x_n_20 = x_g_20 / add_60 + x_g_20 = add_60 = None + view_40 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_41 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_20 = x_234 * x_n_20 + x_n_20 = None + addcmul_20 = torch.addcmul(view_40, view_41, mul_20) + view_40 = view_41 = mul_20 = None + x_235 = x_234 + addcmul_20 + x_234 = addcmul_20 = None + x_236 = torch._C._nn.linear( + x_235, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_235 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_237 = torch.nn.functional.dropout(x_236, 0.0, False, False) + x_236 = None + x_238 = x_237.permute(0, 3, 1, 2) + x_237 = None + x_239 = x_238 + x_228 + x_238 = x_228 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = (None) + x_241 = x_240.permute(0, 2, 3, 1) + x_240 = None + x_242 = torch.nn.functional.layer_norm( + x_241, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_, + 1e-06, + ) + x_241 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = (None) + x_243 = torch._C._nn.linear( + x_242, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_242 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_244 = torch._C._nn.gelu(x_243) + x_243 = None + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_g_21 = x_245.norm(p=2, dim=(1, 2), keepdim=True) + mean_21 = x_g_21.mean(dim=-1, keepdim=True) + add_63 = mean_21 + 1e-06 + mean_21 = None + x_n_21 = x_g_21 / add_63 + x_g_21 = add_63 = None + view_42 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_43 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_21 = x_245 * x_n_21 + x_n_21 = None + addcmul_21 = torch.addcmul(view_42, view_43, mul_21) + view_42 = view_43 = mul_21 = None + x_246 = x_245 + addcmul_21 + x_245 = addcmul_21 = None + x_247 = torch._C._nn.linear( + x_246, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_246 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_248 = torch.nn.functional.dropout(x_247, 0.0, False, False) + x_247 = None + x_249 = x_248.permute(0, 3, 1, 2) + x_248 = None + x_250 = x_249 + x_239 + x_249 = x_239 = None + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = (None) + x_252 = x_251.permute(0, 2, 3, 1) + x_251 = None + x_253 = torch.nn.functional.layer_norm( + x_252, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_, + 1e-06, + ) + x_252 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = (None) + x_254 = torch._C._nn.linear( + x_253, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_253 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_255 = torch._C._nn.gelu(x_254) + x_254 = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_g_22 = x_256.norm(p=2, dim=(1, 2), keepdim=True) + mean_22 = x_g_22.mean(dim=-1, keepdim=True) + add_66 = mean_22 + 1e-06 + mean_22 = None + x_n_22 = x_g_22 / add_66 + x_g_22 = add_66 = None + view_44 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_45 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_22 = x_256 * x_n_22 + x_n_22 = None + addcmul_22 = torch.addcmul(view_44, view_45, mul_22) + view_44 = view_45 = mul_22 = None + x_257 = x_256 + addcmul_22 + x_256 = addcmul_22 = None + x_258 = torch._C._nn.linear( + x_257, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_257 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = x_259.permute(0, 3, 1, 2) + x_259 = None + x_261 = x_260 + x_250 + x_260 = x_250 = None + x_262 = torch.conv2d( + x_261, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = (None) + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = (None) + x_265 = torch._C._nn.linear( + x_264, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_264 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_266 = torch._C._nn.gelu(x_265) + x_265 = None + x_267 = torch.nn.functional.dropout(x_266, 0.0, False, False) + x_266 = None + x_g_23 = x_267.norm(p=2, dim=(1, 2), keepdim=True) + mean_23 = x_g_23.mean(dim=-1, keepdim=True) + add_69 = mean_23 + 1e-06 + mean_23 = None + x_n_23 = x_g_23 / add_69 + x_g_23 = add_69 = None + view_46 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_47 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_23 = x_267 * x_n_23 + x_n_23 = None + addcmul_23 = torch.addcmul(view_46, view_47, mul_23) + view_46 = view_47 = mul_23 = None + x_268 = x_267 + addcmul_23 + x_267 = addcmul_23 = None + x_269 = torch._C._nn.linear( + x_268, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_268 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_270 = torch.nn.functional.dropout(x_269, 0.0, False, False) + x_269 = None + x_271 = x_270.permute(0, 3, 1, 2) + x_270 = None + x_272 = x_271 + x_261 + x_271 = x_261 = None + x_273 = torch.conv2d( + x_272, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = (None) + x_274 = x_273.permute(0, 2, 3, 1) + x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_, + 1e-06, + ) + x_274 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = (None) + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_277 = torch._C._nn.gelu(x_276) + x_276 = None + x_278 = torch.nn.functional.dropout(x_277, 0.0, False, False) + x_277 = None + x_g_24 = x_278.norm(p=2, dim=(1, 2), keepdim=True) + mean_24 = x_g_24.mean(dim=-1, keepdim=True) + add_72 = mean_24 + 1e-06 + mean_24 = None + x_n_24 = x_g_24 / add_72 + x_g_24 = add_72 = None + view_48 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_49 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_24 = x_278 * x_n_24 + x_n_24 = None + addcmul_24 = torch.addcmul(view_48, view_49, mul_24) + view_48 = view_49 = mul_24 = None + x_279 = x_278 + addcmul_24 + x_278 = addcmul_24 = None + x_280 = torch._C._nn.linear( + x_279, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_279 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_281 = torch.nn.functional.dropout(x_280, 0.0, False, False) + x_280 = None + x_282 = x_281.permute(0, 3, 1, 2) + x_281 = None + x_283 = x_282 + x_272 + x_282 = x_272 = None + x_284 = torch.conv2d( + x_283, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = (None) + x_285 = x_284.permute(0, 2, 3, 1) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_, + 1e-06, + ) + x_285 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = (None) + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_288 = torch._C._nn.gelu(x_287) + x_287 = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + x_g_25 = x_289.norm(p=2, dim=(1, 2), keepdim=True) + mean_25 = x_g_25.mean(dim=-1, keepdim=True) + add_75 = mean_25 + 1e-06 + mean_25 = None + x_n_25 = x_g_25 / add_75 + x_g_25 = add_75 = None + view_50 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_51 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_25 = x_289 * x_n_25 + x_n_25 = None + addcmul_25 = torch.addcmul(view_50, view_51, mul_25) + view_50 = view_51 = mul_25 = None + x_290 = x_289 + addcmul_25 + x_289 = addcmul_25 = None + x_291 = torch._C._nn.linear( + x_290, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_290 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_292 = torch.nn.functional.dropout(x_291, 0.0, False, False) + x_291 = None + x_293 = x_292.permute(0, 3, 1, 2) + x_292 = None + x_294 = x_293 + x_283 + x_293 = x_283 = None + x_295 = torch.conv2d( + x_294, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = (None) + x_296 = x_295.permute(0, 2, 3, 1) + x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_, + 1e-06, + ) + x_296 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = (None) + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_299 = torch._C._nn.gelu(x_298) + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_g_26 = x_300.norm(p=2, dim=(1, 2), keepdim=True) + mean_26 = x_g_26.mean(dim=-1, keepdim=True) + add_78 = mean_26 + 1e-06 + mean_26 = None + x_n_26 = x_g_26 / add_78 + x_g_26 = add_78 = None + view_52 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_53 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_26 = x_300 * x_n_26 + x_n_26 = None + addcmul_26 = torch.addcmul(view_52, view_53, mul_26) + view_52 = view_53 = mul_26 = None + x_301 = x_300 + addcmul_26 + x_300 = addcmul_26 = None + x_302 = torch._C._nn.linear( + x_301, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_301 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_303 = torch.nn.functional.dropout(x_302, 0.0, False, False) + x_302 = None + x_304 = x_303.permute(0, 3, 1, 2) + x_303 = None + x_305 = x_304 + x_294 + x_304 = x_294 = None + x_306 = torch.conv2d( + x_305, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = (None) + x_307 = x_306.permute(0, 2, 3, 1) + x_306 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_, + 1e-06, + ) + x_307 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = (None) + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_310 = torch._C._nn.gelu(x_309) + x_309 = None + x_311 = torch.nn.functional.dropout(x_310, 0.0, False, False) + x_310 = None + x_g_27 = x_311.norm(p=2, dim=(1, 2), keepdim=True) + mean_27 = x_g_27.mean(dim=-1, keepdim=True) + add_81 = mean_27 + 1e-06 + mean_27 = None + x_n_27 = x_g_27 / add_81 + x_g_27 = add_81 = None + view_54 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_55 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_27 = x_311 * x_n_27 + x_n_27 = None + addcmul_27 = torch.addcmul(view_54, view_55, mul_27) + view_54 = view_55 = mul_27 = None + x_312 = x_311 + addcmul_27 + x_311 = addcmul_27 = None + x_313 = torch._C._nn.linear( + x_312, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_312 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_314 = torch.nn.functional.dropout(x_313, 0.0, False, False) + x_313 = None + x_315 = x_314.permute(0, 3, 1, 2) + x_314 = None + x_316 = x_315 + x_305 + x_315 = x_305 = None + x_317 = torch.conv2d( + x_316, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = (None) + x_318 = x_317.permute(0, 2, 3, 1) + x_317 = None + x_319 = torch.nn.functional.layer_norm( + x_318, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_, + 1e-06, + ) + x_318 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = (None) + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_321 = torch._C._nn.gelu(x_320) + x_320 = None + x_322 = torch.nn.functional.dropout(x_321, 0.0, False, False) + x_321 = None + x_g_28 = x_322.norm(p=2, dim=(1, 2), keepdim=True) + mean_28 = x_g_28.mean(dim=-1, keepdim=True) + add_84 = mean_28 + 1e-06 + mean_28 = None + x_n_28 = x_g_28 / add_84 + x_g_28 = add_84 = None + view_56 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_57 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_28 = x_322 * x_n_28 + x_n_28 = None + addcmul_28 = torch.addcmul(view_56, view_57, mul_28) + view_56 = view_57 = mul_28 = None + x_323 = x_322 + addcmul_28 + x_322 = addcmul_28 = None + x_324 = torch._C._nn.linear( + x_323, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_323 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_325 = torch.nn.functional.dropout(x_324, 0.0, False, False) + x_324 = None + x_326 = x_325.permute(0, 3, 1, 2) + x_325 = None + x_327 = x_326 + x_316 + x_326 = x_316 = None + x_328 = torch.conv2d( + x_327, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = (None) + x_329 = x_328.permute(0, 2, 3, 1) + x_328 = None + x_330 = torch.nn.functional.layer_norm( + x_329, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_, + 1e-06, + ) + x_329 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = (None) + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_332 = torch._C._nn.gelu(x_331) + x_331 = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_g_29 = x_333.norm(p=2, dim=(1, 2), keepdim=True) + mean_29 = x_g_29.mean(dim=-1, keepdim=True) + add_87 = mean_29 + 1e-06 + mean_29 = None + x_n_29 = x_g_29 / add_87 + x_g_29 = add_87 = None + view_58 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_59 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_29 = x_333 * x_n_29 + x_n_29 = None + addcmul_29 = torch.addcmul(view_58, view_59, mul_29) + view_58 = view_59 = mul_29 = None + x_334 = x_333 + addcmul_29 + x_333 = addcmul_29 = None + x_335 = torch._C._nn.linear( + x_334, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_334 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_336 = torch.nn.functional.dropout(x_335, 0.0, False, False) + x_335 = None + x_337 = x_336.permute(0, 3, 1, 2) + x_336 = None + x_338 = x_337 + x_327 + x_337 = x_327 = None + x_339 = torch.conv2d( + x_338, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = (None) + x_340 = x_339.permute(0, 2, 3, 1) + x_339 = None + x_341 = torch.nn.functional.layer_norm( + x_340, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_, + 1e-06, + ) + x_340 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = (None) + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_343 = torch._C._nn.gelu(x_342) + x_342 = None + x_344 = torch.nn.functional.dropout(x_343, 0.0, False, False) + x_343 = None + x_g_30 = x_344.norm(p=2, dim=(1, 2), keepdim=True) + mean_30 = x_g_30.mean(dim=-1, keepdim=True) + add_90 = mean_30 + 1e-06 + mean_30 = None + x_n_30 = x_g_30 / add_90 + x_g_30 = add_90 = None + view_60 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_61 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_30 = x_344 * x_n_30 + x_n_30 = None + addcmul_30 = torch.addcmul(view_60, view_61, mul_30) + view_60 = view_61 = mul_30 = None + x_345 = x_344 + addcmul_30 + x_344 = addcmul_30 = None + x_346 = torch._C._nn.linear( + x_345, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_345 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_347 = torch.nn.functional.dropout(x_346, 0.0, False, False) + x_346 = None + x_348 = x_347.permute(0, 3, 1, 2) + x_347 = None + x_349 = x_348 + x_338 + x_348 = x_338 = None + x_350 = torch.conv2d( + x_349, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = (None) + x_351 = x_350.permute(0, 2, 3, 1) + x_350 = None + x_352 = torch.nn.functional.layer_norm( + x_351, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_, + 1e-06, + ) + x_351 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = (None) + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_354 = torch._C._nn.gelu(x_353) + x_353 = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_g_31 = x_355.norm(p=2, dim=(1, 2), keepdim=True) + mean_31 = x_g_31.mean(dim=-1, keepdim=True) + add_93 = mean_31 + 1e-06 + mean_31 = None + x_n_31 = x_g_31 / add_93 + x_g_31 = add_93 = None + view_62 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_63 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_31 = x_355 * x_n_31 + x_n_31 = None + addcmul_31 = torch.addcmul(view_62, view_63, mul_31) + view_62 = view_63 = mul_31 = None + x_356 = x_355 + addcmul_31 + x_355 = addcmul_31 = None + x_357 = torch._C._nn.linear( + x_356, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_356 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_358 = torch.nn.functional.dropout(x_357, 0.0, False, False) + x_357 = None + x_359 = x_358.permute(0, 3, 1, 2) + x_358 = None + x_360 = x_359 + x_349 + x_359 = x_349 = None + x_361 = torch.conv2d( + x_360, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = (None) + x_362 = x_361.permute(0, 2, 3, 1) + x_361 = None + x_363 = torch.nn.functional.layer_norm( + x_362, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_, + 1e-06, + ) + x_362 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = (None) + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_365 = torch._C._nn.gelu(x_364) + x_364 = None + x_366 = torch.nn.functional.dropout(x_365, 0.0, False, False) + x_365 = None + x_g_32 = x_366.norm(p=2, dim=(1, 2), keepdim=True) + mean_32 = x_g_32.mean(dim=-1, keepdim=True) + add_96 = mean_32 + 1e-06 + mean_32 = None + x_n_32 = x_g_32 / add_96 + x_g_32 = add_96 = None + view_64 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_65 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_32 = x_366 * x_n_32 + x_n_32 = None + addcmul_32 = torch.addcmul(view_64, view_65, mul_32) + view_64 = view_65 = mul_32 = None + x_367 = x_366 + addcmul_32 + x_366 = addcmul_32 = None + x_368 = torch._C._nn.linear( + x_367, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_367 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_369 = torch.nn.functional.dropout(x_368, 0.0, False, False) + x_368 = None + x_370 = x_369.permute(0, 3, 1, 2) + x_369 = None + x_371 = x_370 + x_360 + x_370 = x_360 = None + x_372 = x_371.permute(0, 2, 3, 1) + x_371 = None + x_373 = torch.nn.functional.layer_norm( + x_372, + (768,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_372 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_374 = x_373.permute(0, 3, 1, 2) + x_373 = None + input_4 = torch.conv2d( + x_374, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_374 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_375 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_376 = x_375.permute(0, 2, 3, 1) + x_375 = None + x_377 = torch.nn.functional.layer_norm( + x_376, + (1536,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_376 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_378 = torch._C._nn.linear( + x_377, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_377 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_379 = torch._C._nn.gelu(x_378) + x_378 = None + x_380 = torch.nn.functional.dropout(x_379, 0.0, False, False) + x_379 = None + x_g_33 = x_380.norm(p=2, dim=(1, 2), keepdim=True) + mean_33 = x_g_33.mean(dim=-1, keepdim=True) + add_99 = mean_33 + 1e-06 + mean_33 = None + x_n_33 = x_g_33 / add_99 + x_g_33 = add_99 = None + view_66 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_67 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_33 = x_380 * x_n_33 + x_n_33 = None + addcmul_33 = torch.addcmul(view_66, view_67, mul_33) + view_66 = view_67 = mul_33 = None + x_381 = x_380 + addcmul_33 + x_380 = addcmul_33 = None + x_382 = torch._C._nn.linear( + x_381, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_381 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_383 = torch.nn.functional.dropout(x_382, 0.0, False, False) + x_382 = None + x_384 = x_383.permute(0, 3, 1, 2) + x_383 = None + x_385 = x_384 + input_4 + x_384 = input_4 = None + x_386 = torch.conv2d( + x_385, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_387 = x_386.permute(0, 2, 3, 1) + x_386 = None + x_388 = torch.nn.functional.layer_norm( + x_387, + (1536,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_387 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_389 = torch._C._nn.linear( + x_388, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_388 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_390 = torch._C._nn.gelu(x_389) + x_389 = None + x_391 = torch.nn.functional.dropout(x_390, 0.0, False, False) + x_390 = None + x_g_34 = x_391.norm(p=2, dim=(1, 2), keepdim=True) + mean_34 = x_g_34.mean(dim=-1, keepdim=True) + add_102 = mean_34 + 1e-06 + mean_34 = None + x_n_34 = x_g_34 / add_102 + x_g_34 = add_102 = None + view_68 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_69 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_34 = x_391 * x_n_34 + x_n_34 = None + addcmul_34 = torch.addcmul(view_68, view_69, mul_34) + view_68 = view_69 = mul_34 = None + x_392 = x_391 + addcmul_34 + x_391 = addcmul_34 = None + x_393 = torch._C._nn.linear( + x_392, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_392 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_394 = torch.nn.functional.dropout(x_393, 0.0, False, False) + x_393 = None + x_395 = x_394.permute(0, 3, 1, 2) + x_394 = None + x_396 = x_395 + x_385 + x_395 = x_385 = None + x_397 = torch.conv2d( + x_396, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_398 = x_397.permute(0, 2, 3, 1) + x_397 = None + x_399 = torch.nn.functional.layer_norm( + x_398, + (1536,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_398 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_400 = torch._C._nn.linear( + x_399, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_399 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_401 = torch._C._nn.gelu(x_400) + x_400 = None + x_402 = torch.nn.functional.dropout(x_401, 0.0, False, False) + x_401 = None + x_g_35 = x_402.norm(p=2, dim=(1, 2), keepdim=True) + mean_35 = x_g_35.mean(dim=-1, keepdim=True) + add_105 = mean_35 + 1e-06 + mean_35 = None + x_n_35 = x_g_35 / add_105 + x_g_35 = add_105 = None + view_70 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_71 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_35 = x_402 * x_n_35 + x_n_35 = None + addcmul_35 = torch.addcmul(view_70, view_71, mul_35) + view_70 = view_71 = mul_35 = None + x_403 = x_402 + addcmul_35 + x_402 = addcmul_35 = None + x_404 = torch._C._nn.linear( + x_403, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_403 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_405 = torch.nn.functional.dropout(x_404, 0.0, False, False) + x_404 = None + x_406 = x_405.permute(0, 3, 1, 2) + x_405 = None + x_407 = x_406 + x_396 + x_406 = x_396 = None + x_408 = torch.nn.functional.adaptive_avg_pool2d(x_407, 1) + x_407 = None + x_409 = x_408.permute(0, 2, 3, 1) + x_408 = None + x_410 = torch.nn.functional.layer_norm( + x_409, + (1536,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_409 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_411 = x_410.permute(0, 3, 1, 2) + x_410 = None + x_412 = x_411.flatten(1, -1) + x_411 = None + x_413 = torch.nn.functional.dropout(x_412, 0.0, False, False) + x_412 = None + return (x_413,) diff --git a/samples/timm/convnextv2_large.fcmae/weight_meta.py b/samples/timm/convnextv2_large.fcmae/weight_meta.py new file mode 100644 index 0000000000..c743c91eaa --- /dev/null +++ b/samples/timm/convnextv2_large.fcmae/weight_meta.py @@ -0,0 +1,3810 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [192, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.223 + std = 1.288 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [384, 192, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [768, 384, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [1536, 768, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_large.fcmae_ft_in1k/graph_hash.txt b/samples/timm/convnextv2_large.fcmae_ft_in1k/graph_hash.txt new file mode 100644 index 0000000000..ca248166d6 --- /dev/null +++ b/samples/timm/convnextv2_large.fcmae_ft_in1k/graph_hash.txt @@ -0,0 +1 @@ +a959225c800ea702edb90763031e8d806ff70463f142397e37c3ad05ece4f96d \ No newline at end of file diff --git a/samples/timm/convnextv2_large.fcmae_ft_in1k/graph_net.json b/samples/timm/convnextv2_large.fcmae_ft_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnextv2_large.fcmae_ft_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_large.fcmae_ft_in1k/input_meta.py b/samples/timm/convnextv2_large.fcmae_ft_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_large.fcmae_ft_in1k/input_tensor_constraints.py b/samples/timm/convnextv2_large.fcmae_ft_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_large.fcmae_ft_in1k/model.py b/samples/timm/convnextv2_large.fcmae_ft_in1k/model.py new file mode 100644 index 0000000000..d78965bafd --- /dev/null +++ b/samples/timm/convnextv2_large.fcmae_ft_in1k/model.py @@ -0,0 +1,3283 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (192,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (192,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_g = x_8.norm(p=2, dim=(1, 2), keepdim=True) + mean = x_g.mean(dim=-1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_8 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_9 = x_8 + addcmul + x_8 = addcmul = None + x_10 = torch._C._nn.linear( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = x_11.permute(0, 3, 1, 2) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (192,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_g_1 = x_19.norm(p=2, dim=(1, 2), keepdim=True) + mean_1 = x_g_1.mean(dim=-1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_19 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_20 = x_19 + addcmul_1 + x_19 = addcmul_1 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (192,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_g_2 = x_30.norm(p=2, dim=(1, 2), keepdim=True) + mean_2 = x_g_2.mean(dim=-1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_30 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_31 = x_30 + addcmul_2 + x_30 = addcmul_2 = None + x_32 = torch._C._nn.linear( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + x_34 = x_33.permute(0, 3, 1, 2) + x_33 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (192,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (384,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_g_3 = x_44.norm(p=2, dim=(1, 2), keepdim=True) + mean_3 = x_g_3.mean(dim=-1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_44 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_45 = x_44 + addcmul_3 + x_44 = addcmul_3 = None + x_46 = torch._C._nn.linear( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (384,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_g_4 = x_55.norm(p=2, dim=(1, 2), keepdim=True) + mean_4 = x_g_4.mean(dim=-1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_55 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_56 = x_55 + addcmul_4 + x_55 = addcmul_4 = None + x_57 = torch._C._nn.linear( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_58 = torch.nn.functional.dropout(x_57, 0.0, False, False) + x_57 = None + x_59 = x_58.permute(0, 3, 1, 2) + x_58 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (384,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_g_5 = x_66.norm(p=2, dim=(1, 2), keepdim=True) + mean_5 = x_g_5.mean(dim=-1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_66 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_67 = x_66 + addcmul_5 + x_66 = addcmul_5 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = x_69.permute(0, 3, 1, 2) + x_69 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (384,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_g_6 = x_80.norm(p=2, dim=(1, 2), keepdim=True) + mean_6 = x_g_6.mean(dim=-1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_80 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_81 = x_80 + addcmul_6 + x_80 = addcmul_6 = None + x_82 = torch._C._nn.linear( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + x_84 = x_83.permute(0, 3, 1, 2) + x_83 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_g_7 = x_91.norm(p=2, dim=(1, 2), keepdim=True) + mean_7 = x_g_7.mean(dim=-1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_91 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_92 = x_91 + addcmul_7 + x_91 = addcmul_7 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_g_8 = x_102.norm(p=2, dim=(1, 2), keepdim=True) + mean_8 = x_g_8.mean(dim=-1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_102 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_103 = x_102 + addcmul_8 + x_102 = addcmul_8 = None + x_104 = torch._C._nn.linear( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = x_105.permute(0, 3, 1, 2) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_g_9 = x_113.norm(p=2, dim=(1, 2), keepdim=True) + mean_9 = x_g_9.mean(dim=-1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_113 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_114 = x_113 + addcmul_9 + x_113 = addcmul_9 = None + x_115 = torch._C._nn.linear( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + x_117 = x_116.permute(0, 3, 1, 2) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_g_10 = x_124.norm(p=2, dim=(1, 2), keepdim=True) + mean_10 = x_g_10.mean(dim=-1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_124 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_125 = x_124 + addcmul_10 + x_124 = addcmul_10 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = x_127.permute(0, 3, 1, 2) + x_127 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_g_11 = x_135.norm(p=2, dim=(1, 2), keepdim=True) + mean_11 = x_g_11.mean(dim=-1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_135 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_136 = x_135 + addcmul_11 + x_135 = addcmul_11 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_138.permute(0, 3, 1, 2) + x_138 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_g_12 = x_146.norm(p=2, dim=(1, 2), keepdim=True) + mean_12 = x_g_12.mean(dim=-1, keepdim=True) + add_36 = mean_12 + 1e-06 + mean_12 = None + x_n_12 = x_g_12 / add_36 + x_g_12 = add_36 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_12 = x_146 * x_n_12 + x_n_12 = None + addcmul_12 = torch.addcmul(view_24, view_25, mul_12) + view_24 = view_25 = mul_12 = None + x_147 = x_146 + addcmul_12 + x_146 = addcmul_12 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_149.permute(0, 3, 1, 2) + x_149 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_g_13 = x_157.norm(p=2, dim=(1, 2), keepdim=True) + mean_13 = x_g_13.mean(dim=-1, keepdim=True) + add_39 = mean_13 + 1e-06 + mean_13 = None + x_n_13 = x_g_13 / add_39 + x_g_13 = add_39 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_13 = x_157 * x_n_13 + x_n_13 = None + addcmul_13 = torch.addcmul(view_26, view_27, mul_13) + view_26 = view_27 = mul_13 = None + x_158 = x_157 + addcmul_13 + x_157 = addcmul_13 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_160.permute(0, 3, 1, 2) + x_160 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_g_14 = x_168.norm(p=2, dim=(1, 2), keepdim=True) + mean_14 = x_g_14.mean(dim=-1, keepdim=True) + add_42 = mean_14 + 1e-06 + mean_14 = None + x_n_14 = x_g_14 / add_42 + x_g_14 = add_42 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_14 = x_168 * x_n_14 + x_n_14 = None + addcmul_14 = torch.addcmul(view_28, view_29, mul_14) + view_28 = view_29 = mul_14 = None + x_169 = x_168 + addcmul_14 + x_168 = addcmul_14 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_171.permute(0, 3, 1, 2) + x_171 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_ = (None) + x_175 = x_174.permute(0, 2, 3, 1) + x_174 = None + x_176 = torch.nn.functional.layer_norm( + x_175, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_, + 1e-06, + ) + x_175 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_ = (None) + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_176 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_178 = torch._C._nn.gelu(x_177) + x_177 = None + x_179 = torch.nn.functional.dropout(x_178, 0.0, False, False) + x_178 = None + x_g_15 = x_179.norm(p=2, dim=(1, 2), keepdim=True) + mean_15 = x_g_15.mean(dim=-1, keepdim=True) + add_45 = mean_15 + 1e-06 + mean_15 = None + x_n_15 = x_g_15 / add_45 + x_g_15 = add_45 = None + view_30 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_31 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_15 = x_179 * x_n_15 + x_n_15 = None + addcmul_15 = torch.addcmul(view_30, view_31, mul_15) + view_30 = view_31 = mul_15 = None + x_180 = x_179 + addcmul_15 + x_179 = addcmul_15 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_182.permute(0, 3, 1, 2) + x_182 = None + x_184 = x_183 + x_173 + x_183 = x_173 = None + x_185 = torch.conv2d( + x_184, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_ = (None) + x_186 = x_185.permute(0, 2, 3, 1) + x_185 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_, + 1e-06, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_ = (None) + x_188 = torch._C._nn.linear( + x_187, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_187 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_189 = torch._C._nn.gelu(x_188) + x_188 = None + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_g_16 = x_190.norm(p=2, dim=(1, 2), keepdim=True) + mean_16 = x_g_16.mean(dim=-1, keepdim=True) + add_48 = mean_16 + 1e-06 + mean_16 = None + x_n_16 = x_g_16 / add_48 + x_g_16 = add_48 = None + view_32 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_33 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_16 = x_190 * x_n_16 + x_n_16 = None + addcmul_16 = torch.addcmul(view_32, view_33, mul_16) + view_32 = view_33 = mul_16 = None + x_191 = x_190 + addcmul_16 + x_190 = addcmul_16 = None + x_192 = torch._C._nn.linear( + x_191, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_191 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_193.permute(0, 3, 1, 2) + x_193 = None + x_195 = x_194 + x_184 + x_194 = x_184 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_ = (None) + x_197 = x_196.permute(0, 2, 3, 1) + x_196 = None + x_198 = torch.nn.functional.layer_norm( + x_197, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_, + 1e-06, + ) + x_197 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_ = (None) + x_199 = torch._C._nn.linear( + x_198, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_198 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_200 = torch._C._nn.gelu(x_199) + x_199 = None + x_201 = torch.nn.functional.dropout(x_200, 0.0, False, False) + x_200 = None + x_g_17 = x_201.norm(p=2, dim=(1, 2), keepdim=True) + mean_17 = x_g_17.mean(dim=-1, keepdim=True) + add_51 = mean_17 + 1e-06 + mean_17 = None + x_n_17 = x_g_17 / add_51 + x_g_17 = add_51 = None + view_34 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_35 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_17 = x_201 * x_n_17 + x_n_17 = None + addcmul_17 = torch.addcmul(view_34, view_35, mul_17) + view_34 = view_35 = mul_17 = None + x_202 = x_201 + addcmul_17 + x_201 = addcmul_17 = None + x_203 = torch._C._nn.linear( + x_202, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_202 = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_205 = x_204.permute(0, 3, 1, 2) + x_204 = None + x_206 = x_205 + x_195 + x_205 = x_195 = None + x_207 = torch.conv2d( + x_206, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_ = (None) + x_208 = x_207.permute(0, 2, 3, 1) + x_207 = None + x_209 = torch.nn.functional.layer_norm( + x_208, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_, + 1e-06, + ) + x_208 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_ = (None) + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_209 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_211 = torch._C._nn.gelu(x_210) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_g_18 = x_212.norm(p=2, dim=(1, 2), keepdim=True) + mean_18 = x_g_18.mean(dim=-1, keepdim=True) + add_54 = mean_18 + 1e-06 + mean_18 = None + x_n_18 = x_g_18 / add_54 + x_g_18 = add_54 = None + view_36 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_37 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_18 = x_212 * x_n_18 + x_n_18 = None + addcmul_18 = torch.addcmul(view_36, view_37, mul_18) + view_36 = view_37 = mul_18 = None + x_213 = x_212 + addcmul_18 + x_212 = addcmul_18 = None + x_214 = torch._C._nn.linear( + x_213, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_213 = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = x_215.permute(0, 3, 1, 2) + x_215 = None + x_217 = x_216 + x_206 + x_216 = x_206 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_ = (None) + x_219 = x_218.permute(0, 2, 3, 1) + x_218 = None + x_220 = torch.nn.functional.layer_norm( + x_219, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_, + 1e-06, + ) + x_219 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_ = (None) + x_221 = torch._C._nn.linear( + x_220, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_220 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_222 = torch._C._nn.gelu(x_221) + x_221 = None + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_g_19 = x_223.norm(p=2, dim=(1, 2), keepdim=True) + mean_19 = x_g_19.mean(dim=-1, keepdim=True) + add_57 = mean_19 + 1e-06 + mean_19 = None + x_n_19 = x_g_19 / add_57 + x_g_19 = add_57 = None + view_38 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_39 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_19 = x_223 * x_n_19 + x_n_19 = None + addcmul_19 = torch.addcmul(view_38, view_39, mul_19) + view_38 = view_39 = mul_19 = None + x_224 = x_223 + addcmul_19 + x_223 = addcmul_19 = None + x_225 = torch._C._nn.linear( + x_224, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_224 = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_226 = torch.nn.functional.dropout(x_225, 0.0, False, False) + x_225 = None + x_227 = x_226.permute(0, 3, 1, 2) + x_226 = None + x_228 = x_227 + x_217 + x_227 = x_217 = None + x_229 = torch.conv2d( + x_228, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_ = (None) + x_230 = x_229.permute(0, 2, 3, 1) + x_229 = None + x_231 = torch.nn.functional.layer_norm( + x_230, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_, + 1e-06, + ) + x_230 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_ = (None) + x_232 = torch._C._nn.linear( + x_231, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_231 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_233 = torch._C._nn.gelu(x_232) + x_232 = None + x_234 = torch.nn.functional.dropout(x_233, 0.0, False, False) + x_233 = None + x_g_20 = x_234.norm(p=2, dim=(1, 2), keepdim=True) + mean_20 = x_g_20.mean(dim=-1, keepdim=True) + add_60 = mean_20 + 1e-06 + mean_20 = None + x_n_20 = x_g_20 / add_60 + x_g_20 = add_60 = None + view_40 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_41 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_20 = x_234 * x_n_20 + x_n_20 = None + addcmul_20 = torch.addcmul(view_40, view_41, mul_20) + view_40 = view_41 = mul_20 = None + x_235 = x_234 + addcmul_20 + x_234 = addcmul_20 = None + x_236 = torch._C._nn.linear( + x_235, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_235 = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_237 = torch.nn.functional.dropout(x_236, 0.0, False, False) + x_236 = None + x_238 = x_237.permute(0, 3, 1, 2) + x_237 = None + x_239 = x_238 + x_228 + x_238 = x_228 = None + x_240 = torch.conv2d( + x_239, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_ = (None) + x_241 = x_240.permute(0, 2, 3, 1) + x_240 = None + x_242 = torch.nn.functional.layer_norm( + x_241, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_, + 1e-06, + ) + x_241 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_ = (None) + x_243 = torch._C._nn.linear( + x_242, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_242 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_244 = torch._C._nn.gelu(x_243) + x_243 = None + x_245 = torch.nn.functional.dropout(x_244, 0.0, False, False) + x_244 = None + x_g_21 = x_245.norm(p=2, dim=(1, 2), keepdim=True) + mean_21 = x_g_21.mean(dim=-1, keepdim=True) + add_63 = mean_21 + 1e-06 + mean_21 = None + x_n_21 = x_g_21 / add_63 + x_g_21 = add_63 = None + view_42 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_43 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_21 = x_245 * x_n_21 + x_n_21 = None + addcmul_21 = torch.addcmul(view_42, view_43, mul_21) + view_42 = view_43 = mul_21 = None + x_246 = x_245 + addcmul_21 + x_245 = addcmul_21 = None + x_247 = torch._C._nn.linear( + x_246, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_246 = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_248 = torch.nn.functional.dropout(x_247, 0.0, False, False) + x_247 = None + x_249 = x_248.permute(0, 3, 1, 2) + x_248 = None + x_250 = x_249 + x_239 + x_249 = x_239 = None + x_251 = torch.conv2d( + x_250, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_ = (None) + x_252 = x_251.permute(0, 2, 3, 1) + x_251 = None + x_253 = torch.nn.functional.layer_norm( + x_252, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_, + 1e-06, + ) + x_252 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_ = (None) + x_254 = torch._C._nn.linear( + x_253, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_253 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_255 = torch._C._nn.gelu(x_254) + x_254 = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_g_22 = x_256.norm(p=2, dim=(1, 2), keepdim=True) + mean_22 = x_g_22.mean(dim=-1, keepdim=True) + add_66 = mean_22 + 1e-06 + mean_22 = None + x_n_22 = x_g_22 / add_66 + x_g_22 = add_66 = None + view_44 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_45 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_22 = x_256 * x_n_22 + x_n_22 = None + addcmul_22 = torch.addcmul(view_44, view_45, mul_22) + view_44 = view_45 = mul_22 = None + x_257 = x_256 + addcmul_22 + x_256 = addcmul_22 = None + x_258 = torch._C._nn.linear( + x_257, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_257 = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_259 = torch.nn.functional.dropout(x_258, 0.0, False, False) + x_258 = None + x_260 = x_259.permute(0, 3, 1, 2) + x_259 = None + x_261 = x_260 + x_250 + x_260 = x_250 = None + x_262 = torch.conv2d( + x_261, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_ = (None) + x_263 = x_262.permute(0, 2, 3, 1) + x_262 = None + x_264 = torch.nn.functional.layer_norm( + x_263, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_, + 1e-06, + ) + x_263 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_ = (None) + x_265 = torch._C._nn.linear( + x_264, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_264 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_266 = torch._C._nn.gelu(x_265) + x_265 = None + x_267 = torch.nn.functional.dropout(x_266, 0.0, False, False) + x_266 = None + x_g_23 = x_267.norm(p=2, dim=(1, 2), keepdim=True) + mean_23 = x_g_23.mean(dim=-1, keepdim=True) + add_69 = mean_23 + 1e-06 + mean_23 = None + x_n_23 = x_g_23 / add_69 + x_g_23 = add_69 = None + view_46 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_47 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_23 = x_267 * x_n_23 + x_n_23 = None + addcmul_23 = torch.addcmul(view_46, view_47, mul_23) + view_46 = view_47 = mul_23 = None + x_268 = x_267 + addcmul_23 + x_267 = addcmul_23 = None + x_269 = torch._C._nn.linear( + x_268, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_268 = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_270 = torch.nn.functional.dropout(x_269, 0.0, False, False) + x_269 = None + x_271 = x_270.permute(0, 3, 1, 2) + x_270 = None + x_272 = x_271 + x_261 + x_271 = x_261 = None + x_273 = torch.conv2d( + x_272, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_ = (None) + x_274 = x_273.permute(0, 2, 3, 1) + x_273 = None + x_275 = torch.nn.functional.layer_norm( + x_274, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_, + 1e-06, + ) + x_274 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_ = (None) + x_276 = torch._C._nn.linear( + x_275, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_275 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_277 = torch._C._nn.gelu(x_276) + x_276 = None + x_278 = torch.nn.functional.dropout(x_277, 0.0, False, False) + x_277 = None + x_g_24 = x_278.norm(p=2, dim=(1, 2), keepdim=True) + mean_24 = x_g_24.mean(dim=-1, keepdim=True) + add_72 = mean_24 + 1e-06 + mean_24 = None + x_n_24 = x_g_24 / add_72 + x_g_24 = add_72 = None + view_48 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_49 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_24 = x_278 * x_n_24 + x_n_24 = None + addcmul_24 = torch.addcmul(view_48, view_49, mul_24) + view_48 = view_49 = mul_24 = None + x_279 = x_278 + addcmul_24 + x_278 = addcmul_24 = None + x_280 = torch._C._nn.linear( + x_279, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_279 = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_281 = torch.nn.functional.dropout(x_280, 0.0, False, False) + x_280 = None + x_282 = x_281.permute(0, 3, 1, 2) + x_281 = None + x_283 = x_282 + x_272 + x_282 = x_272 = None + x_284 = torch.conv2d( + x_283, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_ = (None) + x_285 = x_284.permute(0, 2, 3, 1) + x_284 = None + x_286 = torch.nn.functional.layer_norm( + x_285, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_, + 1e-06, + ) + x_285 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_ = (None) + x_287 = torch._C._nn.linear( + x_286, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_286 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_288 = torch._C._nn.gelu(x_287) + x_287 = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + x_g_25 = x_289.norm(p=2, dim=(1, 2), keepdim=True) + mean_25 = x_g_25.mean(dim=-1, keepdim=True) + add_75 = mean_25 + 1e-06 + mean_25 = None + x_n_25 = x_g_25 / add_75 + x_g_25 = add_75 = None + view_50 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_51 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_25 = x_289 * x_n_25 + x_n_25 = None + addcmul_25 = torch.addcmul(view_50, view_51, mul_25) + view_50 = view_51 = mul_25 = None + x_290 = x_289 + addcmul_25 + x_289 = addcmul_25 = None + x_291 = torch._C._nn.linear( + x_290, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_290 = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_292 = torch.nn.functional.dropout(x_291, 0.0, False, False) + x_291 = None + x_293 = x_292.permute(0, 3, 1, 2) + x_292 = None + x_294 = x_293 + x_283 + x_293 = x_283 = None + x_295 = torch.conv2d( + x_294, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_ = (None) + x_296 = x_295.permute(0, 2, 3, 1) + x_295 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_, + 1e-06, + ) + x_296 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_ = (None) + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_299 = torch._C._nn.gelu(x_298) + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_g_26 = x_300.norm(p=2, dim=(1, 2), keepdim=True) + mean_26 = x_g_26.mean(dim=-1, keepdim=True) + add_78 = mean_26 + 1e-06 + mean_26 = None + x_n_26 = x_g_26 / add_78 + x_g_26 = add_78 = None + view_52 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_53 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_26 = x_300 * x_n_26 + x_n_26 = None + addcmul_26 = torch.addcmul(view_52, view_53, mul_26) + view_52 = view_53 = mul_26 = None + x_301 = x_300 + addcmul_26 + x_300 = addcmul_26 = None + x_302 = torch._C._nn.linear( + x_301, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_301 = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_303 = torch.nn.functional.dropout(x_302, 0.0, False, False) + x_302 = None + x_304 = x_303.permute(0, 3, 1, 2) + x_303 = None + x_305 = x_304 + x_294 + x_304 = x_294 = None + x_306 = torch.conv2d( + x_305, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_ = (None) + x_307 = x_306.permute(0, 2, 3, 1) + x_306 = None + x_308 = torch.nn.functional.layer_norm( + x_307, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_, + 1e-06, + ) + x_307 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_ = (None) + x_309 = torch._C._nn.linear( + x_308, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_308 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_310 = torch._C._nn.gelu(x_309) + x_309 = None + x_311 = torch.nn.functional.dropout(x_310, 0.0, False, False) + x_310 = None + x_g_27 = x_311.norm(p=2, dim=(1, 2), keepdim=True) + mean_27 = x_g_27.mean(dim=-1, keepdim=True) + add_81 = mean_27 + 1e-06 + mean_27 = None + x_n_27 = x_g_27 / add_81 + x_g_27 = add_81 = None + view_54 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_55 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_27 = x_311 * x_n_27 + x_n_27 = None + addcmul_27 = torch.addcmul(view_54, view_55, mul_27) + view_54 = view_55 = mul_27 = None + x_312 = x_311 + addcmul_27 + x_311 = addcmul_27 = None + x_313 = torch._C._nn.linear( + x_312, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_312 = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_314 = torch.nn.functional.dropout(x_313, 0.0, False, False) + x_313 = None + x_315 = x_314.permute(0, 3, 1, 2) + x_314 = None + x_316 = x_315 + x_305 + x_315 = x_305 = None + x_317 = torch.conv2d( + x_316, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_ = (None) + x_318 = x_317.permute(0, 2, 3, 1) + x_317 = None + x_319 = torch.nn.functional.layer_norm( + x_318, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_, + 1e-06, + ) + x_318 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_ = (None) + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_319 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_321 = torch._C._nn.gelu(x_320) + x_320 = None + x_322 = torch.nn.functional.dropout(x_321, 0.0, False, False) + x_321 = None + x_g_28 = x_322.norm(p=2, dim=(1, 2), keepdim=True) + mean_28 = x_g_28.mean(dim=-1, keepdim=True) + add_84 = mean_28 + 1e-06 + mean_28 = None + x_n_28 = x_g_28 / add_84 + x_g_28 = add_84 = None + view_56 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_57 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_28 = x_322 * x_n_28 + x_n_28 = None + addcmul_28 = torch.addcmul(view_56, view_57, mul_28) + view_56 = view_57 = mul_28 = None + x_323 = x_322 + addcmul_28 + x_322 = addcmul_28 = None + x_324 = torch._C._nn.linear( + x_323, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_323 = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_325 = torch.nn.functional.dropout(x_324, 0.0, False, False) + x_324 = None + x_326 = x_325.permute(0, 3, 1, 2) + x_325 = None + x_327 = x_326 + x_316 + x_326 = x_316 = None + x_328 = torch.conv2d( + x_327, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_ = (None) + x_329 = x_328.permute(0, 2, 3, 1) + x_328 = None + x_330 = torch.nn.functional.layer_norm( + x_329, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_, + 1e-06, + ) + x_329 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_ = (None) + x_331 = torch._C._nn.linear( + x_330, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_330 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_332 = torch._C._nn.gelu(x_331) + x_331 = None + x_333 = torch.nn.functional.dropout(x_332, 0.0, False, False) + x_332 = None + x_g_29 = x_333.norm(p=2, dim=(1, 2), keepdim=True) + mean_29 = x_g_29.mean(dim=-1, keepdim=True) + add_87 = mean_29 + 1e-06 + mean_29 = None + x_n_29 = x_g_29 / add_87 + x_g_29 = add_87 = None + view_58 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_59 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_29 = x_333 * x_n_29 + x_n_29 = None + addcmul_29 = torch.addcmul(view_58, view_59, mul_29) + view_58 = view_59 = mul_29 = None + x_334 = x_333 + addcmul_29 + x_333 = addcmul_29 = None + x_335 = torch._C._nn.linear( + x_334, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_334 = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_336 = torch.nn.functional.dropout(x_335, 0.0, False, False) + x_335 = None + x_337 = x_336.permute(0, 3, 1, 2) + x_336 = None + x_338 = x_337 + x_327 + x_337 = x_327 = None + x_339 = torch.conv2d( + x_338, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_ = (None) + x_340 = x_339.permute(0, 2, 3, 1) + x_339 = None + x_341 = torch.nn.functional.layer_norm( + x_340, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_, + 1e-06, + ) + x_340 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_ = (None) + x_342 = torch._C._nn.linear( + x_341, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + x_341 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_343 = torch._C._nn.gelu(x_342) + x_342 = None + x_344 = torch.nn.functional.dropout(x_343, 0.0, False, False) + x_343 = None + x_g_30 = x_344.norm(p=2, dim=(1, 2), keepdim=True) + mean_30 = x_g_30.mean(dim=-1, keepdim=True) + add_90 = mean_30 + 1e-06 + mean_30 = None + x_n_30 = x_g_30 / add_90 + x_g_30 = add_90 = None + view_60 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_61 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_30 = x_344 * x_n_30 + x_n_30 = None + addcmul_30 = torch.addcmul(view_60, view_61, mul_30) + view_60 = view_61 = mul_30 = None + x_345 = x_344 + addcmul_30 + x_344 = addcmul_30 = None + x_346 = torch._C._nn.linear( + x_345, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_345 = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_347 = torch.nn.functional.dropout(x_346, 0.0, False, False) + x_346 = None + x_348 = x_347.permute(0, 3, 1, 2) + x_347 = None + x_349 = x_348 + x_338 + x_348 = x_338 = None + x_350 = torch.conv2d( + x_349, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_ = (None) + x_351 = x_350.permute(0, 2, 3, 1) + x_350 = None + x_352 = torch.nn.functional.layer_norm( + x_351, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_, + 1e-06, + ) + x_351 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_ = (None) + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + x_352 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_354 = torch._C._nn.gelu(x_353) + x_353 = None + x_355 = torch.nn.functional.dropout(x_354, 0.0, False, False) + x_354 = None + x_g_31 = x_355.norm(p=2, dim=(1, 2), keepdim=True) + mean_31 = x_g_31.mean(dim=-1, keepdim=True) + add_93 = mean_31 + 1e-06 + mean_31 = None + x_n_31 = x_g_31 / add_93 + x_g_31 = add_93 = None + view_62 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_63 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_31 = x_355 * x_n_31 + x_n_31 = None + addcmul_31 = torch.addcmul(view_62, view_63, mul_31) + view_62 = view_63 = mul_31 = None + x_356 = x_355 + addcmul_31 + x_355 = addcmul_31 = None + x_357 = torch._C._nn.linear( + x_356, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_356 = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_358 = torch.nn.functional.dropout(x_357, 0.0, False, False) + x_357 = None + x_359 = x_358.permute(0, 3, 1, 2) + x_358 = None + x_360 = x_359 + x_349 + x_359 = x_349 = None + x_361 = torch.conv2d( + x_360, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_ = (None) + x_362 = x_361.permute(0, 2, 3, 1) + x_361 = None + x_363 = torch.nn.functional.layer_norm( + x_362, + (768,), + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_, + 1e-06, + ) + x_362 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_ = (None) + x_364 = torch._C._nn.linear( + x_363, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + x_363 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_365 = torch._C._nn.gelu(x_364) + x_364 = None + x_366 = torch.nn.functional.dropout(x_365, 0.0, False, False) + x_365 = None + x_g_32 = x_366.norm(p=2, dim=(1, 2), keepdim=True) + mean_32 = x_g_32.mean(dim=-1, keepdim=True) + add_96 = mean_32 + 1e-06 + mean_32 = None + x_n_32 = x_g_32 / add_96 + x_g_32 = add_96 = None + view_64 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_65 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_32 = x_366 * x_n_32 + x_n_32 = None + addcmul_32 = torch.addcmul(view_64, view_65, mul_32) + view_64 = view_65 = mul_32 = None + x_367 = x_366 + addcmul_32 + x_366 = addcmul_32 = None + x_368 = torch._C._nn.linear( + x_367, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_367 = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_369 = torch.nn.functional.dropout(x_368, 0.0, False, False) + x_368 = None + x_370 = x_369.permute(0, 3, 1, 2) + x_369 = None + x_371 = x_370 + x_360 + x_370 = x_360 = None + x_372 = x_371.permute(0, 2, 3, 1) + x_371 = None + x_373 = torch.nn.functional.layer_norm( + x_372, + (768,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_372 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_374 = x_373.permute(0, 3, 1, 2) + x_373 = None + input_4 = torch.conv2d( + x_374, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_374 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_375 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_376 = x_375.permute(0, 2, 3, 1) + x_375 = None + x_377 = torch.nn.functional.layer_norm( + x_376, + (1536,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_376 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_378 = torch._C._nn.linear( + x_377, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_377 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_379 = torch._C._nn.gelu(x_378) + x_378 = None + x_380 = torch.nn.functional.dropout(x_379, 0.0, False, False) + x_379 = None + x_g_33 = x_380.norm(p=2, dim=(1, 2), keepdim=True) + mean_33 = x_g_33.mean(dim=-1, keepdim=True) + add_99 = mean_33 + 1e-06 + mean_33 = None + x_n_33 = x_g_33 / add_99 + x_g_33 = add_99 = None + view_66 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_67 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_33 = x_380 * x_n_33 + x_n_33 = None + addcmul_33 = torch.addcmul(view_66, view_67, mul_33) + view_66 = view_67 = mul_33 = None + x_381 = x_380 + addcmul_33 + x_380 = addcmul_33 = None + x_382 = torch._C._nn.linear( + x_381, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_381 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_383 = torch.nn.functional.dropout(x_382, 0.0, False, False) + x_382 = None + x_384 = x_383.permute(0, 3, 1, 2) + x_383 = None + x_385 = x_384 + input_4 + x_384 = input_4 = None + x_386 = torch.conv2d( + x_385, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_387 = x_386.permute(0, 2, 3, 1) + x_386 = None + x_388 = torch.nn.functional.layer_norm( + x_387, + (1536,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_387 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_389 = torch._C._nn.linear( + x_388, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_388 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_390 = torch._C._nn.gelu(x_389) + x_389 = None + x_391 = torch.nn.functional.dropout(x_390, 0.0, False, False) + x_390 = None + x_g_34 = x_391.norm(p=2, dim=(1, 2), keepdim=True) + mean_34 = x_g_34.mean(dim=-1, keepdim=True) + add_102 = mean_34 + 1e-06 + mean_34 = None + x_n_34 = x_g_34 / add_102 + x_g_34 = add_102 = None + view_68 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_69 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_34 = x_391 * x_n_34 + x_n_34 = None + addcmul_34 = torch.addcmul(view_68, view_69, mul_34) + view_68 = view_69 = mul_34 = None + x_392 = x_391 + addcmul_34 + x_391 = addcmul_34 = None + x_393 = torch._C._nn.linear( + x_392, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_392 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_394 = torch.nn.functional.dropout(x_393, 0.0, False, False) + x_393 = None + x_395 = x_394.permute(0, 3, 1, 2) + x_394 = None + x_396 = x_395 + x_385 + x_395 = x_385 = None + x_397 = torch.conv2d( + x_396, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 1536, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_398 = x_397.permute(0, 2, 3, 1) + x_397 = None + x_399 = torch.nn.functional.layer_norm( + x_398, + (1536,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_398 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_400 = torch._C._nn.linear( + x_399, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_399 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_401 = torch._C._nn.gelu(x_400) + x_400 = None + x_402 = torch.nn.functional.dropout(x_401, 0.0, False, False) + x_401 = None + x_g_35 = x_402.norm(p=2, dim=(1, 2), keepdim=True) + mean_35 = x_g_35.mean(dim=-1, keepdim=True) + add_105 = mean_35 + 1e-06 + mean_35 = None + x_n_35 = x_g_35 / add_105 + x_g_35 = add_105 = None + view_70 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_71 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_35 = x_402 * x_n_35 + x_n_35 = None + addcmul_35 = torch.addcmul(view_70, view_71, mul_35) + view_70 = view_71 = mul_35 = None + x_403 = x_402 + addcmul_35 + x_402 = addcmul_35 = None + x_404 = torch._C._nn.linear( + x_403, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_403 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_405 = torch.nn.functional.dropout(x_404, 0.0, False, False) + x_404 = None + x_406 = x_405.permute(0, 3, 1, 2) + x_405 = None + x_407 = x_406 + x_396 + x_406 = x_396 = None + x_408 = torch.nn.functional.adaptive_avg_pool2d(x_407, 1) + x_407 = None + x_409 = x_408.permute(0, 2, 3, 1) + x_408 = None + x_410 = torch.nn.functional.layer_norm( + x_409, + (1536,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_409 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_411 = x_410.permute(0, 3, 1, 2) + x_410 = None + x_412 = x_411.flatten(1, -1) + x_411 = None + x_413 = torch.nn.functional.dropout(x_412, 0.0, False, False) + x_412 = None + x_414 = torch._C._nn.linear( + x_413, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_413 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_414,) diff --git a/samples/timm/convnextv2_large.fcmae_ft_in1k/weight_meta.py b/samples/timm/convnextv2_large.fcmae_ft_in1k/weight_meta.py new file mode 100644 index 0000000000..cedfe37242 --- /dev/null +++ b/samples/timm/convnextv2_large.fcmae_ft_in1k/weight_meta.py @@ -0,0 +1,3830 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [192, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.232 + std = 1.288 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [384, 192, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [768, 384, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [1536, 768, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [1536, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [6144, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1536, 6144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_nano.fcmae/graph_hash.txt b/samples/timm/convnextv2_nano.fcmae/graph_hash.txt new file mode 100644 index 0000000000..acc1eacb11 --- /dev/null +++ b/samples/timm/convnextv2_nano.fcmae/graph_hash.txt @@ -0,0 +1 @@ +914c586df7bf0728ff0403e9e64ba9ace425ac173539baa29c2289106bd509d9 \ No newline at end of file diff --git a/samples/timm/convnextv2_nano.fcmae/graph_net.json b/samples/timm/convnextv2_nano.fcmae/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnextv2_nano.fcmae/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_nano.fcmae/input_meta.py b/samples/timm/convnextv2_nano.fcmae/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_nano.fcmae/input_tensor_constraints.py b/samples/timm/convnextv2_nano.fcmae/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_nano.fcmae/model.py b/samples/timm/convnextv2_nano.fcmae/model.py new file mode 100644 index 0000000000..2f84a1f383 --- /dev/null +++ b/samples/timm/convnextv2_nano.fcmae/model.py @@ -0,0 +1,1487 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (80,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (80,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_g = x_9.norm(p=2, dim=(2, 3), keepdim=True) + mean = x_g.mean(dim=1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_9 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_10 = x_9 + addcmul + x_9 = addcmul = None + x_11 = torch.conv2d( + x_10, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_10 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (80,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_g_1 = x_20.norm(p=2, dim=(2, 3), keepdim=True) + mean_1 = x_g_1.mean(dim=1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_20 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_21 = x_20 + addcmul_1 + x_20 = addcmul_1 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (80,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_2 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (160,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_g_2 = x_34.norm(p=2, dim=(2, 3), keepdim=True) + mean_2 = x_g_2.mean(dim=1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_34 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_35 = x_34 + addcmul_2 + x_34 = addcmul_2 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_37 = torch.nn.functional.dropout(x_36, 0.0, False, False) + x_36 = None + x_38 = x_37 + input_2 + x_37 = input_2 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (160,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_g_3 = x_45.norm(p=2, dim=(2, 3), keepdim=True) + mean_3 = x_g_3.mean(dim=1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_45 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_46 = x_45 + addcmul_3 + x_45 = addcmul_3 = None + x_47 = torch.conv2d( + x_46, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (160,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_3 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_g_4 = x_59.norm(p=2, dim=(2, 3), keepdim=True) + mean_4 = x_g_4.mean(dim=1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_59 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_60 = x_59 + addcmul_4 + x_59 = addcmul_4 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_60 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + x_63 = x_62 + input_3 + x_62 = input_3 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_g_5 = x_70.norm(p=2, dim=(2, 3), keepdim=True) + mean_5 = x_g_5.mean(dim=1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_70 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_71 = x_70 + addcmul_5 + x_70 = addcmul_5 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_g_6 = x_81.norm(p=2, dim=(2, 3), keepdim=True) + mean_6 = x_g_6.mean(dim=1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_81 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_82 = x_81 + addcmul_6 + x_81 = addcmul_6 = None + x_83 = torch.conv2d( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_82 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_g_7 = x_92.norm(p=2, dim=(2, 3), keepdim=True) + mean_7 = x_g_7.mean(dim=1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_92 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_93 = x_92 + addcmul_7 + x_92 = addcmul_7 = None + x_94 = torch.conv2d( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_g_8 = x_103.norm(p=2, dim=(2, 3), keepdim=True) + mean_8 = x_g_8.mean(dim=1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_103 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_104 = x_103 + addcmul_8 + x_103 = addcmul_8 = None + x_105 = torch.conv2d( + x_104, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_104 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_g_9 = x_114.norm(p=2, dim=(2, 3), keepdim=True) + mean_9 = x_g_9.mean(dim=1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_114 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_115 = x_114 + addcmul_9 + x_114 = addcmul_9 = None + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_122 = x_121.permute(0, 3, 1, 2) + x_121 = None + x_123 = torch.conv2d( + x_122, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_124 = torch._C._nn.gelu(x_123) + x_123 = None + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_g_10 = x_125.norm(p=2, dim=(2, 3), keepdim=True) + mean_10 = x_g_10.mean(dim=1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_125 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_126 = x_125 + addcmul_10 + x_125 = addcmul_10 = None + x_127 = torch.conv2d( + x_126, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_126 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_133 = x_132.permute(0, 3, 1, 2) + x_132 = None + x_134 = torch.conv2d( + x_133, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_133 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_135 = torch._C._nn.gelu(x_134) + x_134 = None + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + x_g_11 = x_136.norm(p=2, dim=(2, 3), keepdim=True) + mean_11 = x_g_11.mean(dim=1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_136 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_137 = x_136 + addcmul_11 + x_136 = addcmul_11 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = x_140.permute(0, 2, 3, 1) + x_140 = None + x_142 = torch.nn.functional.layer_norm( + x_141, + (320,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_141 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_143 = x_142.permute(0, 3, 1, 2) + x_142 = None + input_4 = torch.conv2d( + x_143, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_143 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_144 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 640, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (640,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = torch.conv2d( + x_147, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_147 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_149 = torch._C._nn.gelu(x_148) + x_148 = None + x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) + x_149 = None + x_g_12 = x_150.norm(p=2, dim=(2, 3), keepdim=True) + mean_12 = x_g_12.mean(dim=1, keepdim=True) + add_36 = mean_12 + 1e-06 + mean_12 = None + x_n_12 = x_g_12 / add_36 + x_g_12 = add_36 = None + view_24 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_25 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_12 = x_150 * x_n_12 + x_n_12 = None + addcmul_12 = torch.addcmul(view_24, view_25, mul_12) + view_24 = view_25 = mul_12 = None + x_151 = x_150 + addcmul_12 + x_150 = addcmul_12 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_151 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_153 = torch.nn.functional.dropout(x_152, 0.0, False, False) + x_152 = None + x_154 = x_153 + input_4 + x_153 = input_4 = None + x_155 = torch.conv2d( + x_154, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 640, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_156 = x_155.permute(0, 2, 3, 1) + x_155 = None + x_157 = torch.nn.functional.layer_norm( + x_156, + (640,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_156 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_158 = x_157.permute(0, 3, 1, 2) + x_157 = None + x_159 = torch.conv2d( + x_158, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_158 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_160 = torch._C._nn.gelu(x_159) + x_159 = None + x_161 = torch.nn.functional.dropout(x_160, 0.0, False, False) + x_160 = None + x_g_13 = x_161.norm(p=2, dim=(2, 3), keepdim=True) + mean_13 = x_g_13.mean(dim=1, keepdim=True) + add_39 = mean_13 + 1e-06 + mean_13 = None + x_n_13 = x_g_13 / add_39 + x_g_13 = add_39 = None + view_26 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_27 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_13 = x_161 * x_n_13 + x_n_13 = None + addcmul_13 = torch.addcmul(view_26, view_27, mul_13) + view_26 = view_27 = mul_13 = None + x_162 = x_161 + addcmul_13 + x_161 = addcmul_13 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_162 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_164 = torch.nn.functional.dropout(x_163, 0.0, False, False) + x_163 = None + x_165 = x_164 + x_154 + x_164 = x_154 = None + x_166 = torch.nn.functional.adaptive_avg_pool2d(x_165, 1) + x_165 = None + x_167 = x_166.permute(0, 2, 3, 1) + x_166 = None + x_168 = torch.nn.functional.layer_norm( + x_167, + (640,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_167 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_169 = x_168.permute(0, 3, 1, 2) + x_168 = None + x_170 = x_169.flatten(1, -1) + x_169 = None + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + return (x_171,) diff --git a/samples/timm/convnextv2_nano.fcmae/weight_meta.py b/samples/timm/convnextv2_nano.fcmae/weight_meta.py new file mode 100644 index 0000000000..db1b85b2b2 --- /dev/null +++ b/samples/timm/convnextv2_nano.fcmae/weight_meta.py @@ -0,0 +1,1610 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [80, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.229 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [160, 80, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [320, 160, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [640, 320, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [640, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2560, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [640, 2560, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [640, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2560, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [640, 2560, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_nano.fcmae_ft_in1k/graph_hash.txt b/samples/timm/convnextv2_nano.fcmae_ft_in1k/graph_hash.txt new file mode 100644 index 0000000000..da2fc1fbfe --- /dev/null +++ b/samples/timm/convnextv2_nano.fcmae_ft_in1k/graph_hash.txt @@ -0,0 +1 @@ +b8181c9f9ff27ad7cb072ba4161f271d2e379ae487acc6a9fac1a9cc78ee1b70 \ No newline at end of file diff --git a/samples/timm/convnextv2_nano.fcmae_ft_in1k/graph_net.json b/samples/timm/convnextv2_nano.fcmae_ft_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnextv2_nano.fcmae_ft_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_nano.fcmae_ft_in1k/input_meta.py b/samples/timm/convnextv2_nano.fcmae_ft_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_nano.fcmae_ft_in1k/input_tensor_constraints.py b/samples/timm/convnextv2_nano.fcmae_ft_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_nano.fcmae_ft_in1k/model.py b/samples/timm/convnextv2_nano.fcmae_ft_in1k/model.py new file mode 100644 index 0000000000..81a65077f4 --- /dev/null +++ b/samples/timm/convnextv2_nano.fcmae_ft_in1k/model.py @@ -0,0 +1,1503 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (80,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (80,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_g = x_9.norm(p=2, dim=(2, 3), keepdim=True) + mean = x_g.mean(dim=1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_9 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_10 = x_9 + addcmul + x_9 = addcmul = None + x_11 = torch.conv2d( + x_10, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_10 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 80, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (80,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_g_1 = x_20.norm(p=2, dim=(2, 3), keepdim=True) + mean_1 = x_g_1.mean(dim=1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_20 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_21 = x_20 + addcmul_1 + x_20 = addcmul_1 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (80,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_2 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (160,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_g_2 = x_34.norm(p=2, dim=(2, 3), keepdim=True) + mean_2 = x_g_2.mean(dim=1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_34 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_35 = x_34 + addcmul_2 + x_34 = addcmul_2 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_37 = torch.nn.functional.dropout(x_36, 0.0, False, False) + x_36 = None + x_38 = x_37 + input_2 + x_37 = input_2 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 160, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (160,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_g_3 = x_45.norm(p=2, dim=(2, 3), keepdim=True) + mean_3 = x_g_3.mean(dim=1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_45 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_46 = x_45 + addcmul_3 + x_45 = addcmul_3 = None + x_47 = torch.conv2d( + x_46, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (160,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_3 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_g_4 = x_59.norm(p=2, dim=(2, 3), keepdim=True) + mean_4 = x_g_4.mean(dim=1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_59 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_60 = x_59 + addcmul_4 + x_59 = addcmul_4 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_60 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + x_63 = x_62 + input_3 + x_62 = input_3 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_g_5 = x_70.norm(p=2, dim=(2, 3), keepdim=True) + mean_5 = x_g_5.mean(dim=1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_70 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_71 = x_70 + addcmul_5 + x_70 = addcmul_5 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_g_6 = x_81.norm(p=2, dim=(2, 3), keepdim=True) + mean_6 = x_g_6.mean(dim=1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_81 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_82 = x_81 + addcmul_6 + x_81 = addcmul_6 = None + x_83 = torch.conv2d( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_82 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_g_7 = x_92.norm(p=2, dim=(2, 3), keepdim=True) + mean_7 = x_g_7.mean(dim=1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_92 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_93 = x_92 + addcmul_7 + x_92 = addcmul_7 = None + x_94 = torch.conv2d( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_g_8 = x_103.norm(p=2, dim=(2, 3), keepdim=True) + mean_8 = x_g_8.mean(dim=1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_103 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_104 = x_103 + addcmul_8 + x_103 = addcmul_8 = None + x_105 = torch.conv2d( + x_104, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_104 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_g_9 = x_114.norm(p=2, dim=(2, 3), keepdim=True) + mean_9 = x_g_9.mean(dim=1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_114 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_115 = x_114 + addcmul_9 + x_114 = addcmul_9 = None + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_122 = x_121.permute(0, 3, 1, 2) + x_121 = None + x_123 = torch.conv2d( + x_122, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_124 = torch._C._nn.gelu(x_123) + x_123 = None + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_g_10 = x_125.norm(p=2, dim=(2, 3), keepdim=True) + mean_10 = x_g_10.mean(dim=1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_125 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_126 = x_125 + addcmul_10 + x_125 = addcmul_10 = None + x_127 = torch.conv2d( + x_126, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_126 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 320, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (320,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_133 = x_132.permute(0, 3, 1, 2) + x_132 = None + x_134 = torch.conv2d( + x_133, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_133 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_135 = torch._C._nn.gelu(x_134) + x_134 = None + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + x_g_11 = x_136.norm(p=2, dim=(2, 3), keepdim=True) + mean_11 = x_g_11.mean(dim=1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_136 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_137 = x_136 + addcmul_11 + x_136 = addcmul_11 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = x_140.permute(0, 2, 3, 1) + x_140 = None + x_142 = torch.nn.functional.layer_norm( + x_141, + (320,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_141 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_143 = x_142.permute(0, 3, 1, 2) + x_142 = None + input_4 = torch.conv2d( + x_143, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_143 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_144 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 640, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (640,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = torch.conv2d( + x_147, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_147 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_149 = torch._C._nn.gelu(x_148) + x_148 = None + x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) + x_149 = None + x_g_12 = x_150.norm(p=2, dim=(2, 3), keepdim=True) + mean_12 = x_g_12.mean(dim=1, keepdim=True) + add_36 = mean_12 + 1e-06 + mean_12 = None + x_n_12 = x_g_12 / add_36 + x_g_12 = add_36 = None + view_24 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_25 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_12 = x_150 * x_n_12 + x_n_12 = None + addcmul_12 = torch.addcmul(view_24, view_25, mul_12) + view_24 = view_25 = mul_12 = None + x_151 = x_150 + addcmul_12 + x_150 = addcmul_12 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_151 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_153 = torch.nn.functional.dropout(x_152, 0.0, False, False) + x_152 = None + x_154 = x_153 + input_4 + x_153 = input_4 = None + x_155 = torch.conv2d( + x_154, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 640, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_156 = x_155.permute(0, 2, 3, 1) + x_155 = None + x_157 = torch.nn.functional.layer_norm( + x_156, + (640,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_156 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_158 = x_157.permute(0, 3, 1, 2) + x_157 = None + x_159 = torch.conv2d( + x_158, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_158 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_160 = torch._C._nn.gelu(x_159) + x_159 = None + x_161 = torch.nn.functional.dropout(x_160, 0.0, False, False) + x_160 = None + x_g_13 = x_161.norm(p=2, dim=(2, 3), keepdim=True) + mean_13 = x_g_13.mean(dim=1, keepdim=True) + add_39 = mean_13 + 1e-06 + mean_13 = None + x_n_13 = x_g_13 / add_39 + x_g_13 = add_39 = None + view_26 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_27 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_13 = x_161 * x_n_13 + x_n_13 = None + addcmul_13 = torch.addcmul(view_26, view_27, mul_13) + view_26 = view_27 = mul_13 = None + x_162 = x_161 + addcmul_13 + x_161 = addcmul_13 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_162 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_164 = torch.nn.functional.dropout(x_163, 0.0, False, False) + x_163 = None + x_165 = x_164 + x_154 + x_164 = x_154 = None + x_166 = torch.nn.functional.adaptive_avg_pool2d(x_165, 1) + x_165 = None + x_167 = x_166.permute(0, 2, 3, 1) + x_166 = None + x_168 = torch.nn.functional.layer_norm( + x_167, + (640,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_167 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_169 = x_168.permute(0, 3, 1, 2) + x_168 = None + x_170 = x_169.flatten(1, -1) + x_169 = None + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = torch._C._nn.linear( + x_171, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_171 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_172,) diff --git a/samples/timm/convnextv2_nano.fcmae_ft_in1k/weight_meta.py b/samples/timm/convnextv2_nano.fcmae_ft_in1k/weight_meta.py new file mode 100644 index 0000000000..13baa05d48 --- /dev/null +++ b/samples/timm/convnextv2_nano.fcmae_ft_in1k/weight_meta.py @@ -0,0 +1,1630 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [80, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.228 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [80, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [320, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [80, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [160, 80, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [160, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [640, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [160, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [320, 160, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [320, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1280, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [320, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [640, 320, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [640, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2560, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [640, 2560, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [640, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2560, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [2560] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [640, 2560, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_pico.fcmae/graph_hash.txt b/samples/timm/convnextv2_pico.fcmae/graph_hash.txt new file mode 100644 index 0000000000..b62217ee3f --- /dev/null +++ b/samples/timm/convnextv2_pico.fcmae/graph_hash.txt @@ -0,0 +1 @@ +69e21bb20c51b1d85583aa5f82c6408fd0bd4b948662d296dd39624afcc86010 \ No newline at end of file diff --git a/samples/timm/convnextv2_pico.fcmae/graph_net.json b/samples/timm/convnextv2_pico.fcmae/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnextv2_pico.fcmae/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_pico.fcmae/input_meta.py b/samples/timm/convnextv2_pico.fcmae/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_pico.fcmae/input_tensor_constraints.py b/samples/timm/convnextv2_pico.fcmae/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_pico.fcmae/model.py b/samples/timm/convnextv2_pico.fcmae/model.py new file mode 100644 index 0000000000..24ec0ea944 --- /dev/null +++ b/samples/timm/convnextv2_pico.fcmae/model.py @@ -0,0 +1,1299 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (64,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 64, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_g = x_9.norm(p=2, dim=(2, 3), keepdim=True) + mean = x_g.mean(dim=1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_9 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_10 = x_9 + addcmul + x_9 = addcmul = None + x_11 = torch.conv2d( + x_10, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_10 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 64, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_g_1 = x_20.norm(p=2, dim=(2, 3), keepdim=True) + mean_1 = x_g_1.mean(dim=1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_20 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_21 = x_20 + addcmul_1 + x_20 = addcmul_1 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (64,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_2 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_g_2 = x_34.norm(p=2, dim=(2, 3), keepdim=True) + mean_2 = x_g_2.mean(dim=1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_34 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_35 = x_34 + addcmul_2 + x_34 = addcmul_2 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_37 = torch.nn.functional.dropout(x_36, 0.0, False, False) + x_36 = None + x_38 = x_37 + input_2 + x_37 = input_2 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_g_3 = x_45.norm(p=2, dim=(2, 3), keepdim=True) + mean_3 = x_g_3.mean(dim=1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_45 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_46 = x_45 + addcmul_3 + x_45 = addcmul_3 = None + x_47 = torch.conv2d( + x_46, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (128,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_3 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_g_4 = x_59.norm(p=2, dim=(2, 3), keepdim=True) + mean_4 = x_g_4.mean(dim=1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_59 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_60 = x_59 + addcmul_4 + x_59 = addcmul_4 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_60 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + x_63 = x_62 + input_3 + x_62 = input_3 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_g_5 = x_70.norm(p=2, dim=(2, 3), keepdim=True) + mean_5 = x_g_5.mean(dim=1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_70 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_71 = x_70 + addcmul_5 + x_70 = addcmul_5 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_g_6 = x_81.norm(p=2, dim=(2, 3), keepdim=True) + mean_6 = x_g_6.mean(dim=1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_81 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_82 = x_81 + addcmul_6 + x_81 = addcmul_6 = None + x_83 = torch.conv2d( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_82 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_g_7 = x_92.norm(p=2, dim=(2, 3), keepdim=True) + mean_7 = x_g_7.mean(dim=1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_92 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_93 = x_92 + addcmul_7 + x_92 = addcmul_7 = None + x_94 = torch.conv2d( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_g_8 = x_103.norm(p=2, dim=(2, 3), keepdim=True) + mean_8 = x_g_8.mean(dim=1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_103 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_104 = x_103 + addcmul_8 + x_103 = addcmul_8 = None + x_105 = torch.conv2d( + x_104, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_104 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_g_9 = x_114.norm(p=2, dim=(2, 3), keepdim=True) + mean_9 = x_g_9.mean(dim=1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_114 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_115 = x_114 + addcmul_9 + x_114 = addcmul_9 = None + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = x_118.permute(0, 2, 3, 1) + x_118 = None + x_120 = torch.nn.functional.layer_norm( + x_119, + (256,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_119 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_121 = x_120.permute(0, 3, 1, 2) + x_120 = None + input_4 = torch.conv2d( + x_121, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_122 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_123 = x_122.permute(0, 2, 3, 1) + x_122 = None + x_124 = torch.nn.functional.layer_norm( + x_123, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_125 = x_124.permute(0, 3, 1, 2) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126) + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_g_10 = x_128.norm(p=2, dim=(2, 3), keepdim=True) + mean_10 = x_g_10.mean(dim=1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_128 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_129 = x_128 + addcmul_10 + x_128 = addcmul_10 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = x_131 + input_4 + x_131 = input_4 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_134 = x_133.permute(0, 2, 3, 1) + x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_134 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_136 = x_135.permute(0, 3, 1, 2) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137) + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_g_11 = x_139.norm(p=2, dim=(2, 3), keepdim=True) + mean_11 = x_g_11.mean(dim=1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_139 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_140 = x_139 + addcmul_11 + x_139 = addcmul_11 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_140 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_142 + x_132 + x_142 = x_132 = None + x_144 = torch.nn.functional.adaptive_avg_pool2d(x_143, 1) + x_143 = None + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (512,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = x_147.flatten(1, -1) + x_147 = None + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + return (x_149,) diff --git a/samples/timm/convnextv2_pico.fcmae/weight_meta.py b/samples/timm/convnextv2_pico.fcmae/weight_meta.py new file mode 100644 index 0000000000..2a3d473067 --- /dev/null +++ b/samples/timm/convnextv2_pico.fcmae/weight_meta.py @@ -0,0 +1,1410 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [64, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.226 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [64, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [64, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [128, 64, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [256, 128, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [512, 256, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_pico.fcmae_ft_in1k/graph_hash.txt b/samples/timm/convnextv2_pico.fcmae_ft_in1k/graph_hash.txt new file mode 100644 index 0000000000..3f8200a58b --- /dev/null +++ b/samples/timm/convnextv2_pico.fcmae_ft_in1k/graph_hash.txt @@ -0,0 +1 @@ +069e5e4f030b7e114c60364f33ab05ca64cb2970bb19b122e93fc7528c1c2c86 \ No newline at end of file diff --git a/samples/timm/convnextv2_pico.fcmae_ft_in1k/graph_net.json b/samples/timm/convnextv2_pico.fcmae_ft_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnextv2_pico.fcmae_ft_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_pico.fcmae_ft_in1k/input_meta.py b/samples/timm/convnextv2_pico.fcmae_ft_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_pico.fcmae_ft_in1k/input_tensor_constraints.py b/samples/timm/convnextv2_pico.fcmae_ft_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_pico.fcmae_ft_in1k/model.py b/samples/timm/convnextv2_pico.fcmae_ft_in1k/model.py new file mode 100644 index 0000000000..67c81bec95 --- /dev/null +++ b/samples/timm/convnextv2_pico.fcmae_ft_in1k/model.py @@ -0,0 +1,1315 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (64,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 64, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = x_5.permute(0, 3, 1, 2) + x_5 = None + x_7 = torch.conv2d( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_8 = torch._C._nn.gelu(x_7) + x_7 = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_g = x_9.norm(p=2, dim=(2, 3), keepdim=True) + mean = x_g.mean(dim=1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_9 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_10 = x_9 + addcmul + x_9 = addcmul = None + x_11 = torch.conv2d( + x_10, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_10 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 64, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (64,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = x_16.permute(0, 3, 1, 2) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_19 = torch._C._nn.gelu(x_18) + x_18 = None + x_20 = torch.nn.functional.dropout(x_19, 0.0, False, False) + x_19 = None + x_g_1 = x_20.norm(p=2, dim=(2, 3), keepdim=True) + mean_1 = x_g_1.mean(dim=1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_20 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_21 = x_20 + addcmul_1 + x_20 = addcmul_1 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = x_24.permute(0, 2, 3, 1) + x_24 = None + x_26 = torch.nn.functional.layer_norm( + x_25, + (64,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_25 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_27 = x_26.permute(0, 3, 1, 2) + x_26 = None + input_2 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_28 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_29 = x_28.permute(0, 2, 3, 1) + x_28 = None + x_30 = torch.nn.functional.layer_norm( + x_29, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_29 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_31 = x_30.permute(0, 3, 1, 2) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_33 = torch._C._nn.gelu(x_32) + x_32 = None + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_g_2 = x_34.norm(p=2, dim=(2, 3), keepdim=True) + mean_2 = x_g_2.mean(dim=1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_34 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_35 = x_34 + addcmul_2 + x_34 = addcmul_2 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_37 = torch.nn.functional.dropout(x_36, 0.0, False, False) + x_36 = None + x_38 = x_37 + input_2 + x_37 = input_2 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 128, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (128,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_42 = x_41.permute(0, 3, 1, 2) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_44 = torch._C._nn.gelu(x_43) + x_43 = None + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_g_3 = x_45.norm(p=2, dim=(2, 3), keepdim=True) + mean_3 = x_g_3.mean(dim=1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_45 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_46 = x_45 + addcmul_3 + x_45 = addcmul_3 = None + x_47 = torch.conv2d( + x_46, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_48 + x_38 + x_48 = x_38 = None + x_50 = x_49.permute(0, 2, 3, 1) + x_49 = None + x_51 = torch.nn.functional.layer_norm( + x_50, + (128,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_50 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_52 = x_51.permute(0, 3, 1, 2) + x_51 = None + input_3 = torch.conv2d( + x_52, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_52 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_53 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_54 = x_53.permute(0, 2, 3, 1) + x_53 = None + x_55 = torch.nn.functional.layer_norm( + x_54, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_56 = x_55.permute(0, 3, 1, 2) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_58 = torch._C._nn.gelu(x_57) + x_57 = None + x_59 = torch.nn.functional.dropout(x_58, 0.0, False, False) + x_58 = None + x_g_4 = x_59.norm(p=2, dim=(2, 3), keepdim=True) + mean_4 = x_g_4.mean(dim=1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_59 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_60 = x_59 + addcmul_4 + x_59 = addcmul_4 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_60 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + x_63 = x_62 + input_3 + x_62 = input_3 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_65 = x_64.permute(0, 2, 3, 1) + x_64 = None + x_66 = torch.nn.functional.layer_norm( + x_65, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_67 = x_66.permute(0, 3, 1, 2) + x_66 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_67 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_69 = torch._C._nn.gelu(x_68) + x_68 = None + x_70 = torch.nn.functional.dropout(x_69, 0.0, False, False) + x_69 = None + x_g_5 = x_70.norm(p=2, dim=(2, 3), keepdim=True) + mean_5 = x_g_5.mean(dim=1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_70 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_71 = x_70 + addcmul_5 + x_70 = addcmul_5 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = x_73 + x_63 + x_73 = x_63 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_78 = x_77.permute(0, 3, 1, 2) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_80 = torch._C._nn.gelu(x_79) + x_79 = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_g_6 = x_81.norm(p=2, dim=(2, 3), keepdim=True) + mean_6 = x_g_6.mean(dim=1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_81 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_82 = x_81 + addcmul_6 + x_81 = addcmul_6 = None + x_83 = torch.conv2d( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_82 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = x_84 + x_74 + x_84 = x_74 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_89 = x_88.permute(0, 3, 1, 2) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_91 = torch._C._nn.gelu(x_90) + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_g_7 = x_92.norm(p=2, dim=(2, 3), keepdim=True) + mean_7 = x_g_7.mean(dim=1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_92 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_93 = x_92 + addcmul_7 + x_92 = addcmul_7 = None + x_94 = torch.conv2d( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_100 = x_99.permute(0, 3, 1, 2) + x_99 = None + x_101 = torch.conv2d( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_102 = torch._C._nn.gelu(x_101) + x_101 = None + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_g_8 = x_103.norm(p=2, dim=(2, 3), keepdim=True) + mean_8 = x_g_8.mean(dim=1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_103 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_104 = x_103 + addcmul_8 + x_103 = addcmul_8 = None + x_105 = torch.conv2d( + x_104, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_104 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 256, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (256,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_111 = x_110.permute(0, 3, 1, 2) + x_110 = None + x_112 = torch.conv2d( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112) + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_g_9 = x_114.norm(p=2, dim=(2, 3), keepdim=True) + mean_9 = x_g_9.mean(dim=1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_114 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_115 = x_114 + addcmul_9 + x_114 = addcmul_9 = None + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = x_118.permute(0, 2, 3, 1) + x_118 = None + x_120 = torch.nn.functional.layer_norm( + x_119, + (256,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_119 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_121 = x_120.permute(0, 3, 1, 2) + x_120 = None + input_4 = torch.conv2d( + x_121, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_122 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_123 = x_122.permute(0, 2, 3, 1) + x_122 = None + x_124 = torch.nn.functional.layer_norm( + x_123, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_125 = x_124.permute(0, 3, 1, 2) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126) + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_g_10 = x_128.norm(p=2, dim=(2, 3), keepdim=True) + mean_10 = x_g_10.mean(dim=1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_128 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_129 = x_128 + addcmul_10 + x_128 = addcmul_10 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = x_131 + input_4 + x_131 = input_4 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 512, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_134 = x_133.permute(0, 2, 3, 1) + x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (512,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_134 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_136 = x_135.permute(0, 3, 1, 2) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137) + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_g_11 = x_139.norm(p=2, dim=(2, 3), keepdim=True) + mean_11 = x_g_11.mean(dim=1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, -1, 1, 1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_139 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_140 = x_139 + addcmul_11 + x_139 = addcmul_11 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_140 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_142 + x_132 + x_142 = x_132 = None + x_144 = torch.nn.functional.adaptive_avg_pool2d(x_143, 1) + x_143 = None + x_145 = x_144.permute(0, 2, 3, 1) + x_144 = None + x_146 = torch.nn.functional.layer_norm( + x_145, + (512,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_145 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_147 = x_146.permute(0, 3, 1, 2) + x_146 = None + x_148 = x_147.flatten(1, -1) + x_147 = None + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = torch._C._nn.linear( + x_149, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_149 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_150,) diff --git a/samples/timm/convnextv2_pico.fcmae_ft_in1k/weight_meta.py b/samples/timm/convnextv2_pico.fcmae_ft_in1k/weight_meta.py new file mode 100644 index 0000000000..ee956f7d13 --- /dev/null +++ b/samples/timm/convnextv2_pico.fcmae_ft_in1k/weight_meta.py @@ -0,0 +1,1430 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [64, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.223 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [64, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [64, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [128, 64, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [128, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [256, 128, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [256, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [512, 256, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [512, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_tiny.fcmae/graph_hash.txt b/samples/timm/convnextv2_tiny.fcmae/graph_hash.txt new file mode 100644 index 0000000000..8b14ecd9ee --- /dev/null +++ b/samples/timm/convnextv2_tiny.fcmae/graph_hash.txt @@ -0,0 +1 @@ +768dce85976b59285690f156b20646da6537dbcca7eb13625f756f057c303bea \ No newline at end of file diff --git a/samples/timm/convnextv2_tiny.fcmae/graph_net.json b/samples/timm/convnextv2_tiny.fcmae/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnextv2_tiny.fcmae/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_tiny.fcmae/input_meta.py b/samples/timm/convnextv2_tiny.fcmae/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_tiny.fcmae/input_tensor_constraints.py b/samples/timm/convnextv2_tiny.fcmae/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_tiny.fcmae/model.py b/samples/timm/convnextv2_tiny.fcmae/model.py new file mode 100644 index 0000000000..9df891fad3 --- /dev/null +++ b/samples/timm/convnextv2_tiny.fcmae/model.py @@ -0,0 +1,1719 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (96,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_g = x_8.norm(p=2, dim=(1, 2), keepdim=True) + mean = x_g.mean(dim=-1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_8 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_9 = x_8 + addcmul + x_8 = addcmul = None + x_10 = torch._C._nn.linear( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = x_11.permute(0, 3, 1, 2) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_g_1 = x_19.norm(p=2, dim=(1, 2), keepdim=True) + mean_1 = x_g_1.mean(dim=-1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_19 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_20 = x_19 + addcmul_1 + x_19 = addcmul_1 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_g_2 = x_30.norm(p=2, dim=(1, 2), keepdim=True) + mean_2 = x_g_2.mean(dim=-1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_30 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_31 = x_30 + addcmul_2 + x_30 = addcmul_2 = None + x_32 = torch._C._nn.linear( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + x_34 = x_33.permute(0, 3, 1, 2) + x_33 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (96,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_g_3 = x_44.norm(p=2, dim=(1, 2), keepdim=True) + mean_3 = x_g_3.mean(dim=-1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_44 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_45 = x_44 + addcmul_3 + x_44 = addcmul_3 = None + x_46 = torch._C._nn.linear( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_g_4 = x_55.norm(p=2, dim=(1, 2), keepdim=True) + mean_4 = x_g_4.mean(dim=-1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_55 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_56 = x_55 + addcmul_4 + x_55 = addcmul_4 = None + x_57 = torch._C._nn.linear( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_58 = torch.nn.functional.dropout(x_57, 0.0, False, False) + x_57 = None + x_59 = x_58.permute(0, 3, 1, 2) + x_58 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_g_5 = x_66.norm(p=2, dim=(1, 2), keepdim=True) + mean_5 = x_g_5.mean(dim=-1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_66 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_67 = x_66 + addcmul_5 + x_66 = addcmul_5 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = x_69.permute(0, 3, 1, 2) + x_69 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (192,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_g_6 = x_80.norm(p=2, dim=(1, 2), keepdim=True) + mean_6 = x_g_6.mean(dim=-1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_80 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_81 = x_80 + addcmul_6 + x_80 = addcmul_6 = None + x_82 = torch._C._nn.linear( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + x_84 = x_83.permute(0, 3, 1, 2) + x_83 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_g_7 = x_91.norm(p=2, dim=(1, 2), keepdim=True) + mean_7 = x_g_7.mean(dim=-1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_91 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_92 = x_91 + addcmul_7 + x_91 = addcmul_7 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_g_8 = x_102.norm(p=2, dim=(1, 2), keepdim=True) + mean_8 = x_g_8.mean(dim=-1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_102 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_103 = x_102 + addcmul_8 + x_102 = addcmul_8 = None + x_104 = torch._C._nn.linear( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = x_105.permute(0, 3, 1, 2) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_g_9 = x_113.norm(p=2, dim=(1, 2), keepdim=True) + mean_9 = x_g_9.mean(dim=-1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_113 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_114 = x_113 + addcmul_9 + x_113 = addcmul_9 = None + x_115 = torch._C._nn.linear( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + x_117 = x_116.permute(0, 3, 1, 2) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_g_10 = x_124.norm(p=2, dim=(1, 2), keepdim=True) + mean_10 = x_g_10.mean(dim=-1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_124 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_125 = x_124 + addcmul_10 + x_124 = addcmul_10 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = x_127.permute(0, 3, 1, 2) + x_127 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_g_11 = x_135.norm(p=2, dim=(1, 2), keepdim=True) + mean_11 = x_g_11.mean(dim=-1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_135 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_136 = x_135 + addcmul_11 + x_135 = addcmul_11 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_138.permute(0, 3, 1, 2) + x_138 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_g_12 = x_146.norm(p=2, dim=(1, 2), keepdim=True) + mean_12 = x_g_12.mean(dim=-1, keepdim=True) + add_36 = mean_12 + 1e-06 + mean_12 = None + x_n_12 = x_g_12 / add_36 + x_g_12 = add_36 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_12 = x_146 * x_n_12 + x_n_12 = None + addcmul_12 = torch.addcmul(view_24, view_25, mul_12) + view_24 = view_25 = mul_12 = None + x_147 = x_146 + addcmul_12 + x_146 = addcmul_12 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_149.permute(0, 3, 1, 2) + x_149 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_g_13 = x_157.norm(p=2, dim=(1, 2), keepdim=True) + mean_13 = x_g_13.mean(dim=-1, keepdim=True) + add_39 = mean_13 + 1e-06 + mean_13 = None + x_n_13 = x_g_13 / add_39 + x_g_13 = add_39 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_13 = x_157 * x_n_13 + x_n_13 = None + addcmul_13 = torch.addcmul(view_26, view_27, mul_13) + view_26 = view_27 = mul_13 = None + x_158 = x_157 + addcmul_13 + x_157 = addcmul_13 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_160.permute(0, 3, 1, 2) + x_160 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_g_14 = x_168.norm(p=2, dim=(1, 2), keepdim=True) + mean_14 = x_g_14.mean(dim=-1, keepdim=True) + add_42 = mean_14 + 1e-06 + mean_14 = None + x_n_14 = x_g_14 / add_42 + x_g_14 = add_42 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_14 = x_168 * x_n_14 + x_n_14 = None + addcmul_14 = torch.addcmul(view_28, view_29, mul_14) + view_28 = view_29 = mul_14 = None + x_169 = x_168 + addcmul_14 + x_168 = addcmul_14 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_171.permute(0, 3, 1, 2) + x_171 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = x_173.permute(0, 2, 3, 1) + x_173 = None + x_175 = torch.nn.functional.layer_norm( + x_174, + (384,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_174 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_176 = x_175.permute(0, 3, 1, 2) + x_175 = None + input_4 = torch.conv2d( + x_176, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_176 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_177 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_178 = x_177.permute(0, 2, 3, 1) + x_177 = None + x_179 = torch.nn.functional.layer_norm( + x_178, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_178 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_180 = torch._C._nn.linear( + x_179, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_179 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_181 = torch._C._nn.gelu(x_180) + x_180 = None + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_g_15 = x_182.norm(p=2, dim=(1, 2), keepdim=True) + mean_15 = x_g_15.mean(dim=-1, keepdim=True) + add_45 = mean_15 + 1e-06 + mean_15 = None + x_n_15 = x_g_15 / add_45 + x_g_15 = add_45 = None + view_30 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_31 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_15 = x_182 * x_n_15 + x_n_15 = None + addcmul_15 = torch.addcmul(view_30, view_31, mul_15) + view_30 = view_31 = mul_15 = None + x_183 = x_182 + addcmul_15 + x_182 = addcmul_15 = None + x_184 = torch._C._nn.linear( + x_183, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_183 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_185 = torch.nn.functional.dropout(x_184, 0.0, False, False) + x_184 = None + x_186 = x_185.permute(0, 3, 1, 2) + x_185 = None + x_187 = x_186 + input_4 + x_186 = input_4 = None + x_188 = torch.conv2d( + x_187, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_189 = x_188.permute(0, 2, 3, 1) + x_188 = None + x_190 = torch.nn.functional.layer_norm( + x_189, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_189 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_191 = torch._C._nn.linear( + x_190, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_190 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_192 = torch._C._nn.gelu(x_191) + x_191 = None + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_g_16 = x_193.norm(p=2, dim=(1, 2), keepdim=True) + mean_16 = x_g_16.mean(dim=-1, keepdim=True) + add_48 = mean_16 + 1e-06 + mean_16 = None + x_n_16 = x_g_16 / add_48 + x_g_16 = add_48 = None + view_32 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_33 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_16 = x_193 * x_n_16 + x_n_16 = None + addcmul_16 = torch.addcmul(view_32, view_33, mul_16) + view_32 = view_33 = mul_16 = None + x_194 = x_193 + addcmul_16 + x_193 = addcmul_16 = None + x_195 = torch._C._nn.linear( + x_194, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_194 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_196 = torch.nn.functional.dropout(x_195, 0.0, False, False) + x_195 = None + x_197 = x_196.permute(0, 3, 1, 2) + x_196 = None + x_198 = x_197 + x_187 + x_197 = x_187 = None + x_199 = torch.conv2d( + x_198, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_200 = x_199.permute(0, 2, 3, 1) + x_199 = None + x_201 = torch.nn.functional.layer_norm( + x_200, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_200 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_202 = torch._C._nn.linear( + x_201, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_201 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_203 = torch._C._nn.gelu(x_202) + x_202 = None + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_g_17 = x_204.norm(p=2, dim=(1, 2), keepdim=True) + mean_17 = x_g_17.mean(dim=-1, keepdim=True) + add_51 = mean_17 + 1e-06 + mean_17 = None + x_n_17 = x_g_17 / add_51 + x_g_17 = add_51 = None + view_34 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_35 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_17 = x_204 * x_n_17 + x_n_17 = None + addcmul_17 = torch.addcmul(view_34, view_35, mul_17) + view_34 = view_35 = mul_17 = None + x_205 = x_204 + addcmul_17 + x_204 = addcmul_17 = None + x_206 = torch._C._nn.linear( + x_205, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_205 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_207 = torch.nn.functional.dropout(x_206, 0.0, False, False) + x_206 = None + x_208 = x_207.permute(0, 3, 1, 2) + x_207 = None + x_209 = x_208 + x_198 + x_208 = x_198 = None + x_210 = torch.nn.functional.adaptive_avg_pool2d(x_209, 1) + x_209 = None + x_211 = x_210.permute(0, 2, 3, 1) + x_210 = None + x_212 = torch.nn.functional.layer_norm( + x_211, + (768,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_211 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_213 = x_212.permute(0, 3, 1, 2) + x_212 = None + x_214 = x_213.flatten(1, -1) + x_213 = None + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + return (x_215,) diff --git a/samples/timm/convnextv2_tiny.fcmae/weight_meta.py b/samples/timm/convnextv2_tiny.fcmae/weight_meta.py new file mode 100644 index 0000000000..e9058b6446 --- /dev/null +++ b/samples/timm/convnextv2_tiny.fcmae/weight_meta.py @@ -0,0 +1,2010 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [96, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.226 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [192, 96, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [384, 192, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [768, 384, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/convnextv2_tiny.fcmae_ft_in1k/graph_hash.txt b/samples/timm/convnextv2_tiny.fcmae_ft_in1k/graph_hash.txt new file mode 100644 index 0000000000..4a54d453e6 --- /dev/null +++ b/samples/timm/convnextv2_tiny.fcmae_ft_in1k/graph_hash.txt @@ -0,0 +1 @@ +51358064ff58e02a77b4420c059759367a1fd15f6dac8988e275cc60c5a314b4 \ No newline at end of file diff --git a/samples/timm/convnextv2_tiny.fcmae_ft_in1k/graph_net.json b/samples/timm/convnextv2_tiny.fcmae_ft_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/convnextv2_tiny.fcmae_ft_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/convnextv2_tiny.fcmae_ft_in1k/input_meta.py b/samples/timm/convnextv2_tiny.fcmae_ft_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_tiny.fcmae_ft_in1k/input_tensor_constraints.py b/samples/timm/convnextv2_tiny.fcmae_ft_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/convnextv2_tiny.fcmae_ft_in1k/model.py b/samples/timm/convnextv2_tiny.fcmae_ft_in1k/model.py new file mode 100644 index 0000000000..0f35682440 --- /dev/null +++ b/samples/timm/convnextv2_tiny.fcmae_ft_in1k/model.py @@ -0,0 +1,1735 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_0_parameters_weight_ = ( + L_self_modules_stem_modules_0_parameters_weight_ + ) + l_self_modules_stem_modules_0_parameters_bias_ = ( + L_self_modules_stem_modules_0_parameters_bias_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_1_parameters_weight_ = ( + L_self_modules_stem_modules_1_parameters_weight_ + ) + l_self_modules_stem_modules_1_parameters_bias_ = ( + L_self_modules_stem_modules_1_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_head_modules_norm_parameters_weight_ = ( + L_self_modules_head_modules_norm_parameters_weight_ + ) + l_self_modules_head_modules_norm_parameters_bias_ = ( + L_self_modules_head_modules_norm_parameters_bias_ + ) + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_stem_modules_0_parameters_weight_, + l_self_modules_stem_modules_0_parameters_bias_, + (4, 4), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_stem_modules_0_parameters_weight_ + ) = l_self_modules_stem_modules_0_parameters_bias_ = None + x = input_1.permute(0, 2, 3, 1) + input_1 = None + x_1 = torch.nn.functional.layer_norm( + x, + (96,), + l_self_modules_stem_modules_1_parameters_weight_, + l_self_modules_stem_modules_1_parameters_bias_, + 1e-06, + ) + x = ( + l_self_modules_stem_modules_1_parameters_weight_ + ) = l_self_modules_stem_modules_1_parameters_bias_ = None + x_2 = x_1.permute(0, 3, 1, 2) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_4 = x_3.permute(0, 2, 3, 1) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_4 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_7 = torch._C._nn.gelu(x_6) + x_6 = None + x_8 = torch.nn.functional.dropout(x_7, 0.0, False, False) + x_7 = None + x_g = x_8.norm(p=2, dim=(1, 2), keepdim=True) + mean = x_g.mean(dim=-1, keepdim=True) + add = mean + 1e-06 + mean = None + x_n = x_g / add + x_g = add = None + view = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_1 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul = x_8 * x_n + x_n = None + addcmul = torch.addcmul(view, view_1, mul) + view = view_1 = mul = None + x_9 = x_8 + addcmul + x_8 = addcmul = None + x_10 = torch._C._nn.linear( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = x_11.permute(0, 3, 1, 2) + x_11 = None + x_13 = x_12 + x_2 + x_12 = x_2 = None + x_14 = torch.conv2d( + x_13, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_15 = x_14.permute(0, 2, 3, 1) + x_14 = None + x_16 = torch.nn.functional.layer_norm( + x_15, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_18 = torch._C._nn.gelu(x_17) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_g_1 = x_19.norm(p=2, dim=(1, 2), keepdim=True) + mean_1 = x_g_1.mean(dim=-1, keepdim=True) + add_3 = mean_1 + 1e-06 + mean_1 = None + x_n_1 = x_g_1 / add_3 + x_g_1 = add_3 = None + view_2 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_3 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_1 = x_19 * x_n_1 + x_n_1 = None + addcmul_1 = torch.addcmul(view_2, view_3, mul_1) + view_2 = view_3 = mul_1 = None + x_20 = x_19 + addcmul_1 + x_19 = addcmul_1 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_22.permute(0, 3, 1, 2) + x_22 = None + x_24 = x_23 + x_13 + x_23 = x_13 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 96, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_26 = x_25.permute(0, 2, 3, 1) + x_25 = None + x_27 = torch.nn.functional.layer_norm( + x_26, + (96,), + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_29 = torch._C._nn.gelu(x_28) + x_28 = None + x_30 = torch.nn.functional.dropout(x_29, 0.0, False, False) + x_29 = None + x_g_2 = x_30.norm(p=2, dim=(1, 2), keepdim=True) + mean_2 = x_g_2.mean(dim=-1, keepdim=True) + add_6 = mean_2 + 1e-06 + mean_2 = None + x_n_2 = x_g_2 / add_6 + x_g_2 = add_6 = None + view_4 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_5 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_2 = x_30 * x_n_2 + x_n_2 = None + addcmul_2 = torch.addcmul(view_4, view_5, mul_2) + view_4 = view_5 = mul_2 = None + x_31 = x_30 + addcmul_2 + x_30 = addcmul_2 = None + x_32 = torch._C._nn.linear( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + x_34 = x_33.permute(0, 3, 1, 2) + x_33 = None + x_35 = x_34 + x_24 + x_34 = x_24 = None + x_36 = x_35.permute(0, 2, 3, 1) + x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (96,), + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_36 = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_ = (None) + x_38 = x_37.permute(0, 3, 1, 2) + x_37 = None + input_2 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_ = (None) + x_39 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_40 = x_39.permute(0, 2, 3, 1) + x_39 = None + x_41 = torch.nn.functional.layer_norm( + x_40, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_40 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42) + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_g_3 = x_44.norm(p=2, dim=(1, 2), keepdim=True) + mean_3 = x_g_3.mean(dim=-1, keepdim=True) + add_9 = mean_3 + 1e-06 + mean_3 = None + x_n_3 = x_g_3 / add_9 + x_g_3 = add_9 = None + view_6 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_7 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_3 = x_44 * x_n_3 + x_n_3 = None + addcmul_3 = torch.addcmul(view_6, view_7, mul_3) + view_6 = view_7 = mul_3 = None + x_45 = x_44 + addcmul_3 + x_44 = addcmul_3 = None + x_46 = torch._C._nn.linear( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_47 = torch.nn.functional.dropout(x_46, 0.0, False, False) + x_46 = None + x_48 = x_47.permute(0, 3, 1, 2) + x_47 = None + x_49 = x_48 + input_2 + x_48 = input_2 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_51 = x_50.permute(0, 2, 3, 1) + x_50 = None + x_52 = torch.nn.functional.layer_norm( + x_51, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53) + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_g_4 = x_55.norm(p=2, dim=(1, 2), keepdim=True) + mean_4 = x_g_4.mean(dim=-1, keepdim=True) + add_12 = mean_4 + 1e-06 + mean_4 = None + x_n_4 = x_g_4 / add_12 + x_g_4 = add_12 = None + view_8 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_9 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_4 = x_55 * x_n_4 + x_n_4 = None + addcmul_4 = torch.addcmul(view_8, view_9, mul_4) + view_8 = view_9 = mul_4 = None + x_56 = x_55 + addcmul_4 + x_55 = addcmul_4 = None + x_57 = torch._C._nn.linear( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_58 = torch.nn.functional.dropout(x_57, 0.0, False, False) + x_57 = None + x_59 = x_58.permute(0, 3, 1, 2) + x_58 = None + x_60 = x_59 + x_49 + x_59 = x_49 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 192, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_62 = x_61.permute(0, 2, 3, 1) + x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (192,), + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64) + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_g_5 = x_66.norm(p=2, dim=(1, 2), keepdim=True) + mean_5 = x_g_5.mean(dim=-1, keepdim=True) + add_15 = mean_5 + 1e-06 + mean_5 = None + x_n_5 = x_g_5 / add_15 + x_g_5 = add_15 = None + view_10 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_11 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_5 = x_66 * x_n_5 + x_n_5 = None + addcmul_5 = torch.addcmul(view_10, view_11, mul_5) + view_10 = view_11 = mul_5 = None + x_67 = x_66 + addcmul_5 + x_66 = addcmul_5 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = x_69.permute(0, 3, 1, 2) + x_69 = None + x_71 = x_70 + x_60 + x_70 = x_60 = None + x_72 = x_71.permute(0, 2, 3, 1) + x_71 = None + x_73 = torch.nn.functional.layer_norm( + x_72, + (192,), + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_72 = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_ = (None) + x_74 = x_73.permute(0, 3, 1, 2) + x_73 = None + input_3 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_ = (None) + x_75 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_76 = x_75.permute(0, 2, 3, 1) + x_75 = None + x_77 = torch.nn.functional.layer_norm( + x_76, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_79 = torch._C._nn.gelu(x_78) + x_78 = None + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_g_6 = x_80.norm(p=2, dim=(1, 2), keepdim=True) + mean_6 = x_g_6.mean(dim=-1, keepdim=True) + add_18 = mean_6 + 1e-06 + mean_6 = None + x_n_6 = x_g_6 / add_18 + x_g_6 = add_18 = None + view_12 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_13 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_6 = x_80 * x_n_6 + x_n_6 = None + addcmul_6 = torch.addcmul(view_12, view_13, mul_6) + view_12 = view_13 = mul_6 = None + x_81 = x_80 + addcmul_6 + x_80 = addcmul_6 = None + x_82 = torch._C._nn.linear( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + x_84 = x_83.permute(0, 3, 1, 2) + x_83 = None + x_85 = x_84 + input_3 + x_84 = input_3 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_87 = x_86.permute(0, 2, 3, 1) + x_86 = None + x_88 = torch.nn.functional.layer_norm( + x_87, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_89 = torch._C._nn.linear( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_90 = torch._C._nn.gelu(x_89) + x_89 = None + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_g_7 = x_91.norm(p=2, dim=(1, 2), keepdim=True) + mean_7 = x_g_7.mean(dim=-1, keepdim=True) + add_21 = mean_7 + 1e-06 + mean_7 = None + x_n_7 = x_g_7 / add_21 + x_g_7 = add_21 = None + view_14 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_15 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_7 = x_91 * x_n_7 + x_n_7 = None + addcmul_7 = torch.addcmul(view_14, view_15, mul_7) + view_14 = view_15 = mul_7 = None + x_92 = x_91 + addcmul_7 + x_91 = addcmul_7 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_94.permute(0, 3, 1, 2) + x_94 = None + x_96 = x_95 + x_85 + x_95 = x_85 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_98 = x_97.permute(0, 2, 3, 1) + x_97 = None + x_99 = torch.nn.functional.layer_norm( + x_98, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_101 = torch._C._nn.gelu(x_100) + x_100 = None + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_g_8 = x_102.norm(p=2, dim=(1, 2), keepdim=True) + mean_8 = x_g_8.mean(dim=-1, keepdim=True) + add_24 = mean_8 + 1e-06 + mean_8 = None + x_n_8 = x_g_8 / add_24 + x_g_8 = add_24 = None + view_16 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_17 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_8 = x_102 * x_n_8 + x_n_8 = None + addcmul_8 = torch.addcmul(view_16, view_17, mul_8) + view_16 = view_17 = mul_8 = None + x_103 = x_102 + addcmul_8 + x_102 = addcmul_8 = None + x_104 = torch._C._nn.linear( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = x_105.permute(0, 3, 1, 2) + x_105 = None + x_107 = x_106 + x_96 + x_106 = x_96 = None + x_108 = torch.conv2d( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_ = (None) + x_109 = x_108.permute(0, 2, 3, 1) + x_108 = None + x_110 = torch.nn.functional.layer_norm( + x_109, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_, + 1e-06, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_ = (None) + x_111 = torch._C._nn.linear( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_112 = torch._C._nn.gelu(x_111) + x_111 = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_g_9 = x_113.norm(p=2, dim=(1, 2), keepdim=True) + mean_9 = x_g_9.mean(dim=-1, keepdim=True) + add_27 = mean_9 + 1e-06 + mean_9 = None + x_n_9 = x_g_9 / add_27 + x_g_9 = add_27 = None + view_18 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_19 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_9 = x_113 * x_n_9 + x_n_9 = None + addcmul_9 = torch.addcmul(view_18, view_19, mul_9) + view_18 = view_19 = mul_9 = None + x_114 = x_113 + addcmul_9 + x_113 = addcmul_9 = None + x_115 = torch._C._nn.linear( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + x_117 = x_116.permute(0, 3, 1, 2) + x_116 = None + x_118 = x_117 + x_107 + x_117 = x_107 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_ = (None) + x_120 = x_119.permute(0, 2, 3, 1) + x_119 = None + x_121 = torch.nn.functional.layer_norm( + x_120, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_, + 1e-06, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_ = (None) + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_123 = torch._C._nn.gelu(x_122) + x_122 = None + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_g_10 = x_124.norm(p=2, dim=(1, 2), keepdim=True) + mean_10 = x_g_10.mean(dim=-1, keepdim=True) + add_30 = mean_10 + 1e-06 + mean_10 = None + x_n_10 = x_g_10 / add_30 + x_g_10 = add_30 = None + view_20 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_21 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_10 = x_124 * x_n_10 + x_n_10 = None + addcmul_10 = torch.addcmul(view_20, view_21, mul_10) + view_20 = view_21 = mul_10 = None + x_125 = x_124 + addcmul_10 + x_124 = addcmul_10 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = x_127.permute(0, 3, 1, 2) + x_127 = None + x_129 = x_128 + x_118 + x_128 = x_118 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_ = (None) + x_131 = x_130.permute(0, 2, 3, 1) + x_130 = None + x_132 = torch.nn.functional.layer_norm( + x_131, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_, + 1e-06, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_ = (None) + x_133 = torch._C._nn.linear( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_134 = torch._C._nn.gelu(x_133) + x_133 = None + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_g_11 = x_135.norm(p=2, dim=(1, 2), keepdim=True) + mean_11 = x_g_11.mean(dim=-1, keepdim=True) + add_33 = mean_11 + 1e-06 + mean_11 = None + x_n_11 = x_g_11 / add_33 + x_g_11 = add_33 = None + view_22 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_23 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_11 = x_135 * x_n_11 + x_n_11 = None + addcmul_11 = torch.addcmul(view_22, view_23, mul_11) + view_22 = view_23 = mul_11 = None + x_136 = x_135 + addcmul_11 + x_135 = addcmul_11 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_138.permute(0, 3, 1, 2) + x_138 = None + x_140 = x_139 + x_129 + x_139 = x_129 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_ = (None) + x_142 = x_141.permute(0, 2, 3, 1) + x_141 = None + x_143 = torch.nn.functional.layer_norm( + x_142, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_, + 1e-06, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_ = (None) + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_143 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_145 = torch._C._nn.gelu(x_144) + x_144 = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_g_12 = x_146.norm(p=2, dim=(1, 2), keepdim=True) + mean_12 = x_g_12.mean(dim=-1, keepdim=True) + add_36 = mean_12 + 1e-06 + mean_12 = None + x_n_12 = x_g_12 / add_36 + x_g_12 = add_36 = None + view_24 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_25 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_12 = x_146 * x_n_12 + x_n_12 = None + addcmul_12 = torch.addcmul(view_24, view_25, mul_12) + view_24 = view_25 = mul_12 = None + x_147 = x_146 + addcmul_12 + x_146 = addcmul_12 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_149.permute(0, 3, 1, 2) + x_149 = None + x_151 = x_150 + x_140 + x_150 = x_140 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_ = (None) + x_153 = x_152.permute(0, 2, 3, 1) + x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_, + 1e-06, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_ = (None) + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_156 = torch._C._nn.gelu(x_155) + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_g_13 = x_157.norm(p=2, dim=(1, 2), keepdim=True) + mean_13 = x_g_13.mean(dim=-1, keepdim=True) + add_39 = mean_13 + 1e-06 + mean_13 = None + x_n_13 = x_g_13 / add_39 + x_g_13 = add_39 = None + view_26 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_27 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_13 = x_157 * x_n_13 + x_n_13 = None + addcmul_13 = torch.addcmul(view_26, view_27, mul_13) + view_26 = view_27 = mul_13 = None + x_158 = x_157 + addcmul_13 + x_157 = addcmul_13 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_160.permute(0, 3, 1, 2) + x_160 = None + x_162 = x_161 + x_151 + x_161 = x_151 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 384, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_ = (None) + x_164 = x_163.permute(0, 2, 3, 1) + x_163 = None + x_165 = torch.nn.functional.layer_norm( + x_164, + (384,), + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_, + 1e-06, + ) + x_164 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_ = (None) + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_167 = torch._C._nn.gelu(x_166) + x_166 = None + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_g_14 = x_168.norm(p=2, dim=(1, 2), keepdim=True) + mean_14 = x_g_14.mean(dim=-1, keepdim=True) + add_42 = mean_14 + 1e-06 + mean_14 = None + x_n_14 = x_g_14 / add_42 + x_g_14 = add_42 = None + view_28 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_29 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_14 = x_168 * x_n_14 + x_n_14 = None + addcmul_14 = torch.addcmul(view_28, view_29, mul_14) + view_28 = view_29 = mul_14 = None + x_169 = x_168 + addcmul_14 + x_168 = addcmul_14 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_171.permute(0, 3, 1, 2) + x_171 = None + x_173 = x_172 + x_162 + x_172 = x_162 = None + x_174 = x_173.permute(0, 2, 3, 1) + x_173 = None + x_175 = torch.nn.functional.layer_norm( + x_174, + (384,), + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_, + 1e-06, + ) + x_174 = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_ = (None) + x_176 = x_175.permute(0, 3, 1, 2) + x_175 = None + input_4 = torch.conv2d( + x_176, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_, + l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_, + (2, 2), + (0, 0), + (1, 1), + 1, + ) + x_176 = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_ = l_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_ = (None) + x_177 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_ = (None) + x_178 = x_177.permute(0, 2, 3, 1) + x_177 = None + x_179 = torch.nn.functional.layer_norm( + x_178, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_, + 1e-06, + ) + x_178 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_ = (None) + x_180 = torch._C._nn.linear( + x_179, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_179 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_181 = torch._C._nn.gelu(x_180) + x_180 = None + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_g_15 = x_182.norm(p=2, dim=(1, 2), keepdim=True) + mean_15 = x_g_15.mean(dim=-1, keepdim=True) + add_45 = mean_15 + 1e-06 + mean_15 = None + x_n_15 = x_g_15 / add_45 + x_g_15 = add_45 = None + view_30 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_31 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_15 = x_182 * x_n_15 + x_n_15 = None + addcmul_15 = torch.addcmul(view_30, view_31, mul_15) + view_30 = view_31 = mul_15 = None + x_183 = x_182 + addcmul_15 + x_182 = addcmul_15 = None + x_184 = torch._C._nn.linear( + x_183, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_183 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_185 = torch.nn.functional.dropout(x_184, 0.0, False, False) + x_184 = None + x_186 = x_185.permute(0, 3, 1, 2) + x_185 = None + x_187 = x_186 + input_4 + x_186 = input_4 = None + x_188 = torch.conv2d( + x_187, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_ = (None) + x_189 = x_188.permute(0, 2, 3, 1) + x_188 = None + x_190 = torch.nn.functional.layer_norm( + x_189, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_, + 1e-06, + ) + x_189 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_ = (None) + x_191 = torch._C._nn.linear( + x_190, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_190 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_192 = torch._C._nn.gelu(x_191) + x_191 = None + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_g_16 = x_193.norm(p=2, dim=(1, 2), keepdim=True) + mean_16 = x_g_16.mean(dim=-1, keepdim=True) + add_48 = mean_16 + 1e-06 + mean_16 = None + x_n_16 = x_g_16 / add_48 + x_g_16 = add_48 = None + view_32 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_33 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_16 = x_193 * x_n_16 + x_n_16 = None + addcmul_16 = torch.addcmul(view_32, view_33, mul_16) + view_32 = view_33 = mul_16 = None + x_194 = x_193 + addcmul_16 + x_193 = addcmul_16 = None + x_195 = torch._C._nn.linear( + x_194, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_194 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_196 = torch.nn.functional.dropout(x_195, 0.0, False, False) + x_195 = None + x_197 = x_196.permute(0, 3, 1, 2) + x_196 = None + x_198 = x_197 + x_187 + x_197 = x_187 = None + x_199 = torch.conv2d( + x_198, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_, + (1, 1), + (3, 3), + (1, 1), + 768, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_ = (None) + x_200 = x_199.permute(0, 2, 3, 1) + x_199 = None + x_201 = torch.nn.functional.layer_norm( + x_200, + (768,), + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_, + 1e-06, + ) + x_200 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_ = (None) + x_202 = torch._C._nn.linear( + x_201, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_201 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_203 = torch._C._nn.gelu(x_202) + x_202 = None + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_g_17 = x_204.norm(p=2, dim=(1, 2), keepdim=True) + mean_17 = x_g_17.mean(dim=-1, keepdim=True) + add_51 = mean_17 + 1e-06 + mean_17 = None + x_n_17 = x_g_17 / add_51 + x_g_17 = add_51 = None + view_34 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_ = ( + None + ) + view_35 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_.view( + (1, 1, 1, -1) + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_ = ( + None + ) + mul_17 = x_204 * x_n_17 + x_n_17 = None + addcmul_17 = torch.addcmul(view_34, view_35, mul_17) + view_34 = view_35 = mul_17 = None + x_205 = x_204 + addcmul_17 + x_204 = addcmul_17 = None + x_206 = torch._C._nn.linear( + x_205, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_205 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_207 = torch.nn.functional.dropout(x_206, 0.0, False, False) + x_206 = None + x_208 = x_207.permute(0, 3, 1, 2) + x_207 = None + x_209 = x_208 + x_198 + x_208 = x_198 = None + x_210 = torch.nn.functional.adaptive_avg_pool2d(x_209, 1) + x_209 = None + x_211 = x_210.permute(0, 2, 3, 1) + x_210 = None + x_212 = torch.nn.functional.layer_norm( + x_211, + (768,), + l_self_modules_head_modules_norm_parameters_weight_, + l_self_modules_head_modules_norm_parameters_bias_, + 1e-06, + ) + x_211 = ( + l_self_modules_head_modules_norm_parameters_weight_ + ) = l_self_modules_head_modules_norm_parameters_bias_ = None + x_213 = x_212.permute(0, 3, 1, 2) + x_212 = None + x_214 = x_213.flatten(1, -1) + x_213 = None + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = torch._C._nn.linear( + x_215, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_215 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_216,) diff --git a/samples/timm/convnextv2_tiny.fcmae_ft_in1k/weight_meta.py b/samples/timm/convnextv2_tiny.fcmae_ft_in1k/weight_meta.py new file mode 100644 index 0000000000..b7de4f9bed --- /dev/null +++ b/samples/timm/convnextv2_tiny.fcmae_ft_in1k/weight_meta.py @@ -0,0 +1,2030 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_weight_: + name = "L_self_modules_stem_modules_0_parameters_weight_" + shape = [96, 3, 4, 4] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_0_parameters_bias_: + name = "L_self_modules_stem_modules_0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.230 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_weight_: + name = "L_self_modules_stem_modules_1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_1_parameters_bias_: + name = "L_self_modules_stem_modules_1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [96, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_0_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_weight_" + shape = [192, 96, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_downsample_modules_1_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [192, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_0_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_weight_" + shape = [384, 192, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_downsample_modules_1_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_weight_" + shape = [384, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv_dw_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_grn_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_0_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_weight_" + shape = [768, 384, 2, 2] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_downsample_modules_1_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_weight_" + shape = [768, 1, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv_dw_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_grn_parameters_weight_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_weight_: + name = "L_self_modules_head_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_norm_parameters_bias_: + name = "L_self_modules_head_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/crossvit_15_240.in1k/graph_hash.txt b/samples/timm/crossvit_15_240.in1k/graph_hash.txt new file mode 100644 index 0000000000..e8b59ad2a3 --- /dev/null +++ b/samples/timm/crossvit_15_240.in1k/graph_hash.txt @@ -0,0 +1 @@ +3a12e5809ccc0e1fdd5f7c10b900aadef0cd78c86eb87f88a87ef9b393136a85 \ No newline at end of file diff --git a/samples/timm/crossvit_15_240.in1k/graph_net.json b/samples/timm/crossvit_15_240.in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/crossvit_15_240.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/crossvit_15_240.in1k/input_meta.py b/samples/timm/crossvit_15_240.in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/crossvit_15_240.in1k/input_tensor_constraints.py b/samples/timm/crossvit_15_240.in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/crossvit_15_240.in1k/model.py b/samples/timm/crossvit_15_240.in1k/model.py new file mode 100644 index 0000000000..b48e5952ee --- /dev/null +++ b/samples/timm/crossvit_15_240.in1k/model.py @@ -0,0 +1,2733 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_0_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_0_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_1_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ + ) + l_self_parameters_cls_token_0_ = L_self_parameters_cls_token_0_ + l_self_parameters_pos_embed_0_ = L_self_parameters_pos_embed_0_ + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ + ) + l_self_parameters_cls_token_1_ = L_self_parameters_cls_token_1_ + l_self_parameters_pos_embed_1_ = L_self_parameters_pos_embed_1_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_norm_modules_0_parameters_weight_ = ( + L_self_modules_norm_modules_0_parameters_weight_ + ) + l_self_modules_norm_modules_0_parameters_bias_ = ( + L_self_modules_norm_modules_0_parameters_bias_ + ) + l_self_modules_norm_modules_1_parameters_weight_ = ( + L_self_modules_norm_modules_1_parameters_weight_ + ) + l_self_modules_norm_modules_1_parameters_bias_ = ( + L_self_modules_norm_modules_1_parameters_bias_ + ) + l_self_modules_head_modules_0_parameters_weight_ = ( + L_self_modules_head_modules_0_parameters_weight_ + ) + l_self_modules_head_modules_0_parameters_bias_ = ( + L_self_modules_head_modules_0_parameters_bias_ + ) + l_self_modules_head_modules_1_parameters_weight_ = ( + L_self_modules_head_modules_1_parameters_weight_ + ) + l_self_modules_head_modules_1_parameters_bias_ = ( + L_self_modules_head_modules_1_parameters_bias_ + ) + x = torch.nn.functional.interpolate( + l_x_, size=(240, 240), mode="bicubic", align_corners=False + ) + conv2d = torch.conv2d( + x, + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_, + (12, 12), + (0, 0), + (1, 1), + 1, + ) + x = ( + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ = None + flatten = conv2d.flatten(2) + conv2d = None + x_1 = flatten.transpose(1, 2) + flatten = None + cls_tokens = l_self_parameters_cls_token_0_.expand(1, -1, -1) + l_self_parameters_cls_token_0_ = None + x_ = torch.cat((cls_tokens, x_1), dim=1) + cls_tokens = x_1 = None + x__1 = x_ + l_self_parameters_pos_embed_0_ + x_ = l_self_parameters_pos_embed_0_ = None + x__2 = torch.nn.functional.dropout(x__1, 0.0, False, False) + x__1 = None + conv2d_1 = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ = None + flatten_1 = conv2d_1.flatten(2) + conv2d_1 = None + x_2 = flatten_1.transpose(1, 2) + flatten_1 = None + cls_tokens_1 = l_self_parameters_cls_token_1_.expand(1, -1, -1) + l_self_parameters_cls_token_1_ = None + x__3 = torch.cat((cls_tokens_1, x_2), dim=1) + cls_tokens_1 = x_2 = None + x__4 = x__3 + l_self_parameters_pos_embed_1_ + x__3 = l_self_parameters_pos_embed_1_ = None + x__5 = torch.nn.functional.dropout(x__4, 0.0, False, False) + x__4 = None + layer_norm = torch.nn.functional.layer_norm( + x__2, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear = torch._C._nn.linear( + layer_norm, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 401, 3, 6, 32) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_3 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_2 = x_3.transpose(1, 2) + x_3 = None + x_4 = transpose_2.reshape(1, 401, 192) + transpose_2 = None + x_5 = torch._C._nn.linear( + x_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_6 = torch.nn.functional.dropout(x_5, 0.0, False, False) + x_5 = None + x_7 = x__2 + x_6 + x__2 = x_6 = None + layer_norm_1 = torch.nn.functional.layer_norm( + x_7, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_8 = torch._C._nn.linear( + layer_norm_1, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_1 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_9 = torch._C._nn.gelu(x_8, approximate="none") + x_8 = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = torch._C._nn.linear( + x_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_7 + x_12 + x_7 = x_12 = None + layer_norm_2 = torch.nn.functional.layer_norm( + x__5, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_2, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_2 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_2 = linear_4.reshape(1, 197, 3, 6, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_14 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_3 = x_14.transpose(1, 2) + x_14 = None + x_15 = transpose_3.reshape(1, 197, 384) + transpose_3 = None + x_16 = torch._C._nn.linear( + x_15, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_15 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = x__5 + x_17 + x__5 = x_17 = None + layer_norm_3 = torch.nn.functional.layer_norm( + x_18, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_19 = torch._C._nn.linear( + layer_norm_3, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_3 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_20 = torch._C._nn.gelu(x_19, approximate="none") + x_19 = None + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = torch._C._nn.linear( + x_21, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_21 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_18 + x_23 + x_18 = x_23 = None + layer_norm_4 = torch.nn.functional.layer_norm( + x_24, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_8.reshape(1, 197, 3, 6, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_25 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_4 = x_25.transpose(1, 2) + x_25 = None + x_26 = transpose_4.reshape(1, 197, 384) + transpose_4 = None + x_27 = torch._C._nn.linear( + x_26, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_26 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_28 = torch.nn.functional.dropout(x_27, 0.0, False, False) + x_27 = None + x_29 = x_24 + x_28 + x_24 = x_28 = None + layer_norm_5 = torch.nn.functional.layer_norm( + x_29, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_30 = torch._C._nn.linear( + layer_norm_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_31 = torch._C._nn.gelu(x_30, approximate="none") + x_30 = None + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = torch._C._nn.linear( + x_32, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_32 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = x_29 + x_34 + x_29 = x_34 = None + layer_norm_6 = torch.nn.functional.layer_norm( + x_35, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_6, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_6 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_6 = linear_12.reshape(1, 197, 3, 6, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_36 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_5 = x_36.transpose(1, 2) + x_36 = None + x_37 = transpose_5.reshape(1, 197, 384) + transpose_5 = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_37 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_39 = torch.nn.functional.dropout(x_38, 0.0, False, False) + x_38 = None + x_40 = x_35 + x_39 + x_35 = x_39 = None + layer_norm_7 = torch.nn.functional.layer_norm( + x_40, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_41 = torch._C._nn.linear( + layer_norm_7, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_7 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_42 = torch._C._nn.gelu(x_41, approximate="none") + x_41 = None + x_43 = torch.nn.functional.dropout(x_42, 0.0, False, False) + x_42 = None + x_44 = torch._C._nn.linear( + x_43, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_43 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = x_40 + x_45 + x_40 = x_45 = None + layer_norm_8 = torch.nn.functional.layer_norm( + x_46, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_8, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_8 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_8 = linear_16.reshape(1, 197, 3, 6, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_47 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_6 = x_47.transpose(1, 2) + x_47 = None + x_48 = transpose_6.reshape(1, 197, 384) + transpose_6 = None + x_49 = torch._C._nn.linear( + x_48, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_48 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_50 = torch.nn.functional.dropout(x_49, 0.0, False, False) + x_49 = None + x_51 = x_46 + x_50 + x_46 = x_50 = None + layer_norm_9 = torch.nn.functional.layer_norm( + x_51, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_52 = torch._C._nn.linear( + layer_norm_9, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_9 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_53 = torch._C._nn.gelu(x_52, approximate="none") + x_52 = None + x_54 = torch.nn.functional.dropout(x_53, 0.0, False, False) + x_53 = None + x_55 = torch._C._nn.linear( + x_54, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_54 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_56 = torch.nn.functional.dropout(x_55, 0.0, False, False) + x_55 = None + x_57 = x_51 + x_56 + x_51 = x_56 = None + layer_norm_10 = torch.nn.functional.layer_norm( + x_57, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_20 = torch._C._nn.linear( + layer_norm_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_10 = linear_20.reshape(1, 197, 3, 6, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_7 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_7.reshape(1, 197, 384) + transpose_7 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + x_62 = x_57 + x_61 + x_57 = x_61 = None + layer_norm_11 = torch.nn.functional.layer_norm( + x_62, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_63 = torch._C._nn.linear( + layer_norm_11, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_11 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_64 = torch._C._nn.gelu(x_63, approximate="none") + x_63 = None + x_65 = torch.nn.functional.dropout(x_64, 0.0, False, False) + x_64 = None + x_66 = torch._C._nn.linear( + x_65, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_65 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_67 = torch.nn.functional.dropout(x_66, 0.0, False, False) + x_66 = None + x_68 = x_62 + x_67 + x_62 = x_67 = None + getitem_34 = x_13[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_1 = torch.nn.functional.layer_norm( + getitem_34, + (192,), + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_34 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_2 = torch._C._nn.gelu(input_1, approximate="none") + input_1 = None + input_3 = torch._C._nn.linear( + input_2, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_2 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_35 = x_68[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_4 = torch.nn.functional.layer_norm( + getitem_35, + (384,), + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_35 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_5 = torch._C._nn.gelu(input_4, approximate="none") + input_4 = None + input_6 = torch._C._nn.linear( + input_5, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_5 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_36 = x_68[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp = torch.cat((input_3, getitem_36), dim=1) + input_3 = getitem_36 = None + getitem_37 = tmp[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_14 = torch.nn.functional.layer_norm( + tmp, + (384,), + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_38 = layer_norm_14[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_26 = torch._C._nn.linear( + getitem_38, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_38 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_12 = linear_26.reshape(1, 1, 6, 64) + linear_26 = None + q_6 = reshape_12.permute(0, 2, 1, 3) + reshape_12 = None + linear_27 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_13 = linear_27.reshape(1, 197, 6, 64) + linear_27 = None + k_6 = reshape_13.permute(0, 2, 1, 3) + reshape_13 = None + linear_28 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_14 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_14 = linear_28.reshape(1, 197, 6, 64) + linear_28 = None + v_6 = reshape_14.permute(0, 2, 1, 3) + reshape_14 = None + transpose_8 = k_6.transpose(-2, -1) + k_6 = None + matmul = q_6 @ transpose_8 + q_6 = transpose_8 = None + attn = matmul * 0.125 + matmul = None + attn_1 = attn.softmax(dim=-1) + attn = None + attn_2 = torch.nn.functional.dropout(attn_1, 0.0, False, False) + attn_1 = None + matmul_1 = attn_2 @ v_6 + attn_2 = v_6 = None + transpose_9 = matmul_1.transpose(1, 2) + matmul_1 = None + x_69 = transpose_9.reshape(1, 1, 384) + transpose_9 = None + x_70 = torch._C._nn.linear( + x_69, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_69 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_71 = torch.nn.functional.dropout(x_70, 0.0, False, False) + x_70 = None + x_72 = getitem_37 + x_71 + getitem_37 = x_71 = None + getitem_39 = x_72[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_72 = None + input_7 = torch.nn.functional.layer_norm( + getitem_39, + (384,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_39 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_8 = torch._C._nn.gelu(input_7, approximate="none") + input_7 = None + input_9 = torch._C._nn.linear( + input_8, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_8 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_40 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_1 = torch.cat((input_9, getitem_40), dim=1) + input_9 = getitem_40 = None + getitem_41 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_13 = None + tmp_2 = torch.cat((input_6, getitem_41), dim=1) + input_6 = getitem_41 = None + getitem_42 = tmp_2[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_16 = torch.nn.functional.layer_norm( + tmp_2, + (192,), + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_2 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_43 = layer_norm_16[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_31 = torch._C._nn.linear( + getitem_43, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_43 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_16 = linear_31.reshape(1, 1, 6, 32) + linear_31 = None + q_7 = reshape_16.permute(0, 2, 1, 3) + reshape_16 = None + linear_32 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_17 = linear_32.reshape(1, 401, 6, 32) + linear_32 = None + k_7 = reshape_17.permute(0, 2, 1, 3) + reshape_17 = None + linear_33 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_16 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_18 = linear_33.reshape(1, 401, 6, 32) + linear_33 = None + v_7 = reshape_18.permute(0, 2, 1, 3) + reshape_18 = None + transpose_10 = k_7.transpose(-2, -1) + k_7 = None + matmul_2 = q_7 @ transpose_10 + q_7 = transpose_10 = None + attn_3 = matmul_2 * 0.1767766952966369 + matmul_2 = None + attn_4 = attn_3.softmax(dim=-1) + attn_3 = None + attn_5 = torch.nn.functional.dropout(attn_4, 0.0, False, False) + attn_4 = None + matmul_3 = attn_5 @ v_7 + attn_5 = v_7 = None + transpose_11 = matmul_3.transpose(1, 2) + matmul_3 = None + x_73 = transpose_11.reshape(1, 1, 192) + transpose_11 = None + x_74 = torch._C._nn.linear( + x_73, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_73 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_75 = torch.nn.functional.dropout(x_74, 0.0, False, False) + x_74 = None + x_76 = getitem_42 + x_75 + getitem_42 = x_75 = None + getitem_44 = x_76[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_76 = None + input_10 = torch.nn.functional.layer_norm( + getitem_44, + (192,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_44 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_11 = torch._C._nn.gelu(input_10, approximate="none") + input_10 = None + input_12 = torch._C._nn.linear( + input_11, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_11 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_45 = x_68[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_68 = None + tmp_3 = torch.cat((input_12, getitem_45), dim=1) + input_12 = getitem_45 = None + layer_norm_18 = torch.nn.functional.layer_norm( + tmp_1, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_18 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_20 = linear_36.reshape(1, 401, 3, 6, 32) + linear_36 = None + qkv_6 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_8 = unbind_6[0] + k_8 = unbind_6[1] + v_8 = unbind_6[2] + unbind_6 = None + x_77 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_12 = x_77.transpose(1, 2) + x_77 = None + x_78 = transpose_12.reshape(1, 401, 192) + transpose_12 = None + x_79 = torch._C._nn.linear( + x_78, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_78 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = tmp_1 + x_80 + tmp_1 = x_80 = None + layer_norm_19 = torch.nn.functional.layer_norm( + x_81, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_82 = torch._C._nn.linear( + layer_norm_19, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_19 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_83 = torch._C._nn.gelu(x_82, approximate="none") + x_82 = None + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = torch._C._nn.linear( + x_84, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_84 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_86 = torch.nn.functional.dropout(x_85, 0.0, False, False) + x_85 = None + x_87 = x_81 + x_86 + x_81 = x_86 = None + layer_norm_20 = torch.nn.functional.layer_norm( + tmp_3, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_20 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_22 = linear_40.reshape(1, 197, 3, 6, 64) + linear_40 = None + qkv_7 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_9 = unbind_7[0] + k_9 = unbind_7[1] + v_9 = unbind_7[2] + unbind_7 = None + x_88 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_13 = x_88.transpose(1, 2) + x_88 = None + x_89 = transpose_13.reshape(1, 197, 384) + transpose_13 = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_89 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = tmp_3 + x_91 + tmp_3 = x_91 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_92, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_93 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_21 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_94 = torch._C._nn.gelu(x_93, approximate="none") + x_93 = None + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = torch._C._nn.linear( + x_95, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_95 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_97 = torch.nn.functional.dropout(x_96, 0.0, False, False) + x_96 = None + x_98 = x_92 + x_97 + x_92 = x_97 = None + layer_norm_22 = torch.nn.functional.layer_norm( + x_98, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_22, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_22 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_24 = linear_44.reshape(1, 197, 3, 6, 64) + linear_44 = None + qkv_8 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_10 = unbind_8[0] + k_10 = unbind_8[1] + v_10 = unbind_8[2] + unbind_8 = None + x_99 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_14 = x_99.transpose(1, 2) + x_99 = None + x_100 = transpose_14.reshape(1, 197, 384) + transpose_14 = None + x_101 = torch._C._nn.linear( + x_100, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_100 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = x_98 + x_102 + x_98 = x_102 = None + layer_norm_23 = torch.nn.functional.layer_norm( + x_103, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_104 = torch._C._nn.linear( + layer_norm_23, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_23 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_105 = torch._C._nn.gelu(x_104, approximate="none") + x_104 = None + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch._C._nn.linear( + x_106, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_106 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_103 + x_108 + x_103 = x_108 = None + layer_norm_24 = torch.nn.functional.layer_norm( + x_109, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_24 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_26 = linear_48.reshape(1, 197, 3, 6, 64) + linear_48 = None + qkv_9 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_11 = unbind_9[0] + k_11 = unbind_9[1] + v_11 = unbind_9[2] + unbind_9 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_15 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_15.reshape(1, 197, 384) + transpose_15 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_109 + x_113 + x_109 = x_113 = None + layer_norm_25 = torch.nn.functional.layer_norm( + x_114, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_115 = torch._C._nn.linear( + layer_norm_25, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_25 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_116 = torch._C._nn.gelu(x_115, approximate="none") + x_115 = None + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = torch._C._nn.linear( + x_117, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_117 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_119 = torch.nn.functional.dropout(x_118, 0.0, False, False) + x_118 = None + x_120 = x_114 + x_119 + x_114 = x_119 = None + layer_norm_26 = torch.nn.functional.layer_norm( + x_120, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_52 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_26 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_28 = linear_52.reshape(1, 197, 3, 6, 64) + linear_52 = None + qkv_10 = reshape_28.permute(2, 0, 3, 1, 4) + reshape_28 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_12 = unbind_10[0] + k_12 = unbind_10[1] + v_12 = unbind_10[2] + unbind_10 = None + x_121 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_16 = x_121.transpose(1, 2) + x_121 = None + x_122 = transpose_16.reshape(1, 197, 384) + transpose_16 = None + x_123 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_122 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = x_120 + x_124 + x_120 = x_124 = None + layer_norm_27 = torch.nn.functional.layer_norm( + x_125, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_126 = torch._C._nn.linear( + layer_norm_27, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_27 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126, approximate="none") + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_128 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) + x_129 = None + x_131 = x_125 + x_130 + x_125 = x_130 = None + layer_norm_28 = torch.nn.functional.layer_norm( + x_131, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_28 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_30 = linear_56.reshape(1, 197, 3, 6, 64) + linear_56 = None + qkv_11 = reshape_30.permute(2, 0, 3, 1, 4) + reshape_30 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_13 = unbind_11[0] + k_13 = unbind_11[1] + v_13 = unbind_11[2] + unbind_11 = None + x_132 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_17 = x_132.transpose(1, 2) + x_132 = None + x_133 = transpose_17.reshape(1, 197, 384) + transpose_17 = None + x_134 = torch._C._nn.linear( + x_133, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_133 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = x_131 + x_135 + x_131 = x_135 = None + layer_norm_29 = torch.nn.functional.layer_norm( + x_136, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_137 = torch._C._nn.linear( + layer_norm_29, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_29 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137, approximate="none") + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = torch._C._nn.linear( + x_139, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_139 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_141 = torch.nn.functional.dropout(x_140, 0.0, False, False) + x_140 = None + x_142 = x_136 + x_141 + x_136 = x_141 = None + getitem_64 = x_87[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_13 = torch.nn.functional.layer_norm( + getitem_64, + (192,), + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_64 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_14 = torch._C._nn.gelu(input_13, approximate="none") + input_13 = None + input_15 = torch._C._nn.linear( + input_14, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_14 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_65 = x_142[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_16 = torch.nn.functional.layer_norm( + getitem_65, + (384,), + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_65 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_17 = torch._C._nn.gelu(input_16, approximate="none") + input_16 = None + input_18 = torch._C._nn.linear( + input_17, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_17 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_66 = x_142[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_4 = torch.cat((input_15, getitem_66), dim=1) + input_15 = getitem_66 = None + getitem_67 = tmp_4[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_32 = torch.nn.functional.layer_norm( + tmp_4, + (384,), + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_4 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_68 = layer_norm_32[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_62 = torch._C._nn.linear( + getitem_68, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_68 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_32 = linear_62.reshape(1, 1, 6, 64) + linear_62 = None + q_14 = reshape_32.permute(0, 2, 1, 3) + reshape_32 = None + linear_63 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_33 = linear_63.reshape(1, 197, 6, 64) + linear_63 = None + k_14 = reshape_33.permute(0, 2, 1, 3) + reshape_33 = None + linear_64 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_32 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_34 = linear_64.reshape(1, 197, 6, 64) + linear_64 = None + v_14 = reshape_34.permute(0, 2, 1, 3) + reshape_34 = None + transpose_18 = k_14.transpose(-2, -1) + k_14 = None + matmul_4 = q_14 @ transpose_18 + q_14 = transpose_18 = None + attn_6 = matmul_4 * 0.125 + matmul_4 = None + attn_7 = attn_6.softmax(dim=-1) + attn_6 = None + attn_8 = torch.nn.functional.dropout(attn_7, 0.0, False, False) + attn_7 = None + matmul_5 = attn_8 @ v_14 + attn_8 = v_14 = None + transpose_19 = matmul_5.transpose(1, 2) + matmul_5 = None + x_143 = transpose_19.reshape(1, 1, 384) + transpose_19 = None + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_143 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_145 = torch.nn.functional.dropout(x_144, 0.0, False, False) + x_144 = None + x_146 = getitem_67 + x_145 + getitem_67 = x_145 = None + getitem_69 = x_146[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_146 = None + input_19 = torch.nn.functional.layer_norm( + getitem_69, + (384,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_69 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_20 = torch._C._nn.gelu(input_19, approximate="none") + input_19 = None + input_21 = torch._C._nn.linear( + input_20, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_20 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_70 = x_87[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_5 = torch.cat((input_21, getitem_70), dim=1) + input_21 = getitem_70 = None + getitem_71 = x_87[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_87 = None + tmp_6 = torch.cat((input_18, getitem_71), dim=1) + input_18 = getitem_71 = None + getitem_72 = tmp_6[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_34 = torch.nn.functional.layer_norm( + tmp_6, + (192,), + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_6 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_73 = layer_norm_34[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_67 = torch._C._nn.linear( + getitem_73, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_73 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_36 = linear_67.reshape(1, 1, 6, 32) + linear_67 = None + q_15 = reshape_36.permute(0, 2, 1, 3) + reshape_36 = None + linear_68 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_37 = linear_68.reshape(1, 401, 6, 32) + linear_68 = None + k_15 = reshape_37.permute(0, 2, 1, 3) + reshape_37 = None + linear_69 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_34 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_38 = linear_69.reshape(1, 401, 6, 32) + linear_69 = None + v_15 = reshape_38.permute(0, 2, 1, 3) + reshape_38 = None + transpose_20 = k_15.transpose(-2, -1) + k_15 = None + matmul_6 = q_15 @ transpose_20 + q_15 = transpose_20 = None + attn_9 = matmul_6 * 0.1767766952966369 + matmul_6 = None + attn_10 = attn_9.softmax(dim=-1) + attn_9 = None + attn_11 = torch.nn.functional.dropout(attn_10, 0.0, False, False) + attn_10 = None + matmul_7 = attn_11 @ v_15 + attn_11 = v_15 = None + transpose_21 = matmul_7.transpose(1, 2) + matmul_7 = None + x_147 = transpose_21.reshape(1, 1, 192) + transpose_21 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_147 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = getitem_72 + x_149 + getitem_72 = x_149 = None + getitem_74 = x_150[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_150 = None + input_22 = torch.nn.functional.layer_norm( + getitem_74, + (192,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_74 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_23 = torch._C._nn.gelu(input_22, approximate="none") + input_22 = None + input_24 = torch._C._nn.linear( + input_23, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_23 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_75 = x_142[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_142 = None + tmp_7 = torch.cat((input_24, getitem_75), dim=1) + input_24 = getitem_75 = None + layer_norm_36 = torch.nn.functional.layer_norm( + tmp_5, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_72 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_36 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_40 = linear_72.reshape(1, 401, 3, 6, 32) + linear_72 = None + qkv_12 = reshape_40.permute(2, 0, 3, 1, 4) + reshape_40 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_16 = unbind_12[0] + k_16 = unbind_12[1] + v_16 = unbind_12[2] + unbind_12 = None + x_151 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, attn_mask=None, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_22 = x_151.transpose(1, 2) + x_151 = None + x_152 = transpose_22.reshape(1, 401, 192) + transpose_22 = None + x_153 = torch._C._nn.linear( + x_152, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_152 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_154 = torch.nn.functional.dropout(x_153, 0.0, False, False) + x_153 = None + x_155 = tmp_5 + x_154 + tmp_5 = x_154 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_155, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_156 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_37 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_157 = torch._C._nn.gelu(x_156, approximate="none") + x_156 = None + x_158 = torch.nn.functional.dropout(x_157, 0.0, False, False) + x_157 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_155 + x_160 + x_155 = x_160 = None + layer_norm_38 = torch.nn.functional.layer_norm( + tmp_7, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_76 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_38 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_42 = linear_76.reshape(1, 197, 3, 6, 64) + linear_76 = None + qkv_13 = reshape_42.permute(2, 0, 3, 1, 4) + reshape_42 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_17 = unbind_13[0] + k_17 = unbind_13[1] + v_17 = unbind_13[2] + unbind_13 = None + x_162 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, attn_mask=None, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_23 = x_162.transpose(1, 2) + x_162 = None + x_163 = transpose_23.reshape(1, 197, 384) + transpose_23 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_163 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_165 = torch.nn.functional.dropout(x_164, 0.0, False, False) + x_164 = None + x_166 = tmp_7 + x_165 + tmp_7 = x_165 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_166, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_167 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_39 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_168 = torch._C._nn.gelu(x_167, approximate="none") + x_167 = None + x_169 = torch.nn.functional.dropout(x_168, 0.0, False, False) + x_168 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_166 + x_171 + x_166 = x_171 = None + layer_norm_40 = torch.nn.functional.layer_norm( + x_172, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_40 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_44 = linear_80.reshape(1, 197, 3, 6, 64) + linear_80 = None + qkv_14 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_18 = unbind_14[0] + k_18 = unbind_14[1] + v_18 = unbind_14[2] + unbind_14 = None + x_173 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_24 = x_173.transpose(1, 2) + x_173 = None + x_174 = transpose_24.reshape(1, 197, 384) + transpose_24 = None + x_175 = torch._C._nn.linear( + x_174, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_174 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = x_172 + x_176 + x_172 = x_176 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_177, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_178 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_41 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_179 = torch._C._nn.gelu(x_178, approximate="none") + x_178 = None + x_180 = torch.nn.functional.dropout(x_179, 0.0, False, False) + x_179 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_177 + x_182 + x_177 = x_182 = None + layer_norm_42 = torch.nn.functional.layer_norm( + x_183, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_84 = torch._C._nn.linear( + layer_norm_42, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_42 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_46 = linear_84.reshape(1, 197, 3, 6, 64) + linear_84 = None + qkv_15 = reshape_46.permute(2, 0, 3, 1, 4) + reshape_46 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_19 = unbind_15[0] + k_19 = unbind_15[1] + v_19 = unbind_15[2] + unbind_15 = None + x_184 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, attn_mask=None, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_25 = x_184.transpose(1, 2) + x_184 = None + x_185 = transpose_25.reshape(1, 197, 384) + transpose_25 = None + x_186 = torch._C._nn.linear( + x_185, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_185 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_187 = torch.nn.functional.dropout(x_186, 0.0, False, False) + x_186 = None + x_188 = x_183 + x_187 + x_183 = x_187 = None + layer_norm_43 = torch.nn.functional.layer_norm( + x_188, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_189 = torch._C._nn.linear( + layer_norm_43, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_43 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_190 = torch._C._nn.gelu(x_189, approximate="none") + x_189 = None + x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) + x_190 = None + x_192 = torch._C._nn.linear( + x_191, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_191 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_188 + x_193 + x_188 = x_193 = None + layer_norm_44 = torch.nn.functional.layer_norm( + x_194, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_88 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_44 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_48 = linear_88.reshape(1, 197, 3, 6, 64) + linear_88 = None + qkv_16 = reshape_48.permute(2, 0, 3, 1, 4) + reshape_48 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_20 = unbind_16[0] + k_20 = unbind_16[1] + v_20 = unbind_16[2] + unbind_16 = None + x_195 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, attn_mask=None, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_26 = x_195.transpose(1, 2) + x_195 = None + x_196 = transpose_26.reshape(1, 197, 384) + transpose_26 = None + x_197 = torch._C._nn.linear( + x_196, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_196 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_198 = torch.nn.functional.dropout(x_197, 0.0, False, False) + x_197 = None + x_199 = x_194 + x_198 + x_194 = x_198 = None + layer_norm_45 = torch.nn.functional.layer_norm( + x_199, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_200 = torch._C._nn.linear( + layer_norm_45, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_45 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_201 = torch._C._nn.gelu(x_200, approximate="none") + x_200 = None + x_202 = torch.nn.functional.dropout(x_201, 0.0, False, False) + x_201 = None + x_203 = torch._C._nn.linear( + x_202, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_202 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_205 = x_199 + x_204 + x_199 = x_204 = None + layer_norm_46 = torch.nn.functional.layer_norm( + x_205, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_92 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_46 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_50 = linear_92.reshape(1, 197, 3, 6, 64) + linear_92 = None + qkv_17 = reshape_50.permute(2, 0, 3, 1, 4) + reshape_50 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_21 = unbind_17[0] + k_21 = unbind_17[1] + v_21 = unbind_17[2] + unbind_17 = None + x_206 = torch._C._nn.scaled_dot_product_attention( + q_21, k_21, v_21, attn_mask=None, dropout_p=0.0 + ) + q_21 = k_21 = v_21 = None + transpose_27 = x_206.transpose(1, 2) + x_206 = None + x_207 = transpose_27.reshape(1, 197, 384) + transpose_27 = None + x_208 = torch._C._nn.linear( + x_207, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_207 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_209 = torch.nn.functional.dropout(x_208, 0.0, False, False) + x_208 = None + x_210 = x_205 + x_209 + x_205 = x_209 = None + layer_norm_47 = torch.nn.functional.layer_norm( + x_210, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_211 = torch._C._nn.linear( + layer_norm_47, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_47 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_212 = torch._C._nn.gelu(x_211, approximate="none") + x_211 = None + x_213 = torch.nn.functional.dropout(x_212, 0.0, False, False) + x_212 = None + x_214 = torch._C._nn.linear( + x_213, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_213 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = x_210 + x_215 + x_210 = x_215 = None + getitem_94 = x_161[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_25 = torch.nn.functional.layer_norm( + getitem_94, + (192,), + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_94 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_26 = torch._C._nn.gelu(input_25, approximate="none") + input_25 = None + input_27 = torch._C._nn.linear( + input_26, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_26 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_95 = x_216[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_28 = torch.nn.functional.layer_norm( + getitem_95, + (384,), + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_95 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_29 = torch._C._nn.gelu(input_28, approximate="none") + input_28 = None + input_30 = torch._C._nn.linear( + input_29, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_29 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_96 = x_216[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_8 = torch.cat((input_27, getitem_96), dim=1) + input_27 = getitem_96 = None + getitem_97 = tmp_8[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_50 = torch.nn.functional.layer_norm( + tmp_8, + (384,), + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_8 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_98 = layer_norm_50[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_98 = torch._C._nn.linear( + getitem_98, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_98 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_52 = linear_98.reshape(1, 1, 6, 64) + linear_98 = None + q_22 = reshape_52.permute(0, 2, 1, 3) + reshape_52 = None + linear_99 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_53 = linear_99.reshape(1, 197, 6, 64) + linear_99 = None + k_22 = reshape_53.permute(0, 2, 1, 3) + reshape_53 = None + linear_100 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_50 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_54 = linear_100.reshape(1, 197, 6, 64) + linear_100 = None + v_22 = reshape_54.permute(0, 2, 1, 3) + reshape_54 = None + transpose_28 = k_22.transpose(-2, -1) + k_22 = None + matmul_8 = q_22 @ transpose_28 + q_22 = transpose_28 = None + attn_12 = matmul_8 * 0.125 + matmul_8 = None + attn_13 = attn_12.softmax(dim=-1) + attn_12 = None + attn_14 = torch.nn.functional.dropout(attn_13, 0.0, False, False) + attn_13 = None + matmul_9 = attn_14 @ v_22 + attn_14 = v_22 = None + transpose_29 = matmul_9.transpose(1, 2) + matmul_9 = None + x_217 = transpose_29.reshape(1, 1, 384) + transpose_29 = None + x_218 = torch._C._nn.linear( + x_217, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_217 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_219 = torch.nn.functional.dropout(x_218, 0.0, False, False) + x_218 = None + x_220 = getitem_97 + x_219 + getitem_97 = x_219 = None + getitem_99 = x_220[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_220 = None + input_31 = torch.nn.functional.layer_norm( + getitem_99, + (384,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_99 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_32 = torch._C._nn.gelu(input_31, approximate="none") + input_31 = None + input_33 = torch._C._nn.linear( + input_32, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_32 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_100 = x_161[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_9 = torch.cat((input_33, getitem_100), dim=1) + input_33 = getitem_100 = None + getitem_101 = x_161[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_161 = None + tmp_10 = torch.cat((input_30, getitem_101), dim=1) + input_30 = getitem_101 = None + getitem_102 = tmp_10[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_52 = torch.nn.functional.layer_norm( + tmp_10, + (192,), + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_10 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_103 = layer_norm_52[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_103 = torch._C._nn.linear( + getitem_103, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_103 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_56 = linear_103.reshape(1, 1, 6, 32) + linear_103 = None + q_23 = reshape_56.permute(0, 2, 1, 3) + reshape_56 = None + linear_104 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_57 = linear_104.reshape(1, 401, 6, 32) + linear_104 = None + k_23 = reshape_57.permute(0, 2, 1, 3) + reshape_57 = None + linear_105 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_52 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_58 = linear_105.reshape(1, 401, 6, 32) + linear_105 = None + v_23 = reshape_58.permute(0, 2, 1, 3) + reshape_58 = None + transpose_30 = k_23.transpose(-2, -1) + k_23 = None + matmul_10 = q_23 @ transpose_30 + q_23 = transpose_30 = None + attn_15 = matmul_10 * 0.1767766952966369 + matmul_10 = None + attn_16 = attn_15.softmax(dim=-1) + attn_15 = None + attn_17 = torch.nn.functional.dropout(attn_16, 0.0, False, False) + attn_16 = None + matmul_11 = attn_17 @ v_23 + attn_17 = v_23 = None + transpose_31 = matmul_11.transpose(1, 2) + matmul_11 = None + x_221 = transpose_31.reshape(1, 1, 192) + transpose_31 = None + x_222 = torch._C._nn.linear( + x_221, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_221 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_224 = getitem_102 + x_223 + getitem_102 = x_223 = None + getitem_104 = x_224[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_224 = None + input_34 = torch.nn.functional.layer_norm( + getitem_104, + (192,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_104 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_35 = torch._C._nn.gelu(input_34, approximate="none") + input_34 = None + input_36 = torch._C._nn.linear( + input_35, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_35 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_105 = x_216[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_216 = None + tmp_11 = torch.cat((input_36, getitem_105), dim=1) + input_36 = getitem_105 = None + x_225 = torch.nn.functional.layer_norm( + tmp_9, + (192,), + l_self_modules_norm_modules_0_parameters_weight_, + l_self_modules_norm_modules_0_parameters_bias_, + 1e-06, + ) + tmp_9 = ( + l_self_modules_norm_modules_0_parameters_weight_ + ) = l_self_modules_norm_modules_0_parameters_bias_ = None + x_226 = torch.nn.functional.layer_norm( + tmp_11, + (384,), + l_self_modules_norm_modules_1_parameters_weight_, + l_self_modules_norm_modules_1_parameters_bias_, + 1e-06, + ) + tmp_11 = ( + l_self_modules_norm_modules_1_parameters_weight_ + ) = l_self_modules_norm_modules_1_parameters_bias_ = None + x_227 = x_225[(slice(None, None, None), 0)] + x_225 = None + x_228 = x_226[(slice(None, None, None), 0)] + x_226 = None + dropout_68 = torch.nn.functional.dropout(x_227, 0.0, False, False) + x_227 = None + dropout_69 = torch.nn.functional.dropout(x_228, 0.0, False, False) + x_228 = None + linear_108 = torch._C._nn.linear( + dropout_68, + l_self_modules_head_modules_0_parameters_weight_, + l_self_modules_head_modules_0_parameters_bias_, + ) + dropout_68 = ( + l_self_modules_head_modules_0_parameters_weight_ + ) = l_self_modules_head_modules_0_parameters_bias_ = None + linear_109 = torch._C._nn.linear( + dropout_69, + l_self_modules_head_modules_1_parameters_weight_, + l_self_modules_head_modules_1_parameters_bias_, + ) + dropout_69 = ( + l_self_modules_head_modules_1_parameters_weight_ + ) = l_self_modules_head_modules_1_parameters_bias_ = None + stack = torch.stack([linear_108, linear_109], dim=0) + linear_108 = linear_109 = None + x_229 = torch.mean(stack, dim=0) + stack = None + return (x_229,) diff --git a/samples/timm/crossvit_15_240.in1k/weight_meta.py b/samples/timm/crossvit_15_240.in1k/weight_meta.py new file mode 100644 index 0000000000..ac649f4263 --- /dev/null +++ b/samples/timm/crossvit_15_240.in1k/weight_meta.py @@ -0,0 +1,3408 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.223 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_" + shape = [192, 3, 12, 12] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.028 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.028 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_0_: + name = "L_self_parameters_cls_token_0_" + shape = [1, 1, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_0_: + name = "L_self_parameters_pos_embed_0_" + shape = [1, 401, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_" + shape = [384, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_1_: + name = "L_self_parameters_cls_token_1_" + shape = [1, 1, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_1_: + name = "L_self_parameters_pos_embed_1_" + shape = [1, 197, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_weight_: + name = "L_self_modules_norm_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_bias_: + name = "L_self_modules_norm_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_weight_: + name = "L_self_modules_norm_modules_1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_bias_: + name = "L_self_modules_norm_modules_1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_weight_: + name = "L_self_modules_head_modules_0_parameters_weight_" + shape = [1000, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_bias_: + name = "L_self_modules_head_modules_0_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_weight_: + name = "L_self_modules_head_modules_1_parameters_weight_" + shape = [1000, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_bias_: + name = "L_self_modules_head_modules_1_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/crossvit_15_dagger_240.in1k/graph_hash.txt b/samples/timm/crossvit_15_dagger_240.in1k/graph_hash.txt new file mode 100644 index 0000000000..82cac6e4ef --- /dev/null +++ b/samples/timm/crossvit_15_dagger_240.in1k/graph_hash.txt @@ -0,0 +1 @@ +12e4a17846003b8fccb9f45487096cd775aab14f58f5b1129f1bf44433e25f1e \ No newline at end of file diff --git a/samples/timm/crossvit_15_dagger_240.in1k/graph_net.json b/samples/timm/crossvit_15_dagger_240.in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/crossvit_15_dagger_240.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/crossvit_15_dagger_240.in1k/input_meta.py b/samples/timm/crossvit_15_dagger_240.in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/crossvit_15_dagger_240.in1k/input_tensor_constraints.py b/samples/timm/crossvit_15_dagger_240.in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/crossvit_15_dagger_240.in1k/model.py b/samples/timm/crossvit_15_dagger_240.in1k/model.py new file mode 100644 index 0000000000..de3b64c5c2 --- /dev/null +++ b/samples/timm/crossvit_15_dagger_240.in1k/model.py @@ -0,0 +1,2809 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_0_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_0_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_1_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ + ) + l_self_parameters_cls_token_0_ = L_self_parameters_cls_token_0_ + l_self_parameters_pos_embed_0_ = L_self_parameters_pos_embed_0_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ + ) + l_self_parameters_cls_token_1_ = L_self_parameters_cls_token_1_ + l_self_parameters_pos_embed_1_ = L_self_parameters_pos_embed_1_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_norm_modules_0_parameters_weight_ = ( + L_self_modules_norm_modules_0_parameters_weight_ + ) + l_self_modules_norm_modules_0_parameters_bias_ = ( + L_self_modules_norm_modules_0_parameters_bias_ + ) + l_self_modules_norm_modules_1_parameters_weight_ = ( + L_self_modules_norm_modules_1_parameters_weight_ + ) + l_self_modules_norm_modules_1_parameters_bias_ = ( + L_self_modules_norm_modules_1_parameters_bias_ + ) + l_self_modules_head_modules_0_parameters_weight_ = ( + L_self_modules_head_modules_0_parameters_weight_ + ) + l_self_modules_head_modules_0_parameters_bias_ = ( + L_self_modules_head_modules_0_parameters_bias_ + ) + l_self_modules_head_modules_1_parameters_weight_ = ( + L_self_modules_head_modules_1_parameters_weight_ + ) + l_self_modules_head_modules_1_parameters_bias_ = ( + L_self_modules_head_modules_1_parameters_bias_ + ) + x = torch.nn.functional.interpolate( + l_x_, size=(240, 240), mode="bicubic", align_corners=False + ) + input_1 = torch.conv2d( + x, + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_, + (4, 4), + (3, 3), + (1, 1), + 1, + ) + x = l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ + ) = None + input_2 = torch.nn.functional.relu(input_1, inplace=True) + input_1 = None + input_3 = torch.conv2d( + input_2, + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_, + (3, 3), + (0, 0), + (1, 1), + 1, + ) + input_2 = l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ + ) = None + input_4 = torch.nn.functional.relu(input_3, inplace=True) + input_3 = None + input_5 = torch.conv2d( + input_4, + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + input_4 = l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ + ) = None + flatten = input_5.flatten(2) + input_5 = None + x_1 = flatten.transpose(1, 2) + flatten = None + cls_tokens = l_self_parameters_cls_token_0_.expand(1, -1, -1) + l_self_parameters_cls_token_0_ = None + x_ = torch.cat((cls_tokens, x_1), dim=1) + cls_tokens = x_1 = None + x__1 = x_ + l_self_parameters_pos_embed_0_ + x_ = l_self_parameters_pos_embed_0_ = None + x__2 = torch.nn.functional.dropout(x__1, 0.0, False, False) + x__1 = None + input_6 = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_, + (4, 4), + (3, 3), + (1, 1), + 1, + ) + l_x_ = l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ + ) = None + input_7 = torch.nn.functional.relu(input_6, inplace=True) + input_6 = None + input_8 = torch.conv2d( + input_7, + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_7 = l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ + ) = None + input_9 = torch.nn.functional.relu(input_8, inplace=True) + input_8 = None + input_10 = torch.conv2d( + input_9, + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_9 = l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ + ) = None + flatten_1 = input_10.flatten(2) + input_10 = None + x_2 = flatten_1.transpose(1, 2) + flatten_1 = None + cls_tokens_1 = l_self_parameters_cls_token_1_.expand(1, -1, -1) + l_self_parameters_cls_token_1_ = None + x__3 = torch.cat((cls_tokens_1, x_2), dim=1) + cls_tokens_1 = x_2 = None + x__4 = x__3 + l_self_parameters_pos_embed_1_ + x__3 = l_self_parameters_pos_embed_1_ = None + x__5 = torch.nn.functional.dropout(x__4, 0.0, False, False) + x__4 = None + layer_norm = torch.nn.functional.layer_norm( + x__2, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear = torch._C._nn.linear( + layer_norm, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 401, 3, 6, 32) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_3 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_2 = x_3.transpose(1, 2) + x_3 = None + x_4 = transpose_2.reshape(1, 401, 192) + transpose_2 = None + x_5 = torch._C._nn.linear( + x_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_6 = torch.nn.functional.dropout(x_5, 0.0, False, False) + x_5 = None + x_7 = x__2 + x_6 + x__2 = x_6 = None + layer_norm_1 = torch.nn.functional.layer_norm( + x_7, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_8 = torch._C._nn.linear( + layer_norm_1, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_1 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_9 = torch._C._nn.gelu(x_8, approximate="none") + x_8 = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = torch._C._nn.linear( + x_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_7 + x_12 + x_7 = x_12 = None + layer_norm_2 = torch.nn.functional.layer_norm( + x__5, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_2, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_2 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_2 = linear_4.reshape(1, 197, 3, 6, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_14 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_3 = x_14.transpose(1, 2) + x_14 = None + x_15 = transpose_3.reshape(1, 197, 384) + transpose_3 = None + x_16 = torch._C._nn.linear( + x_15, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_15 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = x__5 + x_17 + x__5 = x_17 = None + layer_norm_3 = torch.nn.functional.layer_norm( + x_18, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_19 = torch._C._nn.linear( + layer_norm_3, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_3 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_20 = torch._C._nn.gelu(x_19, approximate="none") + x_19 = None + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = torch._C._nn.linear( + x_21, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_21 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_18 + x_23 + x_18 = x_23 = None + layer_norm_4 = torch.nn.functional.layer_norm( + x_24, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_8.reshape(1, 197, 3, 6, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_25 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_4 = x_25.transpose(1, 2) + x_25 = None + x_26 = transpose_4.reshape(1, 197, 384) + transpose_4 = None + x_27 = torch._C._nn.linear( + x_26, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_26 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_28 = torch.nn.functional.dropout(x_27, 0.0, False, False) + x_27 = None + x_29 = x_24 + x_28 + x_24 = x_28 = None + layer_norm_5 = torch.nn.functional.layer_norm( + x_29, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_30 = torch._C._nn.linear( + layer_norm_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_31 = torch._C._nn.gelu(x_30, approximate="none") + x_30 = None + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = torch._C._nn.linear( + x_32, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_32 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = x_29 + x_34 + x_29 = x_34 = None + layer_norm_6 = torch.nn.functional.layer_norm( + x_35, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_6, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_6 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_6 = linear_12.reshape(1, 197, 3, 6, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_36 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_5 = x_36.transpose(1, 2) + x_36 = None + x_37 = transpose_5.reshape(1, 197, 384) + transpose_5 = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_37 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_39 = torch.nn.functional.dropout(x_38, 0.0, False, False) + x_38 = None + x_40 = x_35 + x_39 + x_35 = x_39 = None + layer_norm_7 = torch.nn.functional.layer_norm( + x_40, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_41 = torch._C._nn.linear( + layer_norm_7, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_7 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_42 = torch._C._nn.gelu(x_41, approximate="none") + x_41 = None + x_43 = torch.nn.functional.dropout(x_42, 0.0, False, False) + x_42 = None + x_44 = torch._C._nn.linear( + x_43, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_43 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = x_40 + x_45 + x_40 = x_45 = None + layer_norm_8 = torch.nn.functional.layer_norm( + x_46, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_8, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_8 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_8 = linear_16.reshape(1, 197, 3, 6, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_47 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_6 = x_47.transpose(1, 2) + x_47 = None + x_48 = transpose_6.reshape(1, 197, 384) + transpose_6 = None + x_49 = torch._C._nn.linear( + x_48, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_48 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_50 = torch.nn.functional.dropout(x_49, 0.0, False, False) + x_49 = None + x_51 = x_46 + x_50 + x_46 = x_50 = None + layer_norm_9 = torch.nn.functional.layer_norm( + x_51, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_52 = torch._C._nn.linear( + layer_norm_9, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_9 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_53 = torch._C._nn.gelu(x_52, approximate="none") + x_52 = None + x_54 = torch.nn.functional.dropout(x_53, 0.0, False, False) + x_53 = None + x_55 = torch._C._nn.linear( + x_54, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_54 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_56 = torch.nn.functional.dropout(x_55, 0.0, False, False) + x_55 = None + x_57 = x_51 + x_56 + x_51 = x_56 = None + layer_norm_10 = torch.nn.functional.layer_norm( + x_57, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_20 = torch._C._nn.linear( + layer_norm_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_10 = linear_20.reshape(1, 197, 3, 6, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_7 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_7.reshape(1, 197, 384) + transpose_7 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + x_62 = x_57 + x_61 + x_57 = x_61 = None + layer_norm_11 = torch.nn.functional.layer_norm( + x_62, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_63 = torch._C._nn.linear( + layer_norm_11, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_11 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_64 = torch._C._nn.gelu(x_63, approximate="none") + x_63 = None + x_65 = torch.nn.functional.dropout(x_64, 0.0, False, False) + x_64 = None + x_66 = torch._C._nn.linear( + x_65, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_65 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_67 = torch.nn.functional.dropout(x_66, 0.0, False, False) + x_66 = None + x_68 = x_62 + x_67 + x_62 = x_67 = None + getitem_34 = x_13[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_11 = torch.nn.functional.layer_norm( + getitem_34, + (192,), + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_34 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_12 = torch._C._nn.gelu(input_11, approximate="none") + input_11 = None + input_13 = torch._C._nn.linear( + input_12, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_12 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_35 = x_68[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_14 = torch.nn.functional.layer_norm( + getitem_35, + (384,), + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_35 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_15 = torch._C._nn.gelu(input_14, approximate="none") + input_14 = None + input_16 = torch._C._nn.linear( + input_15, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_15 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_36 = x_68[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp = torch.cat((input_13, getitem_36), dim=1) + input_13 = getitem_36 = None + getitem_37 = tmp[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_14 = torch.nn.functional.layer_norm( + tmp, + (384,), + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_38 = layer_norm_14[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_26 = torch._C._nn.linear( + getitem_38, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_38 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_12 = linear_26.reshape(1, 1, 6, 64) + linear_26 = None + q_6 = reshape_12.permute(0, 2, 1, 3) + reshape_12 = None + linear_27 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_13 = linear_27.reshape(1, 197, 6, 64) + linear_27 = None + k_6 = reshape_13.permute(0, 2, 1, 3) + reshape_13 = None + linear_28 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_14 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_14 = linear_28.reshape(1, 197, 6, 64) + linear_28 = None + v_6 = reshape_14.permute(0, 2, 1, 3) + reshape_14 = None + transpose_8 = k_6.transpose(-2, -1) + k_6 = None + matmul = q_6 @ transpose_8 + q_6 = transpose_8 = None + attn = matmul * 0.125 + matmul = None + attn_1 = attn.softmax(dim=-1) + attn = None + attn_2 = torch.nn.functional.dropout(attn_1, 0.0, False, False) + attn_1 = None + matmul_1 = attn_2 @ v_6 + attn_2 = v_6 = None + transpose_9 = matmul_1.transpose(1, 2) + matmul_1 = None + x_69 = transpose_9.reshape(1, 1, 384) + transpose_9 = None + x_70 = torch._C._nn.linear( + x_69, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_69 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_71 = torch.nn.functional.dropout(x_70, 0.0, False, False) + x_70 = None + x_72 = getitem_37 + x_71 + getitem_37 = x_71 = None + getitem_39 = x_72[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_72 = None + input_17 = torch.nn.functional.layer_norm( + getitem_39, + (384,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_39 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_18 = torch._C._nn.gelu(input_17, approximate="none") + input_17 = None + input_19 = torch._C._nn.linear( + input_18, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_18 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_40 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_1 = torch.cat((input_19, getitem_40), dim=1) + input_19 = getitem_40 = None + getitem_41 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_13 = None + tmp_2 = torch.cat((input_16, getitem_41), dim=1) + input_16 = getitem_41 = None + getitem_42 = tmp_2[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_16 = torch.nn.functional.layer_norm( + tmp_2, + (192,), + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_2 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_43 = layer_norm_16[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_31 = torch._C._nn.linear( + getitem_43, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_43 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_16 = linear_31.reshape(1, 1, 6, 32) + linear_31 = None + q_7 = reshape_16.permute(0, 2, 1, 3) + reshape_16 = None + linear_32 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_17 = linear_32.reshape(1, 401, 6, 32) + linear_32 = None + k_7 = reshape_17.permute(0, 2, 1, 3) + reshape_17 = None + linear_33 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_16 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_18 = linear_33.reshape(1, 401, 6, 32) + linear_33 = None + v_7 = reshape_18.permute(0, 2, 1, 3) + reshape_18 = None + transpose_10 = k_7.transpose(-2, -1) + k_7 = None + matmul_2 = q_7 @ transpose_10 + q_7 = transpose_10 = None + attn_3 = matmul_2 * 0.1767766952966369 + matmul_2 = None + attn_4 = attn_3.softmax(dim=-1) + attn_3 = None + attn_5 = torch.nn.functional.dropout(attn_4, 0.0, False, False) + attn_4 = None + matmul_3 = attn_5 @ v_7 + attn_5 = v_7 = None + transpose_11 = matmul_3.transpose(1, 2) + matmul_3 = None + x_73 = transpose_11.reshape(1, 1, 192) + transpose_11 = None + x_74 = torch._C._nn.linear( + x_73, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_73 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_75 = torch.nn.functional.dropout(x_74, 0.0, False, False) + x_74 = None + x_76 = getitem_42 + x_75 + getitem_42 = x_75 = None + getitem_44 = x_76[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_76 = None + input_20 = torch.nn.functional.layer_norm( + getitem_44, + (192,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_44 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_21 = torch._C._nn.gelu(input_20, approximate="none") + input_20 = None + input_22 = torch._C._nn.linear( + input_21, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_21 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_45 = x_68[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_68 = None + tmp_3 = torch.cat((input_22, getitem_45), dim=1) + input_22 = getitem_45 = None + layer_norm_18 = torch.nn.functional.layer_norm( + tmp_1, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_18 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_20 = linear_36.reshape(1, 401, 3, 6, 32) + linear_36 = None + qkv_6 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_8 = unbind_6[0] + k_8 = unbind_6[1] + v_8 = unbind_6[2] + unbind_6 = None + x_77 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_12 = x_77.transpose(1, 2) + x_77 = None + x_78 = transpose_12.reshape(1, 401, 192) + transpose_12 = None + x_79 = torch._C._nn.linear( + x_78, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_78 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = tmp_1 + x_80 + tmp_1 = x_80 = None + layer_norm_19 = torch.nn.functional.layer_norm( + x_81, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_82 = torch._C._nn.linear( + layer_norm_19, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_19 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_83 = torch._C._nn.gelu(x_82, approximate="none") + x_82 = None + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = torch._C._nn.linear( + x_84, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_84 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_86 = torch.nn.functional.dropout(x_85, 0.0, False, False) + x_85 = None + x_87 = x_81 + x_86 + x_81 = x_86 = None + layer_norm_20 = torch.nn.functional.layer_norm( + tmp_3, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_20 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_22 = linear_40.reshape(1, 197, 3, 6, 64) + linear_40 = None + qkv_7 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_9 = unbind_7[0] + k_9 = unbind_7[1] + v_9 = unbind_7[2] + unbind_7 = None + x_88 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_13 = x_88.transpose(1, 2) + x_88 = None + x_89 = transpose_13.reshape(1, 197, 384) + transpose_13 = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_89 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = tmp_3 + x_91 + tmp_3 = x_91 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_92, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_93 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_21 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_94 = torch._C._nn.gelu(x_93, approximate="none") + x_93 = None + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = torch._C._nn.linear( + x_95, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_95 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_97 = torch.nn.functional.dropout(x_96, 0.0, False, False) + x_96 = None + x_98 = x_92 + x_97 + x_92 = x_97 = None + layer_norm_22 = torch.nn.functional.layer_norm( + x_98, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_22, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_22 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_24 = linear_44.reshape(1, 197, 3, 6, 64) + linear_44 = None + qkv_8 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_10 = unbind_8[0] + k_10 = unbind_8[1] + v_10 = unbind_8[2] + unbind_8 = None + x_99 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_14 = x_99.transpose(1, 2) + x_99 = None + x_100 = transpose_14.reshape(1, 197, 384) + transpose_14 = None + x_101 = torch._C._nn.linear( + x_100, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_100 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = x_98 + x_102 + x_98 = x_102 = None + layer_norm_23 = torch.nn.functional.layer_norm( + x_103, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_104 = torch._C._nn.linear( + layer_norm_23, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_23 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_105 = torch._C._nn.gelu(x_104, approximate="none") + x_104 = None + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch._C._nn.linear( + x_106, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_106 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_103 + x_108 + x_103 = x_108 = None + layer_norm_24 = torch.nn.functional.layer_norm( + x_109, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_24 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_26 = linear_48.reshape(1, 197, 3, 6, 64) + linear_48 = None + qkv_9 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_11 = unbind_9[0] + k_11 = unbind_9[1] + v_11 = unbind_9[2] + unbind_9 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_15 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_15.reshape(1, 197, 384) + transpose_15 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_109 + x_113 + x_109 = x_113 = None + layer_norm_25 = torch.nn.functional.layer_norm( + x_114, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_115 = torch._C._nn.linear( + layer_norm_25, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_25 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_116 = torch._C._nn.gelu(x_115, approximate="none") + x_115 = None + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = torch._C._nn.linear( + x_117, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_117 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_119 = torch.nn.functional.dropout(x_118, 0.0, False, False) + x_118 = None + x_120 = x_114 + x_119 + x_114 = x_119 = None + layer_norm_26 = torch.nn.functional.layer_norm( + x_120, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_52 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_26 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_28 = linear_52.reshape(1, 197, 3, 6, 64) + linear_52 = None + qkv_10 = reshape_28.permute(2, 0, 3, 1, 4) + reshape_28 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_12 = unbind_10[0] + k_12 = unbind_10[1] + v_12 = unbind_10[2] + unbind_10 = None + x_121 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_16 = x_121.transpose(1, 2) + x_121 = None + x_122 = transpose_16.reshape(1, 197, 384) + transpose_16 = None + x_123 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_122 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = x_120 + x_124 + x_120 = x_124 = None + layer_norm_27 = torch.nn.functional.layer_norm( + x_125, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_126 = torch._C._nn.linear( + layer_norm_27, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_27 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126, approximate="none") + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_128 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) + x_129 = None + x_131 = x_125 + x_130 + x_125 = x_130 = None + layer_norm_28 = torch.nn.functional.layer_norm( + x_131, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_28 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_30 = linear_56.reshape(1, 197, 3, 6, 64) + linear_56 = None + qkv_11 = reshape_30.permute(2, 0, 3, 1, 4) + reshape_30 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_13 = unbind_11[0] + k_13 = unbind_11[1] + v_13 = unbind_11[2] + unbind_11 = None + x_132 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_17 = x_132.transpose(1, 2) + x_132 = None + x_133 = transpose_17.reshape(1, 197, 384) + transpose_17 = None + x_134 = torch._C._nn.linear( + x_133, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_133 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = x_131 + x_135 + x_131 = x_135 = None + layer_norm_29 = torch.nn.functional.layer_norm( + x_136, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_137 = torch._C._nn.linear( + layer_norm_29, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_29 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137, approximate="none") + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = torch._C._nn.linear( + x_139, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_139 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_141 = torch.nn.functional.dropout(x_140, 0.0, False, False) + x_140 = None + x_142 = x_136 + x_141 + x_136 = x_141 = None + getitem_64 = x_87[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_23 = torch.nn.functional.layer_norm( + getitem_64, + (192,), + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_64 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_24 = torch._C._nn.gelu(input_23, approximate="none") + input_23 = None + input_25 = torch._C._nn.linear( + input_24, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_24 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_65 = x_142[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_26 = torch.nn.functional.layer_norm( + getitem_65, + (384,), + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_65 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_27 = torch._C._nn.gelu(input_26, approximate="none") + input_26 = None + input_28 = torch._C._nn.linear( + input_27, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_27 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_66 = x_142[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_4 = torch.cat((input_25, getitem_66), dim=1) + input_25 = getitem_66 = None + getitem_67 = tmp_4[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_32 = torch.nn.functional.layer_norm( + tmp_4, + (384,), + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_4 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_68 = layer_norm_32[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_62 = torch._C._nn.linear( + getitem_68, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_68 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_32 = linear_62.reshape(1, 1, 6, 64) + linear_62 = None + q_14 = reshape_32.permute(0, 2, 1, 3) + reshape_32 = None + linear_63 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_33 = linear_63.reshape(1, 197, 6, 64) + linear_63 = None + k_14 = reshape_33.permute(0, 2, 1, 3) + reshape_33 = None + linear_64 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_32 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_34 = linear_64.reshape(1, 197, 6, 64) + linear_64 = None + v_14 = reshape_34.permute(0, 2, 1, 3) + reshape_34 = None + transpose_18 = k_14.transpose(-2, -1) + k_14 = None + matmul_4 = q_14 @ transpose_18 + q_14 = transpose_18 = None + attn_6 = matmul_4 * 0.125 + matmul_4 = None + attn_7 = attn_6.softmax(dim=-1) + attn_6 = None + attn_8 = torch.nn.functional.dropout(attn_7, 0.0, False, False) + attn_7 = None + matmul_5 = attn_8 @ v_14 + attn_8 = v_14 = None + transpose_19 = matmul_5.transpose(1, 2) + matmul_5 = None + x_143 = transpose_19.reshape(1, 1, 384) + transpose_19 = None + x_144 = torch._C._nn.linear( + x_143, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_143 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_145 = torch.nn.functional.dropout(x_144, 0.0, False, False) + x_144 = None + x_146 = getitem_67 + x_145 + getitem_67 = x_145 = None + getitem_69 = x_146[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_146 = None + input_29 = torch.nn.functional.layer_norm( + getitem_69, + (384,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_69 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_30 = torch._C._nn.gelu(input_29, approximate="none") + input_29 = None + input_31 = torch._C._nn.linear( + input_30, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_30 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_70 = x_87[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_5 = torch.cat((input_31, getitem_70), dim=1) + input_31 = getitem_70 = None + getitem_71 = x_87[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_87 = None + tmp_6 = torch.cat((input_28, getitem_71), dim=1) + input_28 = getitem_71 = None + getitem_72 = tmp_6[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_34 = torch.nn.functional.layer_norm( + tmp_6, + (192,), + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_6 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_73 = layer_norm_34[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_67 = torch._C._nn.linear( + getitem_73, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_73 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_36 = linear_67.reshape(1, 1, 6, 32) + linear_67 = None + q_15 = reshape_36.permute(0, 2, 1, 3) + reshape_36 = None + linear_68 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_37 = linear_68.reshape(1, 401, 6, 32) + linear_68 = None + k_15 = reshape_37.permute(0, 2, 1, 3) + reshape_37 = None + linear_69 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_34 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_38 = linear_69.reshape(1, 401, 6, 32) + linear_69 = None + v_15 = reshape_38.permute(0, 2, 1, 3) + reshape_38 = None + transpose_20 = k_15.transpose(-2, -1) + k_15 = None + matmul_6 = q_15 @ transpose_20 + q_15 = transpose_20 = None + attn_9 = matmul_6 * 0.1767766952966369 + matmul_6 = None + attn_10 = attn_9.softmax(dim=-1) + attn_9 = None + attn_11 = torch.nn.functional.dropout(attn_10, 0.0, False, False) + attn_10 = None + matmul_7 = attn_11 @ v_15 + attn_11 = v_15 = None + transpose_21 = matmul_7.transpose(1, 2) + matmul_7 = None + x_147 = transpose_21.reshape(1, 1, 192) + transpose_21 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_147 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = getitem_72 + x_149 + getitem_72 = x_149 = None + getitem_74 = x_150[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_150 = None + input_32 = torch.nn.functional.layer_norm( + getitem_74, + (192,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_74 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_33 = torch._C._nn.gelu(input_32, approximate="none") + input_32 = None + input_34 = torch._C._nn.linear( + input_33, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_33 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_75 = x_142[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_142 = None + tmp_7 = torch.cat((input_34, getitem_75), dim=1) + input_34 = getitem_75 = None + layer_norm_36 = torch.nn.functional.layer_norm( + tmp_5, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_72 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_36 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_40 = linear_72.reshape(1, 401, 3, 6, 32) + linear_72 = None + qkv_12 = reshape_40.permute(2, 0, 3, 1, 4) + reshape_40 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_16 = unbind_12[0] + k_16 = unbind_12[1] + v_16 = unbind_12[2] + unbind_12 = None + x_151 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, attn_mask=None, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_22 = x_151.transpose(1, 2) + x_151 = None + x_152 = transpose_22.reshape(1, 401, 192) + transpose_22 = None + x_153 = torch._C._nn.linear( + x_152, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_152 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_154 = torch.nn.functional.dropout(x_153, 0.0, False, False) + x_153 = None + x_155 = tmp_5 + x_154 + tmp_5 = x_154 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_155, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_156 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_37 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_157 = torch._C._nn.gelu(x_156, approximate="none") + x_156 = None + x_158 = torch.nn.functional.dropout(x_157, 0.0, False, False) + x_157 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_155 + x_160 + x_155 = x_160 = None + layer_norm_38 = torch.nn.functional.layer_norm( + tmp_7, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_76 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_38 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_42 = linear_76.reshape(1, 197, 3, 6, 64) + linear_76 = None + qkv_13 = reshape_42.permute(2, 0, 3, 1, 4) + reshape_42 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_17 = unbind_13[0] + k_17 = unbind_13[1] + v_17 = unbind_13[2] + unbind_13 = None + x_162 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, attn_mask=None, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_23 = x_162.transpose(1, 2) + x_162 = None + x_163 = transpose_23.reshape(1, 197, 384) + transpose_23 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_163 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_165 = torch.nn.functional.dropout(x_164, 0.0, False, False) + x_164 = None + x_166 = tmp_7 + x_165 + tmp_7 = x_165 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_166, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_167 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_39 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_168 = torch._C._nn.gelu(x_167, approximate="none") + x_167 = None + x_169 = torch.nn.functional.dropout(x_168, 0.0, False, False) + x_168 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_166 + x_171 + x_166 = x_171 = None + layer_norm_40 = torch.nn.functional.layer_norm( + x_172, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_40 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_44 = linear_80.reshape(1, 197, 3, 6, 64) + linear_80 = None + qkv_14 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_18 = unbind_14[0] + k_18 = unbind_14[1] + v_18 = unbind_14[2] + unbind_14 = None + x_173 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_24 = x_173.transpose(1, 2) + x_173 = None + x_174 = transpose_24.reshape(1, 197, 384) + transpose_24 = None + x_175 = torch._C._nn.linear( + x_174, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_174 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = x_172 + x_176 + x_172 = x_176 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_177, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_178 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_41 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_179 = torch._C._nn.gelu(x_178, approximate="none") + x_178 = None + x_180 = torch.nn.functional.dropout(x_179, 0.0, False, False) + x_179 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_177 + x_182 + x_177 = x_182 = None + layer_norm_42 = torch.nn.functional.layer_norm( + x_183, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_84 = torch._C._nn.linear( + layer_norm_42, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_42 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_46 = linear_84.reshape(1, 197, 3, 6, 64) + linear_84 = None + qkv_15 = reshape_46.permute(2, 0, 3, 1, 4) + reshape_46 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_19 = unbind_15[0] + k_19 = unbind_15[1] + v_19 = unbind_15[2] + unbind_15 = None + x_184 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, attn_mask=None, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_25 = x_184.transpose(1, 2) + x_184 = None + x_185 = transpose_25.reshape(1, 197, 384) + transpose_25 = None + x_186 = torch._C._nn.linear( + x_185, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_185 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_187 = torch.nn.functional.dropout(x_186, 0.0, False, False) + x_186 = None + x_188 = x_183 + x_187 + x_183 = x_187 = None + layer_norm_43 = torch.nn.functional.layer_norm( + x_188, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_189 = torch._C._nn.linear( + layer_norm_43, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_43 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_190 = torch._C._nn.gelu(x_189, approximate="none") + x_189 = None + x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) + x_190 = None + x_192 = torch._C._nn.linear( + x_191, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_191 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_188 + x_193 + x_188 = x_193 = None + layer_norm_44 = torch.nn.functional.layer_norm( + x_194, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_88 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_44 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_48 = linear_88.reshape(1, 197, 3, 6, 64) + linear_88 = None + qkv_16 = reshape_48.permute(2, 0, 3, 1, 4) + reshape_48 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_20 = unbind_16[0] + k_20 = unbind_16[1] + v_20 = unbind_16[2] + unbind_16 = None + x_195 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, attn_mask=None, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_26 = x_195.transpose(1, 2) + x_195 = None + x_196 = transpose_26.reshape(1, 197, 384) + transpose_26 = None + x_197 = torch._C._nn.linear( + x_196, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_196 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_198 = torch.nn.functional.dropout(x_197, 0.0, False, False) + x_197 = None + x_199 = x_194 + x_198 + x_194 = x_198 = None + layer_norm_45 = torch.nn.functional.layer_norm( + x_199, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_200 = torch._C._nn.linear( + layer_norm_45, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_45 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_201 = torch._C._nn.gelu(x_200, approximate="none") + x_200 = None + x_202 = torch.nn.functional.dropout(x_201, 0.0, False, False) + x_201 = None + x_203 = torch._C._nn.linear( + x_202, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_202 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_205 = x_199 + x_204 + x_199 = x_204 = None + layer_norm_46 = torch.nn.functional.layer_norm( + x_205, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_92 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_46 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_50 = linear_92.reshape(1, 197, 3, 6, 64) + linear_92 = None + qkv_17 = reshape_50.permute(2, 0, 3, 1, 4) + reshape_50 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_21 = unbind_17[0] + k_21 = unbind_17[1] + v_21 = unbind_17[2] + unbind_17 = None + x_206 = torch._C._nn.scaled_dot_product_attention( + q_21, k_21, v_21, attn_mask=None, dropout_p=0.0 + ) + q_21 = k_21 = v_21 = None + transpose_27 = x_206.transpose(1, 2) + x_206 = None + x_207 = transpose_27.reshape(1, 197, 384) + transpose_27 = None + x_208 = torch._C._nn.linear( + x_207, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_207 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_209 = torch.nn.functional.dropout(x_208, 0.0, False, False) + x_208 = None + x_210 = x_205 + x_209 + x_205 = x_209 = None + layer_norm_47 = torch.nn.functional.layer_norm( + x_210, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_211 = torch._C._nn.linear( + layer_norm_47, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_47 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_212 = torch._C._nn.gelu(x_211, approximate="none") + x_211 = None + x_213 = torch.nn.functional.dropout(x_212, 0.0, False, False) + x_212 = None + x_214 = torch._C._nn.linear( + x_213, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_213 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = x_210 + x_215 + x_210 = x_215 = None + getitem_94 = x_161[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_35 = torch.nn.functional.layer_norm( + getitem_94, + (192,), + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_94 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_36 = torch._C._nn.gelu(input_35, approximate="none") + input_35 = None + input_37 = torch._C._nn.linear( + input_36, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_36 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_95 = x_216[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_38 = torch.nn.functional.layer_norm( + getitem_95, + (384,), + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_95 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_39 = torch._C._nn.gelu(input_38, approximate="none") + input_38 = None + input_40 = torch._C._nn.linear( + input_39, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_39 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_96 = x_216[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_8 = torch.cat((input_37, getitem_96), dim=1) + input_37 = getitem_96 = None + getitem_97 = tmp_8[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_50 = torch.nn.functional.layer_norm( + tmp_8, + (384,), + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_8 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_98 = layer_norm_50[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_98 = torch._C._nn.linear( + getitem_98, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_98 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_52 = linear_98.reshape(1, 1, 6, 64) + linear_98 = None + q_22 = reshape_52.permute(0, 2, 1, 3) + reshape_52 = None + linear_99 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_53 = linear_99.reshape(1, 197, 6, 64) + linear_99 = None + k_22 = reshape_53.permute(0, 2, 1, 3) + reshape_53 = None + linear_100 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_50 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_54 = linear_100.reshape(1, 197, 6, 64) + linear_100 = None + v_22 = reshape_54.permute(0, 2, 1, 3) + reshape_54 = None + transpose_28 = k_22.transpose(-2, -1) + k_22 = None + matmul_8 = q_22 @ transpose_28 + q_22 = transpose_28 = None + attn_12 = matmul_8 * 0.125 + matmul_8 = None + attn_13 = attn_12.softmax(dim=-1) + attn_12 = None + attn_14 = torch.nn.functional.dropout(attn_13, 0.0, False, False) + attn_13 = None + matmul_9 = attn_14 @ v_22 + attn_14 = v_22 = None + transpose_29 = matmul_9.transpose(1, 2) + matmul_9 = None + x_217 = transpose_29.reshape(1, 1, 384) + transpose_29 = None + x_218 = torch._C._nn.linear( + x_217, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_217 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_219 = torch.nn.functional.dropout(x_218, 0.0, False, False) + x_218 = None + x_220 = getitem_97 + x_219 + getitem_97 = x_219 = None + getitem_99 = x_220[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_220 = None + input_41 = torch.nn.functional.layer_norm( + getitem_99, + (384,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_99 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_42 = torch._C._nn.gelu(input_41, approximate="none") + input_41 = None + input_43 = torch._C._nn.linear( + input_42, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_42 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_100 = x_161[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_9 = torch.cat((input_43, getitem_100), dim=1) + input_43 = getitem_100 = None + getitem_101 = x_161[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_161 = None + tmp_10 = torch.cat((input_40, getitem_101), dim=1) + input_40 = getitem_101 = None + getitem_102 = tmp_10[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_52 = torch.nn.functional.layer_norm( + tmp_10, + (192,), + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_10 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_103 = layer_norm_52[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_103 = torch._C._nn.linear( + getitem_103, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_103 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_56 = linear_103.reshape(1, 1, 6, 32) + linear_103 = None + q_23 = reshape_56.permute(0, 2, 1, 3) + reshape_56 = None + linear_104 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_57 = linear_104.reshape(1, 401, 6, 32) + linear_104 = None + k_23 = reshape_57.permute(0, 2, 1, 3) + reshape_57 = None + linear_105 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_52 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_58 = linear_105.reshape(1, 401, 6, 32) + linear_105 = None + v_23 = reshape_58.permute(0, 2, 1, 3) + reshape_58 = None + transpose_30 = k_23.transpose(-2, -1) + k_23 = None + matmul_10 = q_23 @ transpose_30 + q_23 = transpose_30 = None + attn_15 = matmul_10 * 0.1767766952966369 + matmul_10 = None + attn_16 = attn_15.softmax(dim=-1) + attn_15 = None + attn_17 = torch.nn.functional.dropout(attn_16, 0.0, False, False) + attn_16 = None + matmul_11 = attn_17 @ v_23 + attn_17 = v_23 = None + transpose_31 = matmul_11.transpose(1, 2) + matmul_11 = None + x_221 = transpose_31.reshape(1, 1, 192) + transpose_31 = None + x_222 = torch._C._nn.linear( + x_221, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_221 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_223 = torch.nn.functional.dropout(x_222, 0.0, False, False) + x_222 = None + x_224 = getitem_102 + x_223 + getitem_102 = x_223 = None + getitem_104 = x_224[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_224 = None + input_44 = torch.nn.functional.layer_norm( + getitem_104, + (192,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_104 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_45 = torch._C._nn.gelu(input_44, approximate="none") + input_44 = None + input_46 = torch._C._nn.linear( + input_45, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_45 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_105 = x_216[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_216 = None + tmp_11 = torch.cat((input_46, getitem_105), dim=1) + input_46 = getitem_105 = None + x_225 = torch.nn.functional.layer_norm( + tmp_9, + (192,), + l_self_modules_norm_modules_0_parameters_weight_, + l_self_modules_norm_modules_0_parameters_bias_, + 1e-06, + ) + tmp_9 = ( + l_self_modules_norm_modules_0_parameters_weight_ + ) = l_self_modules_norm_modules_0_parameters_bias_ = None + x_226 = torch.nn.functional.layer_norm( + tmp_11, + (384,), + l_self_modules_norm_modules_1_parameters_weight_, + l_self_modules_norm_modules_1_parameters_bias_, + 1e-06, + ) + tmp_11 = ( + l_self_modules_norm_modules_1_parameters_weight_ + ) = l_self_modules_norm_modules_1_parameters_bias_ = None + x_227 = x_225[(slice(None, None, None), 0)] + x_225 = None + x_228 = x_226[(slice(None, None, None), 0)] + x_226 = None + dropout_68 = torch.nn.functional.dropout(x_227, 0.0, False, False) + x_227 = None + dropout_69 = torch.nn.functional.dropout(x_228, 0.0, False, False) + x_228 = None + linear_108 = torch._C._nn.linear( + dropout_68, + l_self_modules_head_modules_0_parameters_weight_, + l_self_modules_head_modules_0_parameters_bias_, + ) + dropout_68 = ( + l_self_modules_head_modules_0_parameters_weight_ + ) = l_self_modules_head_modules_0_parameters_bias_ = None + linear_109 = torch._C._nn.linear( + dropout_69, + l_self_modules_head_modules_1_parameters_weight_, + l_self_modules_head_modules_1_parameters_bias_, + ) + dropout_69 = ( + l_self_modules_head_modules_1_parameters_weight_ + ) = l_self_modules_head_modules_1_parameters_bias_ = None + stack = torch.stack([linear_108, linear_109], dim=0) + linear_108 = linear_109 = None + x_229 = torch.mean(stack, dim=0) + stack = None + return (x_229,) diff --git a/samples/timm/crossvit_15_dagger_240.in1k/weight_meta.py b/samples/timm/crossvit_15_dagger_240.in1k/weight_meta.py new file mode 100644 index 0000000000..175c745a8c --- /dev/null +++ b/samples/timm/crossvit_15_dagger_240.in1k/weight_meta.py @@ -0,0 +1,3512 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.222 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_" + ) + shape = [48, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_" + ) + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.010 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_" + ) + shape = [96, 48, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.028 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.004 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_" + ) + shape = [192, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_0_: + name = "L_self_parameters_cls_token_0_" + shape = [1, 1, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_0_: + name = "L_self_parameters_pos_embed_0_" + shape = [1, 401, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_" + ) + shape = [96, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.007 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_" + ) + shape = [192, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_" + ) + shape = [384, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.014 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.014 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_1_: + name = "L_self_parameters_cls_token_1_" + shape = [1, 1, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_1_: + name = "L_self_parameters_pos_embed_1_" + shape = [1, 197, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_weight_: + name = "L_self_modules_norm_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_bias_: + name = "L_self_modules_norm_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_weight_: + name = "L_self_modules_norm_modules_1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_bias_: + name = "L_self_modules_norm_modules_1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_weight_: + name = "L_self_modules_head_modules_0_parameters_weight_" + shape = [1000, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_bias_: + name = "L_self_modules_head_modules_0_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_weight_: + name = "L_self_modules_head_modules_1_parameters_weight_" + shape = [1000, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_bias_: + name = "L_self_modules_head_modules_1_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/crossvit_15_dagger_408.in1k/graph_hash.txt b/samples/timm/crossvit_15_dagger_408.in1k/graph_hash.txt new file mode 100644 index 0000000000..49fae451f2 --- /dev/null +++ b/samples/timm/crossvit_15_dagger_408.in1k/graph_hash.txt @@ -0,0 +1 @@ +66d5c1cd52f9c8e484bc9cce56fd0cfdb5cd0a275d6b9efd65469dda69dd7143 \ No newline at end of file diff --git a/samples/timm/crossvit_15_dagger_408.in1k/graph_net.json b/samples/timm/crossvit_15_dagger_408.in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/crossvit_15_dagger_408.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/crossvit_15_dagger_408.in1k/input_meta.py b/samples/timm/crossvit_15_dagger_408.in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/crossvit_15_dagger_408.in1k/input_tensor_constraints.py b/samples/timm/crossvit_15_dagger_408.in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/crossvit_15_dagger_408.in1k/model.py b/samples/timm/crossvit_15_dagger_408.in1k/model.py new file mode 100644 index 0000000000..b648a7ea02 --- /dev/null +++ b/samples/timm/crossvit_15_dagger_408.in1k/model.py @@ -0,0 +1,2814 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_0_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_0_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_1_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ + ) + l_self_parameters_cls_token_0_ = L_self_parameters_cls_token_0_ + l_self_parameters_pos_embed_0_ = L_self_parameters_pos_embed_0_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ + ) + l_self_parameters_cls_token_1_ = L_self_parameters_cls_token_1_ + l_self_parameters_pos_embed_1_ = L_self_parameters_pos_embed_1_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_norm_modules_0_parameters_weight_ = ( + L_self_modules_norm_modules_0_parameters_weight_ + ) + l_self_modules_norm_modules_0_parameters_bias_ = ( + L_self_modules_norm_modules_0_parameters_bias_ + ) + l_self_modules_norm_modules_1_parameters_weight_ = ( + L_self_modules_norm_modules_1_parameters_weight_ + ) + l_self_modules_norm_modules_1_parameters_bias_ = ( + L_self_modules_norm_modules_1_parameters_bias_ + ) + l_self_modules_head_modules_0_parameters_weight_ = ( + L_self_modules_head_modules_0_parameters_weight_ + ) + l_self_modules_head_modules_0_parameters_bias_ = ( + L_self_modules_head_modules_0_parameters_bias_ + ) + l_self_modules_head_modules_1_parameters_weight_ = ( + L_self_modules_head_modules_1_parameters_weight_ + ) + l_self_modules_head_modules_1_parameters_bias_ = ( + L_self_modules_head_modules_1_parameters_bias_ + ) + x = torch.nn.functional.interpolate( + l_x_, size=(408, 408), mode="bicubic", align_corners=False + ) + input_1 = torch.conv2d( + x, + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_, + (4, 4), + (3, 3), + (1, 1), + 1, + ) + x = l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ + ) = None + input_2 = torch.nn.functional.relu(input_1, inplace=True) + input_1 = None + input_3 = torch.conv2d( + input_2, + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_, + (3, 3), + (0, 0), + (1, 1), + 1, + ) + input_2 = l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ + ) = None + input_4 = torch.nn.functional.relu(input_3, inplace=True) + input_3 = None + input_5 = torch.conv2d( + input_4, + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + input_4 = l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ + ) = None + flatten = input_5.flatten(2) + input_5 = None + x_1 = flatten.transpose(1, 2) + flatten = None + cls_tokens = l_self_parameters_cls_token_0_.expand(1, -1, -1) + l_self_parameters_cls_token_0_ = None + x_ = torch.cat((cls_tokens, x_1), dim=1) + cls_tokens = x_1 = None + x__1 = x_ + l_self_parameters_pos_embed_0_ + x_ = l_self_parameters_pos_embed_0_ = None + x__2 = torch.nn.functional.dropout(x__1, 0.0, False, False) + x__1 = None + x_2 = torch.nn.functional.interpolate( + l_x_, size=(384, 384), mode="bicubic", align_corners=False + ) + l_x_ = None + input_6 = torch.conv2d( + x_2, + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_, + (4, 4), + (3, 3), + (1, 1), + 1, + ) + x_2 = l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ + ) = None + input_7 = torch.nn.functional.relu(input_6, inplace=True) + input_6 = None + input_8 = torch.conv2d( + input_7, + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_7 = l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ + ) = None + input_9 = torch.nn.functional.relu(input_8, inplace=True) + input_8 = None + input_10 = torch.conv2d( + input_9, + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_9 = l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ + ) = None + flatten_1 = input_10.flatten(2) + input_10 = None + x_3 = flatten_1.transpose(1, 2) + flatten_1 = None + cls_tokens_1 = l_self_parameters_cls_token_1_.expand(1, -1, -1) + l_self_parameters_cls_token_1_ = None + x__3 = torch.cat((cls_tokens_1, x_3), dim=1) + cls_tokens_1 = x_3 = None + x__4 = x__3 + l_self_parameters_pos_embed_1_ + x__3 = l_self_parameters_pos_embed_1_ = None + x__5 = torch.nn.functional.dropout(x__4, 0.0, False, False) + x__4 = None + layer_norm = torch.nn.functional.layer_norm( + x__2, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear = torch._C._nn.linear( + layer_norm, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 1157, 3, 6, 32) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_4 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_2 = x_4.transpose(1, 2) + x_4 = None + x_5 = transpose_2.reshape(1, 1157, 192) + transpose_2 = None + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_7 = torch.nn.functional.dropout(x_6, 0.0, False, False) + x_6 = None + x_8 = x__2 + x_7 + x__2 = x_7 = None + layer_norm_1 = torch.nn.functional.layer_norm( + x_8, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_9 = torch._C._nn.linear( + layer_norm_1, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_1 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_10 = torch._C._nn.gelu(x_9, approximate="none") + x_9 = None + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_11 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_13 = torch.nn.functional.dropout(x_12, 0.0, False, False) + x_12 = None + x_14 = x_8 + x_13 + x_8 = x_13 = None + layer_norm_2 = torch.nn.functional.layer_norm( + x__5, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_2, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_2 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_2 = linear_4.reshape(1, 577, 3, 6, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_15 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_3 = x_15.transpose(1, 2) + x_15 = None + x_16 = transpose_3.reshape(1, 577, 384) + transpose_3 = None + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_16 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_18 = torch.nn.functional.dropout(x_17, 0.0, False, False) + x_17 = None + x_19 = x__5 + x_18 + x__5 = x_18 = None + layer_norm_3 = torch.nn.functional.layer_norm( + x_19, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_20 = torch._C._nn.linear( + layer_norm_3, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_3 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_21 = torch._C._nn.gelu(x_20, approximate="none") + x_20 = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = torch._C._nn.linear( + x_22, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_22 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_24 = torch.nn.functional.dropout(x_23, 0.0, False, False) + x_23 = None + x_25 = x_19 + x_24 + x_19 = x_24 = None + layer_norm_4 = torch.nn.functional.layer_norm( + x_25, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_8.reshape(1, 577, 3, 6, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_26 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_4 = x_26.transpose(1, 2) + x_26 = None + x_27 = transpose_4.reshape(1, 577, 384) + transpose_4 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_27 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + x_30 = x_25 + x_29 + x_25 = x_29 = None + layer_norm_5 = torch.nn.functional.layer_norm( + x_30, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_31 = torch._C._nn.linear( + layer_norm_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_32 = torch._C._nn.gelu(x_31, approximate="none") + x_31 = None + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_33 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + x_36 = x_30 + x_35 + x_30 = x_35 = None + layer_norm_6 = torch.nn.functional.layer_norm( + x_36, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_6, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_6 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_6 = linear_12.reshape(1, 577, 3, 6, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_37 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_5 = x_37.transpose(1, 2) + x_37 = None + x_38 = transpose_5.reshape(1, 577, 384) + transpose_5 = None + x_39 = torch._C._nn.linear( + x_38, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_38 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = x_36 + x_40 + x_36 = x_40 = None + layer_norm_7 = torch.nn.functional.layer_norm( + x_41, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + layer_norm_7, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_7 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42, approximate="none") + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_45 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_44 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_46 = torch.nn.functional.dropout(x_45, 0.0, False, False) + x_45 = None + x_47 = x_41 + x_46 + x_41 = x_46 = None + layer_norm_8 = torch.nn.functional.layer_norm( + x_47, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_8, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_8 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_8 = linear_16.reshape(1, 577, 3, 6, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_48 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_6 = x_48.transpose(1, 2) + x_48 = None + x_49 = transpose_6.reshape(1, 577, 384) + transpose_6 = None + x_50 = torch._C._nn.linear( + x_49, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_49 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) + x_50 = None + x_52 = x_47 + x_51 + x_47 = x_51 = None + layer_norm_9 = torch.nn.functional.layer_norm( + x_52, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + layer_norm_9, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_9 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53, approximate="none") + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = torch._C._nn.linear( + x_55, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_55 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_57 = torch.nn.functional.dropout(x_56, 0.0, False, False) + x_56 = None + x_58 = x_52 + x_57 + x_52 = x_57 = None + layer_norm_10 = torch.nn.functional.layer_norm( + x_58, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_20 = torch._C._nn.linear( + layer_norm_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_10 = linear_20.reshape(1, 577, 3, 6, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_59 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_7 = x_59.transpose(1, 2) + x_59 = None + x_60 = transpose_7.reshape(1, 577, 384) + transpose_7 = None + x_61 = torch._C._nn.linear( + x_60, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_60 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + x_63 = x_58 + x_62 + x_58 = x_62 = None + layer_norm_11 = torch.nn.functional.layer_norm( + x_63, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + layer_norm_11, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_11 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_63 + x_68 + x_63 = x_68 = None + getitem_34 = x_14[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_11 = torch.nn.functional.layer_norm( + getitem_34, + (192,), + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_34 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_12 = torch._C._nn.gelu(input_11, approximate="none") + input_11 = None + input_13 = torch._C._nn.linear( + input_12, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_12 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_35 = x_69[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_14 = torch.nn.functional.layer_norm( + getitem_35, + (384,), + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_35 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_15 = torch._C._nn.gelu(input_14, approximate="none") + input_14 = None + input_16 = torch._C._nn.linear( + input_15, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_15 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_36 = x_69[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp = torch.cat((input_13, getitem_36), dim=1) + input_13 = getitem_36 = None + getitem_37 = tmp[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_14 = torch.nn.functional.layer_norm( + tmp, + (384,), + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_38 = layer_norm_14[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_26 = torch._C._nn.linear( + getitem_38, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_38 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_12 = linear_26.reshape(1, 1, 6, 64) + linear_26 = None + q_6 = reshape_12.permute(0, 2, 1, 3) + reshape_12 = None + linear_27 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_13 = linear_27.reshape(1, 577, 6, 64) + linear_27 = None + k_6 = reshape_13.permute(0, 2, 1, 3) + reshape_13 = None + linear_28 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_14 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_14 = linear_28.reshape(1, 577, 6, 64) + linear_28 = None + v_6 = reshape_14.permute(0, 2, 1, 3) + reshape_14 = None + transpose_8 = k_6.transpose(-2, -1) + k_6 = None + matmul = q_6 @ transpose_8 + q_6 = transpose_8 = None + attn = matmul * 0.125 + matmul = None + attn_1 = attn.softmax(dim=-1) + attn = None + attn_2 = torch.nn.functional.dropout(attn_1, 0.0, False, False) + attn_1 = None + matmul_1 = attn_2 @ v_6 + attn_2 = v_6 = None + transpose_9 = matmul_1.transpose(1, 2) + matmul_1 = None + x_70 = transpose_9.reshape(1, 1, 384) + transpose_9 = None + x_71 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_70 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + x_73 = getitem_37 + x_72 + getitem_37 = x_72 = None + getitem_39 = x_73[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_73 = None + input_17 = torch.nn.functional.layer_norm( + getitem_39, + (384,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_39 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_18 = torch._C._nn.gelu(input_17, approximate="none") + input_17 = None + input_19 = torch._C._nn.linear( + input_18, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_18 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_40 = x_14[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_1 = torch.cat((input_19, getitem_40), dim=1) + input_19 = getitem_40 = None + getitem_41 = x_14[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_14 = None + tmp_2 = torch.cat((input_16, getitem_41), dim=1) + input_16 = getitem_41 = None + getitem_42 = tmp_2[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_16 = torch.nn.functional.layer_norm( + tmp_2, + (192,), + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_2 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_43 = layer_norm_16[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_31 = torch._C._nn.linear( + getitem_43, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_43 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_16 = linear_31.reshape(1, 1, 6, 32) + linear_31 = None + q_7 = reshape_16.permute(0, 2, 1, 3) + reshape_16 = None + linear_32 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_17 = linear_32.reshape(1, 1157, 6, 32) + linear_32 = None + k_7 = reshape_17.permute(0, 2, 1, 3) + reshape_17 = None + linear_33 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_16 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_18 = linear_33.reshape(1, 1157, 6, 32) + linear_33 = None + v_7 = reshape_18.permute(0, 2, 1, 3) + reshape_18 = None + transpose_10 = k_7.transpose(-2, -1) + k_7 = None + matmul_2 = q_7 @ transpose_10 + q_7 = transpose_10 = None + attn_3 = matmul_2 * 0.1767766952966369 + matmul_2 = None + attn_4 = attn_3.softmax(dim=-1) + attn_3 = None + attn_5 = torch.nn.functional.dropout(attn_4, 0.0, False, False) + attn_4 = None + matmul_3 = attn_5 @ v_7 + attn_5 = v_7 = None + transpose_11 = matmul_3.transpose(1, 2) + matmul_3 = None + x_74 = transpose_11.reshape(1, 1, 192) + transpose_11 = None + x_75 = torch._C._nn.linear( + x_74, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_74 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_76 = torch.nn.functional.dropout(x_75, 0.0, False, False) + x_75 = None + x_77 = getitem_42 + x_76 + getitem_42 = x_76 = None + getitem_44 = x_77[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_77 = None + input_20 = torch.nn.functional.layer_norm( + getitem_44, + (192,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_44 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_21 = torch._C._nn.gelu(input_20, approximate="none") + input_20 = None + input_22 = torch._C._nn.linear( + input_21, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_21 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_45 = x_69[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_69 = None + tmp_3 = torch.cat((input_22, getitem_45), dim=1) + input_22 = getitem_45 = None + layer_norm_18 = torch.nn.functional.layer_norm( + tmp_1, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_18 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_20 = linear_36.reshape(1, 1157, 3, 6, 32) + linear_36 = None + qkv_6 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_8 = unbind_6[0] + k_8 = unbind_6[1] + v_8 = unbind_6[2] + unbind_6 = None + x_78 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_12 = x_78.transpose(1, 2) + x_78 = None + x_79 = transpose_12.reshape(1, 1157, 192) + transpose_12 = None + x_80 = torch._C._nn.linear( + x_79, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_79 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = tmp_1 + x_81 + tmp_1 = x_81 = None + layer_norm_19 = torch.nn.functional.layer_norm( + x_82, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_83 = torch._C._nn.linear( + layer_norm_19, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_19 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_84 = torch._C._nn.gelu(x_83, approximate="none") + x_83 = None + x_85 = torch.nn.functional.dropout(x_84, 0.0, False, False) + x_84 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_85 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + x_88 = x_82 + x_87 + x_82 = x_87 = None + layer_norm_20 = torch.nn.functional.layer_norm( + tmp_3, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_20 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_22 = linear_40.reshape(1, 577, 3, 6, 64) + linear_40 = None + qkv_7 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_9 = unbind_7[0] + k_9 = unbind_7[1] + v_9 = unbind_7[2] + unbind_7 = None + x_89 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_13 = x_89.transpose(1, 2) + x_89 = None + x_90 = transpose_13.reshape(1, 577, 384) + transpose_13 = None + x_91 = torch._C._nn.linear( + x_90, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_90 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = tmp_3 + x_92 + tmp_3 = x_92 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_93, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_94 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_21 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_95 = torch._C._nn.gelu(x_94, approximate="none") + x_94 = None + x_96 = torch.nn.functional.dropout(x_95, 0.0, False, False) + x_95 = None + x_97 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_96 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_98 = torch.nn.functional.dropout(x_97, 0.0, False, False) + x_97 = None + x_99 = x_93 + x_98 + x_93 = x_98 = None + layer_norm_22 = torch.nn.functional.layer_norm( + x_99, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_22, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_22 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_24 = linear_44.reshape(1, 577, 3, 6, 64) + linear_44 = None + qkv_8 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_10 = unbind_8[0] + k_10 = unbind_8[1] + v_10 = unbind_8[2] + unbind_8 = None + x_100 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_14 = x_100.transpose(1, 2) + x_100 = None + x_101 = transpose_14.reshape(1, 577, 384) + transpose_14 = None + x_102 = torch._C._nn.linear( + x_101, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_101 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_104 = x_99 + x_103 + x_99 = x_103 = None + layer_norm_23 = torch.nn.functional.layer_norm( + x_104, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_105 = torch._C._nn.linear( + layer_norm_23, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_23 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_106 = torch._C._nn.gelu(x_105, approximate="none") + x_105 = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + x_108 = torch._C._nn.linear( + x_107, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_107 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_109 = torch.nn.functional.dropout(x_108, 0.0, False, False) + x_108 = None + x_110 = x_104 + x_109 + x_104 = x_109 = None + layer_norm_24 = torch.nn.functional.layer_norm( + x_110, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_24 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_26 = linear_48.reshape(1, 577, 3, 6, 64) + linear_48 = None + qkv_9 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_11 = unbind_9[0] + k_11 = unbind_9[1] + v_11 = unbind_9[2] + unbind_9 = None + x_111 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_15 = x_111.transpose(1, 2) + x_111 = None + x_112 = transpose_15.reshape(1, 577, 384) + transpose_15 = None + x_113 = torch._C._nn.linear( + x_112, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_112 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_115 = x_110 + x_114 + x_110 = x_114 = None + layer_norm_25 = torch.nn.functional.layer_norm( + x_115, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_116 = torch._C._nn.linear( + layer_norm_25, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_25 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + x_121 = x_115 + x_120 + x_115 = x_120 = None + layer_norm_26 = torch.nn.functional.layer_norm( + x_121, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_52 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_26 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_28 = linear_52.reshape(1, 577, 3, 6, 64) + linear_52 = None + qkv_10 = reshape_28.permute(2, 0, 3, 1, 4) + reshape_28 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_12 = unbind_10[0] + k_12 = unbind_10[1] + v_12 = unbind_10[2] + unbind_10 = None + x_122 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_16 = x_122.transpose(1, 2) + x_122 = None + x_123 = transpose_16.reshape(1, 577, 384) + transpose_16 = None + x_124 = torch._C._nn.linear( + x_123, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_123 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_126 = x_121 + x_125 + x_121 = x_125 = None + layer_norm_27 = torch.nn.functional.layer_norm( + x_126, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_127 = torch._C._nn.linear( + layer_norm_27, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_27 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_128 = torch._C._nn.gelu(x_127, approximate="none") + x_127 = None + x_129 = torch.nn.functional.dropout(x_128, 0.0, False, False) + x_128 = None + x_130 = torch._C._nn.linear( + x_129, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_129 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = x_126 + x_131 + x_126 = x_131 = None + layer_norm_28 = torch.nn.functional.layer_norm( + x_132, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_28 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_30 = linear_56.reshape(1, 577, 3, 6, 64) + linear_56 = None + qkv_11 = reshape_30.permute(2, 0, 3, 1, 4) + reshape_30 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_13 = unbind_11[0] + k_13 = unbind_11[1] + v_13 = unbind_11[2] + unbind_11 = None + x_133 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_17 = x_133.transpose(1, 2) + x_133 = None + x_134 = transpose_17.reshape(1, 577, 384) + transpose_17 = None + x_135 = torch._C._nn.linear( + x_134, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_134 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + x_137 = x_132 + x_136 + x_132 = x_136 = None + layer_norm_29 = torch.nn.functional.layer_norm( + x_137, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_138 = torch._C._nn.linear( + layer_norm_29, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_29 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_139 = torch._C._nn.gelu(x_138, approximate="none") + x_138 = None + x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) + x_139 = None + x_141 = torch._C._nn.linear( + x_140, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_140 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_137 + x_142 + x_137 = x_142 = None + getitem_64 = x_88[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_23 = torch.nn.functional.layer_norm( + getitem_64, + (192,), + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_64 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_24 = torch._C._nn.gelu(input_23, approximate="none") + input_23 = None + input_25 = torch._C._nn.linear( + input_24, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_24 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_65 = x_143[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_26 = torch.nn.functional.layer_norm( + getitem_65, + (384,), + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_65 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_27 = torch._C._nn.gelu(input_26, approximate="none") + input_26 = None + input_28 = torch._C._nn.linear( + input_27, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_27 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_66 = x_143[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_4 = torch.cat((input_25, getitem_66), dim=1) + input_25 = getitem_66 = None + getitem_67 = tmp_4[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_32 = torch.nn.functional.layer_norm( + tmp_4, + (384,), + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_4 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_68 = layer_norm_32[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_62 = torch._C._nn.linear( + getitem_68, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_68 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_32 = linear_62.reshape(1, 1, 6, 64) + linear_62 = None + q_14 = reshape_32.permute(0, 2, 1, 3) + reshape_32 = None + linear_63 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_33 = linear_63.reshape(1, 577, 6, 64) + linear_63 = None + k_14 = reshape_33.permute(0, 2, 1, 3) + reshape_33 = None + linear_64 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_32 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_34 = linear_64.reshape(1, 577, 6, 64) + linear_64 = None + v_14 = reshape_34.permute(0, 2, 1, 3) + reshape_34 = None + transpose_18 = k_14.transpose(-2, -1) + k_14 = None + matmul_4 = q_14 @ transpose_18 + q_14 = transpose_18 = None + attn_6 = matmul_4 * 0.125 + matmul_4 = None + attn_7 = attn_6.softmax(dim=-1) + attn_6 = None + attn_8 = torch.nn.functional.dropout(attn_7, 0.0, False, False) + attn_7 = None + matmul_5 = attn_8 @ v_14 + attn_8 = v_14 = None + transpose_19 = matmul_5.transpose(1, 2) + matmul_5 = None + x_144 = transpose_19.reshape(1, 1, 384) + transpose_19 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_144 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = getitem_67 + x_146 + getitem_67 = x_146 = None + getitem_69 = x_147[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_147 = None + input_29 = torch.nn.functional.layer_norm( + getitem_69, + (384,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_69 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_30 = torch._C._nn.gelu(input_29, approximate="none") + input_29 = None + input_31 = torch._C._nn.linear( + input_30, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_30 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_70 = x_88[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_5 = torch.cat((input_31, getitem_70), dim=1) + input_31 = getitem_70 = None + getitem_71 = x_88[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_88 = None + tmp_6 = torch.cat((input_28, getitem_71), dim=1) + input_28 = getitem_71 = None + getitem_72 = tmp_6[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_34 = torch.nn.functional.layer_norm( + tmp_6, + (192,), + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_6 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_73 = layer_norm_34[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_67 = torch._C._nn.linear( + getitem_73, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_73 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_36 = linear_67.reshape(1, 1, 6, 32) + linear_67 = None + q_15 = reshape_36.permute(0, 2, 1, 3) + reshape_36 = None + linear_68 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_37 = linear_68.reshape(1, 1157, 6, 32) + linear_68 = None + k_15 = reshape_37.permute(0, 2, 1, 3) + reshape_37 = None + linear_69 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_34 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_38 = linear_69.reshape(1, 1157, 6, 32) + linear_69 = None + v_15 = reshape_38.permute(0, 2, 1, 3) + reshape_38 = None + transpose_20 = k_15.transpose(-2, -1) + k_15 = None + matmul_6 = q_15 @ transpose_20 + q_15 = transpose_20 = None + attn_9 = matmul_6 * 0.1767766952966369 + matmul_6 = None + attn_10 = attn_9.softmax(dim=-1) + attn_9 = None + attn_11 = torch.nn.functional.dropout(attn_10, 0.0, False, False) + attn_10 = None + matmul_7 = attn_11 @ v_15 + attn_11 = v_15 = None + transpose_21 = matmul_7.transpose(1, 2) + matmul_7 = None + x_148 = transpose_21.reshape(1, 1, 192) + transpose_21 = None + x_149 = torch._C._nn.linear( + x_148, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_148 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) + x_149 = None + x_151 = getitem_72 + x_150 + getitem_72 = x_150 = None + getitem_74 = x_151[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_151 = None + input_32 = torch.nn.functional.layer_norm( + getitem_74, + (192,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_74 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_33 = torch._C._nn.gelu(input_32, approximate="none") + input_32 = None + input_34 = torch._C._nn.linear( + input_33, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_33 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_75 = x_143[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_143 = None + tmp_7 = torch.cat((input_34, getitem_75), dim=1) + input_34 = getitem_75 = None + layer_norm_36 = torch.nn.functional.layer_norm( + tmp_5, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_72 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_36 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_40 = linear_72.reshape(1, 1157, 3, 6, 32) + linear_72 = None + qkv_12 = reshape_40.permute(2, 0, 3, 1, 4) + reshape_40 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_16 = unbind_12[0] + k_16 = unbind_12[1] + v_16 = unbind_12[2] + unbind_12 = None + x_152 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, attn_mask=None, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_22 = x_152.transpose(1, 2) + x_152 = None + x_153 = transpose_22.reshape(1, 1157, 192) + transpose_22 = None + x_154 = torch._C._nn.linear( + x_153, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_153 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_155 = torch.nn.functional.dropout(x_154, 0.0, False, False) + x_154 = None + x_156 = tmp_5 + x_155 + tmp_5 = x_155 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_156, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_157 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_37 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_158 = torch._C._nn.gelu(x_157, approximate="none") + x_157 = None + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = torch._C._nn.linear( + x_159, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_159 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_161 = torch.nn.functional.dropout(x_160, 0.0, False, False) + x_160 = None + x_162 = x_156 + x_161 + x_156 = x_161 = None + layer_norm_38 = torch.nn.functional.layer_norm( + tmp_7, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_76 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_38 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_42 = linear_76.reshape(1, 577, 3, 6, 64) + linear_76 = None + qkv_13 = reshape_42.permute(2, 0, 3, 1, 4) + reshape_42 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_17 = unbind_13[0] + k_17 = unbind_13[1] + v_17 = unbind_13[2] + unbind_13 = None + x_163 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, attn_mask=None, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_23 = x_163.transpose(1, 2) + x_163 = None + x_164 = transpose_23.reshape(1, 577, 384) + transpose_23 = None + x_165 = torch._C._nn.linear( + x_164, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_164 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_166 = torch.nn.functional.dropout(x_165, 0.0, False, False) + x_165 = None + x_167 = tmp_7 + x_166 + tmp_7 = x_166 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_167, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_168 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_39 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_169 = torch._C._nn.gelu(x_168, approximate="none") + x_168 = None + x_170 = torch.nn.functional.dropout(x_169, 0.0, False, False) + x_169 = None + x_171 = torch._C._nn.linear( + x_170, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_170 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_172 = torch.nn.functional.dropout(x_171, 0.0, False, False) + x_171 = None + x_173 = x_167 + x_172 + x_167 = x_172 = None + layer_norm_40 = torch.nn.functional.layer_norm( + x_173, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_40 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_44 = linear_80.reshape(1, 577, 3, 6, 64) + linear_80 = None + qkv_14 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_18 = unbind_14[0] + k_18 = unbind_14[1] + v_18 = unbind_14[2] + unbind_14 = None + x_174 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_24 = x_174.transpose(1, 2) + x_174 = None + x_175 = transpose_24.reshape(1, 577, 384) + transpose_24 = None + x_176 = torch._C._nn.linear( + x_175, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_175 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_177 = torch.nn.functional.dropout(x_176, 0.0, False, False) + x_176 = None + x_178 = x_173 + x_177 + x_173 = x_177 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_178, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_179 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_41 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_180 = torch._C._nn.gelu(x_179, approximate="none") + x_179 = None + x_181 = torch.nn.functional.dropout(x_180, 0.0, False, False) + x_180 = None + x_182 = torch._C._nn.linear( + x_181, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_181 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_183 = torch.nn.functional.dropout(x_182, 0.0, False, False) + x_182 = None + x_184 = x_178 + x_183 + x_178 = x_183 = None + layer_norm_42 = torch.nn.functional.layer_norm( + x_184, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_84 = torch._C._nn.linear( + layer_norm_42, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_42 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_46 = linear_84.reshape(1, 577, 3, 6, 64) + linear_84 = None + qkv_15 = reshape_46.permute(2, 0, 3, 1, 4) + reshape_46 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_19 = unbind_15[0] + k_19 = unbind_15[1] + v_19 = unbind_15[2] + unbind_15 = None + x_185 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, attn_mask=None, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_25 = x_185.transpose(1, 2) + x_185 = None + x_186 = transpose_25.reshape(1, 577, 384) + transpose_25 = None + x_187 = torch._C._nn.linear( + x_186, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_186 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_188 = torch.nn.functional.dropout(x_187, 0.0, False, False) + x_187 = None + x_189 = x_184 + x_188 + x_184 = x_188 = None + layer_norm_43 = torch.nn.functional.layer_norm( + x_189, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_190 = torch._C._nn.linear( + layer_norm_43, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_43 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_191 = torch._C._nn.gelu(x_190, approximate="none") + x_190 = None + x_192 = torch.nn.functional.dropout(x_191, 0.0, False, False) + x_191 = None + x_193 = torch._C._nn.linear( + x_192, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_192 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_194 = torch.nn.functional.dropout(x_193, 0.0, False, False) + x_193 = None + x_195 = x_189 + x_194 + x_189 = x_194 = None + layer_norm_44 = torch.nn.functional.layer_norm( + x_195, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_88 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_44 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_48 = linear_88.reshape(1, 577, 3, 6, 64) + linear_88 = None + qkv_16 = reshape_48.permute(2, 0, 3, 1, 4) + reshape_48 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_20 = unbind_16[0] + k_20 = unbind_16[1] + v_20 = unbind_16[2] + unbind_16 = None + x_196 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, attn_mask=None, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_26 = x_196.transpose(1, 2) + x_196 = None + x_197 = transpose_26.reshape(1, 577, 384) + transpose_26 = None + x_198 = torch._C._nn.linear( + x_197, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_197 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_199 = torch.nn.functional.dropout(x_198, 0.0, False, False) + x_198 = None + x_200 = x_195 + x_199 + x_195 = x_199 = None + layer_norm_45 = torch.nn.functional.layer_norm( + x_200, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_201 = torch._C._nn.linear( + layer_norm_45, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_45 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_202 = torch._C._nn.gelu(x_201, approximate="none") + x_201 = None + x_203 = torch.nn.functional.dropout(x_202, 0.0, False, False) + x_202 = None + x_204 = torch._C._nn.linear( + x_203, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_203 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_205 = torch.nn.functional.dropout(x_204, 0.0, False, False) + x_204 = None + x_206 = x_200 + x_205 + x_200 = x_205 = None + layer_norm_46 = torch.nn.functional.layer_norm( + x_206, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_92 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_46 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_50 = linear_92.reshape(1, 577, 3, 6, 64) + linear_92 = None + qkv_17 = reshape_50.permute(2, 0, 3, 1, 4) + reshape_50 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_21 = unbind_17[0] + k_21 = unbind_17[1] + v_21 = unbind_17[2] + unbind_17 = None + x_207 = torch._C._nn.scaled_dot_product_attention( + q_21, k_21, v_21, attn_mask=None, dropout_p=0.0 + ) + q_21 = k_21 = v_21 = None + transpose_27 = x_207.transpose(1, 2) + x_207 = None + x_208 = transpose_27.reshape(1, 577, 384) + transpose_27 = None + x_209 = torch._C._nn.linear( + x_208, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_208 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) + x_209 = None + x_211 = x_206 + x_210 + x_206 = x_210 = None + layer_norm_47 = torch.nn.functional.layer_norm( + x_211, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_212 = torch._C._nn.linear( + layer_norm_47, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_47 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_213 = torch._C._nn.gelu(x_212, approximate="none") + x_212 = None + x_214 = torch.nn.functional.dropout(x_213, 0.0, False, False) + x_213 = None + x_215 = torch._C._nn.linear( + x_214, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_214 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_216 = torch.nn.functional.dropout(x_215, 0.0, False, False) + x_215 = None + x_217 = x_211 + x_216 + x_211 = x_216 = None + getitem_94 = x_162[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_35 = torch.nn.functional.layer_norm( + getitem_94, + (192,), + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_94 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_36 = torch._C._nn.gelu(input_35, approximate="none") + input_35 = None + input_37 = torch._C._nn.linear( + input_36, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_36 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_95 = x_217[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_38 = torch.nn.functional.layer_norm( + getitem_95, + (384,), + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_95 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_39 = torch._C._nn.gelu(input_38, approximate="none") + input_38 = None + input_40 = torch._C._nn.linear( + input_39, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_39 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_96 = x_217[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_8 = torch.cat((input_37, getitem_96), dim=1) + input_37 = getitem_96 = None + getitem_97 = tmp_8[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_50 = torch.nn.functional.layer_norm( + tmp_8, + (384,), + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_8 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_98 = layer_norm_50[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_98 = torch._C._nn.linear( + getitem_98, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_98 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_52 = linear_98.reshape(1, 1, 6, 64) + linear_98 = None + q_22 = reshape_52.permute(0, 2, 1, 3) + reshape_52 = None + linear_99 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_53 = linear_99.reshape(1, 577, 6, 64) + linear_99 = None + k_22 = reshape_53.permute(0, 2, 1, 3) + reshape_53 = None + linear_100 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_50 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_54 = linear_100.reshape(1, 577, 6, 64) + linear_100 = None + v_22 = reshape_54.permute(0, 2, 1, 3) + reshape_54 = None + transpose_28 = k_22.transpose(-2, -1) + k_22 = None + matmul_8 = q_22 @ transpose_28 + q_22 = transpose_28 = None + attn_12 = matmul_8 * 0.125 + matmul_8 = None + attn_13 = attn_12.softmax(dim=-1) + attn_12 = None + attn_14 = torch.nn.functional.dropout(attn_13, 0.0, False, False) + attn_13 = None + matmul_9 = attn_14 @ v_22 + attn_14 = v_22 = None + transpose_29 = matmul_9.transpose(1, 2) + matmul_9 = None + x_218 = transpose_29.reshape(1, 1, 384) + transpose_29 = None + x_219 = torch._C._nn.linear( + x_218, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_218 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_220 = torch.nn.functional.dropout(x_219, 0.0, False, False) + x_219 = None + x_221 = getitem_97 + x_220 + getitem_97 = x_220 = None + getitem_99 = x_221[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_221 = None + input_41 = torch.nn.functional.layer_norm( + getitem_99, + (384,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_99 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_42 = torch._C._nn.gelu(input_41, approximate="none") + input_41 = None + input_43 = torch._C._nn.linear( + input_42, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_42 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_100 = x_162[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_9 = torch.cat((input_43, getitem_100), dim=1) + input_43 = getitem_100 = None + getitem_101 = x_162[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_162 = None + tmp_10 = torch.cat((input_40, getitem_101), dim=1) + input_40 = getitem_101 = None + getitem_102 = tmp_10[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_52 = torch.nn.functional.layer_norm( + tmp_10, + (192,), + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_10 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_103 = layer_norm_52[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_103 = torch._C._nn.linear( + getitem_103, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_103 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_56 = linear_103.reshape(1, 1, 6, 32) + linear_103 = None + q_23 = reshape_56.permute(0, 2, 1, 3) + reshape_56 = None + linear_104 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_57 = linear_104.reshape(1, 1157, 6, 32) + linear_104 = None + k_23 = reshape_57.permute(0, 2, 1, 3) + reshape_57 = None + linear_105 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_52 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_58 = linear_105.reshape(1, 1157, 6, 32) + linear_105 = None + v_23 = reshape_58.permute(0, 2, 1, 3) + reshape_58 = None + transpose_30 = k_23.transpose(-2, -1) + k_23 = None + matmul_10 = q_23 @ transpose_30 + q_23 = transpose_30 = None + attn_15 = matmul_10 * 0.1767766952966369 + matmul_10 = None + attn_16 = attn_15.softmax(dim=-1) + attn_15 = None + attn_17 = torch.nn.functional.dropout(attn_16, 0.0, False, False) + attn_16 = None + matmul_11 = attn_17 @ v_23 + attn_17 = v_23 = None + transpose_31 = matmul_11.transpose(1, 2) + matmul_11 = None + x_222 = transpose_31.reshape(1, 1, 192) + transpose_31 = None + x_223 = torch._C._nn.linear( + x_222, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_222 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_224 = torch.nn.functional.dropout(x_223, 0.0, False, False) + x_223 = None + x_225 = getitem_102 + x_224 + getitem_102 = x_224 = None + getitem_104 = x_225[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_225 = None + input_44 = torch.nn.functional.layer_norm( + getitem_104, + (192,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_104 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_45 = torch._C._nn.gelu(input_44, approximate="none") + input_44 = None + input_46 = torch._C._nn.linear( + input_45, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_45 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_105 = x_217[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_217 = None + tmp_11 = torch.cat((input_46, getitem_105), dim=1) + input_46 = getitem_105 = None + x_226 = torch.nn.functional.layer_norm( + tmp_9, + (192,), + l_self_modules_norm_modules_0_parameters_weight_, + l_self_modules_norm_modules_0_parameters_bias_, + 1e-06, + ) + tmp_9 = ( + l_self_modules_norm_modules_0_parameters_weight_ + ) = l_self_modules_norm_modules_0_parameters_bias_ = None + x_227 = torch.nn.functional.layer_norm( + tmp_11, + (384,), + l_self_modules_norm_modules_1_parameters_weight_, + l_self_modules_norm_modules_1_parameters_bias_, + 1e-06, + ) + tmp_11 = ( + l_self_modules_norm_modules_1_parameters_weight_ + ) = l_self_modules_norm_modules_1_parameters_bias_ = None + x_228 = x_226[(slice(None, None, None), 0)] + x_226 = None + x_229 = x_227[(slice(None, None, None), 0)] + x_227 = None + dropout_68 = torch.nn.functional.dropout(x_228, 0.0, False, False) + x_228 = None + dropout_69 = torch.nn.functional.dropout(x_229, 0.0, False, False) + x_229 = None + linear_108 = torch._C._nn.linear( + dropout_68, + l_self_modules_head_modules_0_parameters_weight_, + l_self_modules_head_modules_0_parameters_bias_, + ) + dropout_68 = ( + l_self_modules_head_modules_0_parameters_weight_ + ) = l_self_modules_head_modules_0_parameters_bias_ = None + linear_109 = torch._C._nn.linear( + dropout_69, + l_self_modules_head_modules_1_parameters_weight_, + l_self_modules_head_modules_1_parameters_bias_, + ) + dropout_69 = ( + l_self_modules_head_modules_1_parameters_weight_ + ) = l_self_modules_head_modules_1_parameters_bias_ = None + stack = torch.stack([linear_108, linear_109], dim=0) + linear_108 = linear_109 = None + x_230 = torch.mean(stack, dim=0) + stack = None + return (x_230,) diff --git a/samples/timm/crossvit_15_dagger_408.in1k/weight_meta.py b/samples/timm/crossvit_15_dagger_408.in1k/weight_meta.py new file mode 100644 index 0000000000..980a0d7bcf --- /dev/null +++ b/samples/timm/crossvit_15_dagger_408.in1k/weight_meta.py @@ -0,0 +1,3522 @@ +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.226 + std = 1.288 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_" + ) + shape = [48, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_" + ) + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.008 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_" + ) + shape = [96, 48, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.028 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.027 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_" + ) + shape = [192, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_0_: + name = "L_self_parameters_cls_token_0_" + shape = [1, 1, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_0_: + name = "L_self_parameters_pos_embed_0_" + shape = [1, 1157, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_" + ) + shape = [96, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_" + ) + shape = [192, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_" + ) + shape = [384, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.014 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.014 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_1_: + name = "L_self_parameters_cls_token_1_" + shape = [1, 1, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_1_: + name = "L_self_parameters_pos_embed_1_" + shape = [1, 577, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 576] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_weight_: + name = "L_self_modules_norm_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_bias_: + name = "L_self_modules_norm_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_weight_: + name = "L_self_modules_norm_modules_1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_bias_: + name = "L_self_modules_norm_modules_1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_weight_: + name = "L_self_modules_head_modules_0_parameters_weight_" + shape = [1000, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_bias_: + name = "L_self_modules_head_modules_0_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_weight_: + name = "L_self_modules_head_modules_1_parameters_weight_" + shape = [1000, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_bias_: + name = "L_self_modules_head_modules_1_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/crossvit_18_240.in1k/graph_hash.txt b/samples/timm/crossvit_18_240.in1k/graph_hash.txt new file mode 100644 index 0000000000..b989878424 --- /dev/null +++ b/samples/timm/crossvit_18_240.in1k/graph_hash.txt @@ -0,0 +1 @@ +191d9cd47d0904f532b3160647b8b1caaa2bf2c6c8c73b9a6b64dbe3c5d6e6c2 \ No newline at end of file diff --git a/samples/timm/crossvit_18_240.in1k/graph_net.json b/samples/timm/crossvit_18_240.in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/crossvit_18_240.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/crossvit_18_240.in1k/input_meta.py b/samples/timm/crossvit_18_240.in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/crossvit_18_240.in1k/input_tensor_constraints.py b/samples/timm/crossvit_18_240.in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/crossvit_18_240.in1k/model.py b/samples/timm/crossvit_18_240.in1k/model.py new file mode 100644 index 0000000000..b5b7817d62 --- /dev/null +++ b/samples/timm/crossvit_18_240.in1k/model.py @@ -0,0 +1,3015 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_0_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_0_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_1_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ + ) + l_self_parameters_cls_token_0_ = L_self_parameters_cls_token_0_ + l_self_parameters_pos_embed_0_ = L_self_parameters_pos_embed_0_ + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ + ) + l_self_parameters_cls_token_1_ = L_self_parameters_cls_token_1_ + l_self_parameters_pos_embed_1_ = L_self_parameters_pos_embed_1_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_norm_modules_0_parameters_weight_ = ( + L_self_modules_norm_modules_0_parameters_weight_ + ) + l_self_modules_norm_modules_0_parameters_bias_ = ( + L_self_modules_norm_modules_0_parameters_bias_ + ) + l_self_modules_norm_modules_1_parameters_weight_ = ( + L_self_modules_norm_modules_1_parameters_weight_ + ) + l_self_modules_norm_modules_1_parameters_bias_ = ( + L_self_modules_norm_modules_1_parameters_bias_ + ) + l_self_modules_head_modules_0_parameters_weight_ = ( + L_self_modules_head_modules_0_parameters_weight_ + ) + l_self_modules_head_modules_0_parameters_bias_ = ( + L_self_modules_head_modules_0_parameters_bias_ + ) + l_self_modules_head_modules_1_parameters_weight_ = ( + L_self_modules_head_modules_1_parameters_weight_ + ) + l_self_modules_head_modules_1_parameters_bias_ = ( + L_self_modules_head_modules_1_parameters_bias_ + ) + x = torch.nn.functional.interpolate( + l_x_, size=(240, 240), mode="bicubic", align_corners=False + ) + conv2d = torch.conv2d( + x, + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_, + (12, 12), + (0, 0), + (1, 1), + 1, + ) + x = ( + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ = None + flatten = conv2d.flatten(2) + conv2d = None + x_1 = flatten.transpose(1, 2) + flatten = None + cls_tokens = l_self_parameters_cls_token_0_.expand(1, -1, -1) + l_self_parameters_cls_token_0_ = None + x_ = torch.cat((cls_tokens, x_1), dim=1) + cls_tokens = x_1 = None + x__1 = x_ + l_self_parameters_pos_embed_0_ + x_ = l_self_parameters_pos_embed_0_ = None + x__2 = torch.nn.functional.dropout(x__1, 0.0, False, False) + x__1 = None + conv2d_1 = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ = None + flatten_1 = conv2d_1.flatten(2) + conv2d_1 = None + x_2 = flatten_1.transpose(1, 2) + flatten_1 = None + cls_tokens_1 = l_self_parameters_cls_token_1_.expand(1, -1, -1) + l_self_parameters_cls_token_1_ = None + x__3 = torch.cat((cls_tokens_1, x_2), dim=1) + cls_tokens_1 = x_2 = None + x__4 = x__3 + l_self_parameters_pos_embed_1_ + x__3 = l_self_parameters_pos_embed_1_ = None + x__5 = torch.nn.functional.dropout(x__4, 0.0, False, False) + x__4 = None + layer_norm = torch.nn.functional.layer_norm( + x__2, + (224,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear = torch._C._nn.linear( + layer_norm, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 401, 3, 7, 32) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_3 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_2 = x_3.transpose(1, 2) + x_3 = None + x_4 = transpose_2.reshape(1, 401, 224) + transpose_2 = None + x_5 = torch._C._nn.linear( + x_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_6 = torch.nn.functional.dropout(x_5, 0.0, False, False) + x_5 = None + x_7 = x__2 + x_6 + x__2 = x_6 = None + layer_norm_1 = torch.nn.functional.layer_norm( + x_7, + (224,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_8 = torch._C._nn.linear( + layer_norm_1, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_1 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_9 = torch._C._nn.gelu(x_8, approximate="none") + x_8 = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = torch._C._nn.linear( + x_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_7 + x_12 + x_7 = x_12 = None + layer_norm_2 = torch.nn.functional.layer_norm( + x__5, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_2, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_2 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_2 = linear_4.reshape(1, 197, 3, 7, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_14 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_3 = x_14.transpose(1, 2) + x_14 = None + x_15 = transpose_3.reshape(1, 197, 448) + transpose_3 = None + x_16 = torch._C._nn.linear( + x_15, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_15 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = x__5 + x_17 + x__5 = x_17 = None + layer_norm_3 = torch.nn.functional.layer_norm( + x_18, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_19 = torch._C._nn.linear( + layer_norm_3, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_3 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_20 = torch._C._nn.gelu(x_19, approximate="none") + x_19 = None + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = torch._C._nn.linear( + x_21, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_21 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_18 + x_23 + x_18 = x_23 = None + layer_norm_4 = torch.nn.functional.layer_norm( + x_24, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_8.reshape(1, 197, 3, 7, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_25 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_4 = x_25.transpose(1, 2) + x_25 = None + x_26 = transpose_4.reshape(1, 197, 448) + transpose_4 = None + x_27 = torch._C._nn.linear( + x_26, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_26 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_28 = torch.nn.functional.dropout(x_27, 0.0, False, False) + x_27 = None + x_29 = x_24 + x_28 + x_24 = x_28 = None + layer_norm_5 = torch.nn.functional.layer_norm( + x_29, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_30 = torch._C._nn.linear( + layer_norm_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_31 = torch._C._nn.gelu(x_30, approximate="none") + x_30 = None + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = torch._C._nn.linear( + x_32, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_32 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = x_29 + x_34 + x_29 = x_34 = None + layer_norm_6 = torch.nn.functional.layer_norm( + x_35, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_6, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_6 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_6 = linear_12.reshape(1, 197, 3, 7, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_36 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_5 = x_36.transpose(1, 2) + x_36 = None + x_37 = transpose_5.reshape(1, 197, 448) + transpose_5 = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_37 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_39 = torch.nn.functional.dropout(x_38, 0.0, False, False) + x_38 = None + x_40 = x_35 + x_39 + x_35 = x_39 = None + layer_norm_7 = torch.nn.functional.layer_norm( + x_40, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_41 = torch._C._nn.linear( + layer_norm_7, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_7 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_42 = torch._C._nn.gelu(x_41, approximate="none") + x_41 = None + x_43 = torch.nn.functional.dropout(x_42, 0.0, False, False) + x_42 = None + x_44 = torch._C._nn.linear( + x_43, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_43 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = x_40 + x_45 + x_40 = x_45 = None + layer_norm_8 = torch.nn.functional.layer_norm( + x_46, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_8, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_8 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_8 = linear_16.reshape(1, 197, 3, 7, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_47 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_6 = x_47.transpose(1, 2) + x_47 = None + x_48 = transpose_6.reshape(1, 197, 448) + transpose_6 = None + x_49 = torch._C._nn.linear( + x_48, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_48 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_50 = torch.nn.functional.dropout(x_49, 0.0, False, False) + x_49 = None + x_51 = x_46 + x_50 + x_46 = x_50 = None + layer_norm_9 = torch.nn.functional.layer_norm( + x_51, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_52 = torch._C._nn.linear( + layer_norm_9, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_9 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_53 = torch._C._nn.gelu(x_52, approximate="none") + x_52 = None + x_54 = torch.nn.functional.dropout(x_53, 0.0, False, False) + x_53 = None + x_55 = torch._C._nn.linear( + x_54, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_54 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_56 = torch.nn.functional.dropout(x_55, 0.0, False, False) + x_55 = None + x_57 = x_51 + x_56 + x_51 = x_56 = None + layer_norm_10 = torch.nn.functional.layer_norm( + x_57, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_20 = torch._C._nn.linear( + layer_norm_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_10 = linear_20.reshape(1, 197, 3, 7, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_7 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_7.reshape(1, 197, 448) + transpose_7 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + x_62 = x_57 + x_61 + x_57 = x_61 = None + layer_norm_11 = torch.nn.functional.layer_norm( + x_62, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_63 = torch._C._nn.linear( + layer_norm_11, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_11 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_64 = torch._C._nn.gelu(x_63, approximate="none") + x_63 = None + x_65 = torch.nn.functional.dropout(x_64, 0.0, False, False) + x_64 = None + x_66 = torch._C._nn.linear( + x_65, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_65 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_67 = torch.nn.functional.dropout(x_66, 0.0, False, False) + x_66 = None + x_68 = x_62 + x_67 + x_62 = x_67 = None + layer_norm_12 = torch.nn.functional.layer_norm( + x_68, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = (None) + linear_24 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_12 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_12 = linear_24.reshape(1, 197, 3, 7, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_69 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_8 = x_69.transpose(1, 2) + x_69 = None + x_70 = transpose_8.reshape(1, 197, 448) + transpose_8 = None + x_71 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_70 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + x_73 = x_68 + x_72 + x_68 = x_72 = None + layer_norm_13 = torch.nn.functional.layer_norm( + x_73, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = (None) + x_74 = torch._C._nn.linear( + layer_norm_13, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_13 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_75 = torch._C._nn.gelu(x_74, approximate="none") + x_74 = None + x_76 = torch.nn.functional.dropout(x_75, 0.0, False, False) + x_75 = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_76 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_78 = torch.nn.functional.dropout(x_77, 0.0, False, False) + x_77 = None + x_79 = x_73 + x_78 + x_73 = x_78 = None + getitem_37 = x_13[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_1 = torch.nn.functional.layer_norm( + getitem_37, + (224,), + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_37 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_2 = torch._C._nn.gelu(input_1, approximate="none") + input_1 = None + input_3 = torch._C._nn.linear( + input_2, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_2 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_38 = x_79[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_4 = torch.nn.functional.layer_norm( + getitem_38, + (448,), + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_38 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_5 = torch._C._nn.gelu(input_4, approximate="none") + input_4 = None + input_6 = torch._C._nn.linear( + input_5, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_5 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_39 = x_79[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp = torch.cat((input_3, getitem_39), dim=1) + input_3 = getitem_39 = None + getitem_40 = tmp[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_16 = torch.nn.functional.layer_norm( + tmp, + (448,), + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_41 = layer_norm_16[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_30 = torch._C._nn.linear( + getitem_41, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_41 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_14 = linear_30.reshape(1, 1, 7, 64) + linear_30 = None + q_7 = reshape_14.permute(0, 2, 1, 3) + reshape_14 = None + linear_31 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_15 = linear_31.reshape(1, 197, 7, 64) + linear_31 = None + k_7 = reshape_15.permute(0, 2, 1, 3) + reshape_15 = None + linear_32 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_16 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_16 = linear_32.reshape(1, 197, 7, 64) + linear_32 = None + v_7 = reshape_16.permute(0, 2, 1, 3) + reshape_16 = None + transpose_9 = k_7.transpose(-2, -1) + k_7 = None + matmul = q_7 @ transpose_9 + q_7 = transpose_9 = None + attn = matmul * 0.125 + matmul = None + attn_1 = attn.softmax(dim=-1) + attn = None + attn_2 = torch.nn.functional.dropout(attn_1, 0.0, False, False) + attn_1 = None + matmul_1 = attn_2 @ v_7 + attn_2 = v_7 = None + transpose_10 = matmul_1.transpose(1, 2) + matmul_1 = None + x_80 = transpose_10.reshape(1, 1, 448) + transpose_10 = None + x_81 = torch._C._nn.linear( + x_80, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_80 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_82 = torch.nn.functional.dropout(x_81, 0.0, False, False) + x_81 = None + x_83 = getitem_40 + x_82 + getitem_40 = x_82 = None + getitem_42 = x_83[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_83 = None + input_7 = torch.nn.functional.layer_norm( + getitem_42, + (448,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_42 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_8 = torch._C._nn.gelu(input_7, approximate="none") + input_7 = None + input_9 = torch._C._nn.linear( + input_8, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_8 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_43 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_1 = torch.cat((input_9, getitem_43), dim=1) + input_9 = getitem_43 = None + getitem_44 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_13 = None + tmp_2 = torch.cat((input_6, getitem_44), dim=1) + input_6 = getitem_44 = None + getitem_45 = tmp_2[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_18 = torch.nn.functional.layer_norm( + tmp_2, + (224,), + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_2 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_46 = layer_norm_18[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_35 = torch._C._nn.linear( + getitem_46, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_46 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_18 = linear_35.reshape(1, 1, 7, 32) + linear_35 = None + q_8 = reshape_18.permute(0, 2, 1, 3) + reshape_18 = None + linear_36 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_19 = linear_36.reshape(1, 401, 7, 32) + linear_36 = None + k_8 = reshape_19.permute(0, 2, 1, 3) + reshape_19 = None + linear_37 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_18 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_20 = linear_37.reshape(1, 401, 7, 32) + linear_37 = None + v_8 = reshape_20.permute(0, 2, 1, 3) + reshape_20 = None + transpose_11 = k_8.transpose(-2, -1) + k_8 = None + matmul_2 = q_8 @ transpose_11 + q_8 = transpose_11 = None + attn_3 = matmul_2 * 0.1767766952966369 + matmul_2 = None + attn_4 = attn_3.softmax(dim=-1) + attn_3 = None + attn_5 = torch.nn.functional.dropout(attn_4, 0.0, False, False) + attn_4 = None + matmul_3 = attn_5 @ v_8 + attn_5 = v_8 = None + transpose_12 = matmul_3.transpose(1, 2) + matmul_3 = None + x_84 = transpose_12.reshape(1, 1, 224) + transpose_12 = None + x_85 = torch._C._nn.linear( + x_84, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_84 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_86 = torch.nn.functional.dropout(x_85, 0.0, False, False) + x_85 = None + x_87 = getitem_45 + x_86 + getitem_45 = x_86 = None + getitem_47 = x_87[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_87 = None + input_10 = torch.nn.functional.layer_norm( + getitem_47, + (224,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_47 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_11 = torch._C._nn.gelu(input_10, approximate="none") + input_10 = None + input_12 = torch._C._nn.linear( + input_11, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_11 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_48 = x_79[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_79 = None + tmp_3 = torch.cat((input_12, getitem_48), dim=1) + input_12 = getitem_48 = None + layer_norm_20 = torch.nn.functional.layer_norm( + tmp_1, + (224,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_20 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_22 = linear_40.reshape(1, 401, 3, 7, 32) + linear_40 = None + qkv_7 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_9 = unbind_7[0] + k_9 = unbind_7[1] + v_9 = unbind_7[2] + unbind_7 = None + x_88 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_13 = x_88.transpose(1, 2) + x_88 = None + x_89 = transpose_13.reshape(1, 401, 224) + transpose_13 = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_89 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = tmp_1 + x_91 + tmp_1 = x_91 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_92, + (224,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_93 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_21 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_94 = torch._C._nn.gelu(x_93, approximate="none") + x_93 = None + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = torch._C._nn.linear( + x_95, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_95 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_97 = torch.nn.functional.dropout(x_96, 0.0, False, False) + x_96 = None + x_98 = x_92 + x_97 + x_92 = x_97 = None + layer_norm_22 = torch.nn.functional.layer_norm( + tmp_3, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_22, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_22 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_24 = linear_44.reshape(1, 197, 3, 7, 64) + linear_44 = None + qkv_8 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_10 = unbind_8[0] + k_10 = unbind_8[1] + v_10 = unbind_8[2] + unbind_8 = None + x_99 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_14 = x_99.transpose(1, 2) + x_99 = None + x_100 = transpose_14.reshape(1, 197, 448) + transpose_14 = None + x_101 = torch._C._nn.linear( + x_100, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_100 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = tmp_3 + x_102 + tmp_3 = x_102 = None + layer_norm_23 = torch.nn.functional.layer_norm( + x_103, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_104 = torch._C._nn.linear( + layer_norm_23, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_23 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_105 = torch._C._nn.gelu(x_104, approximate="none") + x_104 = None + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch._C._nn.linear( + x_106, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_106 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_103 + x_108 + x_103 = x_108 = None + layer_norm_24 = torch.nn.functional.layer_norm( + x_109, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_24 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_26 = linear_48.reshape(1, 197, 3, 7, 64) + linear_48 = None + qkv_9 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_11 = unbind_9[0] + k_11 = unbind_9[1] + v_11 = unbind_9[2] + unbind_9 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_15 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_15.reshape(1, 197, 448) + transpose_15 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_109 + x_113 + x_109 = x_113 = None + layer_norm_25 = torch.nn.functional.layer_norm( + x_114, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_115 = torch._C._nn.linear( + layer_norm_25, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_25 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_116 = torch._C._nn.gelu(x_115, approximate="none") + x_115 = None + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = torch._C._nn.linear( + x_117, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_117 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_119 = torch.nn.functional.dropout(x_118, 0.0, False, False) + x_118 = None + x_120 = x_114 + x_119 + x_114 = x_119 = None + layer_norm_26 = torch.nn.functional.layer_norm( + x_120, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_52 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_26 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_28 = linear_52.reshape(1, 197, 3, 7, 64) + linear_52 = None + qkv_10 = reshape_28.permute(2, 0, 3, 1, 4) + reshape_28 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_12 = unbind_10[0] + k_12 = unbind_10[1] + v_12 = unbind_10[2] + unbind_10 = None + x_121 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_16 = x_121.transpose(1, 2) + x_121 = None + x_122 = transpose_16.reshape(1, 197, 448) + transpose_16 = None + x_123 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_122 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = x_120 + x_124 + x_120 = x_124 = None + layer_norm_27 = torch.nn.functional.layer_norm( + x_125, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_126 = torch._C._nn.linear( + layer_norm_27, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_27 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126, approximate="none") + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_128 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) + x_129 = None + x_131 = x_125 + x_130 + x_125 = x_130 = None + layer_norm_28 = torch.nn.functional.layer_norm( + x_131, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_28 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_30 = linear_56.reshape(1, 197, 3, 7, 64) + linear_56 = None + qkv_11 = reshape_30.permute(2, 0, 3, 1, 4) + reshape_30 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_13 = unbind_11[0] + k_13 = unbind_11[1] + v_13 = unbind_11[2] + unbind_11 = None + x_132 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_17 = x_132.transpose(1, 2) + x_132 = None + x_133 = transpose_17.reshape(1, 197, 448) + transpose_17 = None + x_134 = torch._C._nn.linear( + x_133, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_133 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = x_131 + x_135 + x_131 = x_135 = None + layer_norm_29 = torch.nn.functional.layer_norm( + x_136, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_137 = torch._C._nn.linear( + layer_norm_29, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_29 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137, approximate="none") + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = torch._C._nn.linear( + x_139, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_139 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_141 = torch.nn.functional.dropout(x_140, 0.0, False, False) + x_140 = None + x_142 = x_136 + x_141 + x_136 = x_141 = None + layer_norm_30 = torch.nn.functional.layer_norm( + x_142, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_60 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_30 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_32 = linear_60.reshape(1, 197, 3, 7, 64) + linear_60 = None + qkv_12 = reshape_32.permute(2, 0, 3, 1, 4) + reshape_32 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_14 = unbind_12[0] + k_14 = unbind_12[1] + v_14 = unbind_12[2] + unbind_12 = None + x_143 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_18 = x_143.transpose(1, 2) + x_143 = None + x_144 = transpose_18.reshape(1, 197, 448) + transpose_18 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_144 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = x_142 + x_146 + x_142 = x_146 = None + layer_norm_31 = torch.nn.functional.layer_norm( + x_147, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_148 = torch._C._nn.linear( + layer_norm_31, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_31 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_149 = torch._C._nn.gelu(x_148, approximate="none") + x_148 = None + x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) + x_149 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + x_153 = x_147 + x_152 + x_147 = x_152 = None + layer_norm_32 = torch.nn.functional.layer_norm( + x_153, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_32 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_34 = linear_64.reshape(1, 197, 3, 7, 64) + linear_64 = None + qkv_13 = reshape_34.permute(2, 0, 3, 1, 4) + reshape_34 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_15 = unbind_13[0] + k_15 = unbind_13[1] + v_15 = unbind_13[2] + unbind_13 = None + x_154 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_19 = x_154.transpose(1, 2) + x_154 = None + x_155 = transpose_19.reshape(1, 197, 448) + transpose_19 = None + x_156 = torch._C._nn.linear( + x_155, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_155 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = (None) + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = x_153 + x_157 + x_153 = x_157 = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_158, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = (None) + x_159 = torch._C._nn.linear( + layer_norm_33, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_33 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_160 = torch._C._nn.gelu(x_159, approximate="none") + x_159 = None + x_161 = torch.nn.functional.dropout(x_160, 0.0, False, False) + x_160 = None + x_162 = torch._C._nn.linear( + x_161, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_161 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_163 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + x_164 = x_158 + x_163 + x_158 = x_163 = None + getitem_70 = x_98[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_13 = torch.nn.functional.layer_norm( + getitem_70, + (224,), + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_70 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_14 = torch._C._nn.gelu(input_13, approximate="none") + input_13 = None + input_15 = torch._C._nn.linear( + input_14, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_14 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_71 = x_164[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_16 = torch.nn.functional.layer_norm( + getitem_71, + (448,), + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_71 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_17 = torch._C._nn.gelu(input_16, approximate="none") + input_16 = None + input_18 = torch._C._nn.linear( + input_17, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_17 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_72 = x_164[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_4 = torch.cat((input_15, getitem_72), dim=1) + input_15 = getitem_72 = None + getitem_73 = tmp_4[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_36 = torch.nn.functional.layer_norm( + tmp_4, + (448,), + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_4 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_74 = layer_norm_36[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_70 = torch._C._nn.linear( + getitem_74, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_74 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_36 = linear_70.reshape(1, 1, 7, 64) + linear_70 = None + q_16 = reshape_36.permute(0, 2, 1, 3) + reshape_36 = None + linear_71 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_37 = linear_71.reshape(1, 197, 7, 64) + linear_71 = None + k_16 = reshape_37.permute(0, 2, 1, 3) + reshape_37 = None + linear_72 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_36 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_38 = linear_72.reshape(1, 197, 7, 64) + linear_72 = None + v_16 = reshape_38.permute(0, 2, 1, 3) + reshape_38 = None + transpose_20 = k_16.transpose(-2, -1) + k_16 = None + matmul_4 = q_16 @ transpose_20 + q_16 = transpose_20 = None + attn_6 = matmul_4 * 0.125 + matmul_4 = None + attn_7 = attn_6.softmax(dim=-1) + attn_6 = None + attn_8 = torch.nn.functional.dropout(attn_7, 0.0, False, False) + attn_7 = None + matmul_5 = attn_8 @ v_16 + attn_8 = v_16 = None + transpose_21 = matmul_5.transpose(1, 2) + matmul_5 = None + x_165 = transpose_21.reshape(1, 1, 448) + transpose_21 = None + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_165 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_167 = torch.nn.functional.dropout(x_166, 0.0, False, False) + x_166 = None + x_168 = getitem_73 + x_167 + getitem_73 = x_167 = None + getitem_75 = x_168[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_168 = None + input_19 = torch.nn.functional.layer_norm( + getitem_75, + (448,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_75 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_20 = torch._C._nn.gelu(input_19, approximate="none") + input_19 = None + input_21 = torch._C._nn.linear( + input_20, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_20 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_76 = x_98[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_5 = torch.cat((input_21, getitem_76), dim=1) + input_21 = getitem_76 = None + getitem_77 = x_98[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_98 = None + tmp_6 = torch.cat((input_18, getitem_77), dim=1) + input_18 = getitem_77 = None + getitem_78 = tmp_6[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_38 = torch.nn.functional.layer_norm( + tmp_6, + (224,), + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_6 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_79 = layer_norm_38[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_75 = torch._C._nn.linear( + getitem_79, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_79 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_40 = linear_75.reshape(1, 1, 7, 32) + linear_75 = None + q_17 = reshape_40.permute(0, 2, 1, 3) + reshape_40 = None + linear_76 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_41 = linear_76.reshape(1, 401, 7, 32) + linear_76 = None + k_17 = reshape_41.permute(0, 2, 1, 3) + reshape_41 = None + linear_77 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_38 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_42 = linear_77.reshape(1, 401, 7, 32) + linear_77 = None + v_17 = reshape_42.permute(0, 2, 1, 3) + reshape_42 = None + transpose_22 = k_17.transpose(-2, -1) + k_17 = None + matmul_6 = q_17 @ transpose_22 + q_17 = transpose_22 = None + attn_9 = matmul_6 * 0.1767766952966369 + matmul_6 = None + attn_10 = attn_9.softmax(dim=-1) + attn_9 = None + attn_11 = torch.nn.functional.dropout(attn_10, 0.0, False, False) + attn_10 = None + matmul_7 = attn_11 @ v_17 + attn_11 = v_17 = None + transpose_23 = matmul_7.transpose(1, 2) + matmul_7 = None + x_169 = transpose_23.reshape(1, 1, 224) + transpose_23 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_169 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = getitem_78 + x_171 + getitem_78 = x_171 = None + getitem_80 = x_172[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_172 = None + input_22 = torch.nn.functional.layer_norm( + getitem_80, + (224,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_80 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_23 = torch._C._nn.gelu(input_22, approximate="none") + input_22 = None + input_24 = torch._C._nn.linear( + input_23, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_23 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_81 = x_164[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_164 = None + tmp_7 = torch.cat((input_24, getitem_81), dim=1) + input_24 = getitem_81 = None + layer_norm_40 = torch.nn.functional.layer_norm( + tmp_5, + (224,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_40 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_44 = linear_80.reshape(1, 401, 3, 7, 32) + linear_80 = None + qkv_14 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_18 = unbind_14[0] + k_18 = unbind_14[1] + v_18 = unbind_14[2] + unbind_14 = None + x_173 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_24 = x_173.transpose(1, 2) + x_173 = None + x_174 = transpose_24.reshape(1, 401, 224) + transpose_24 = None + x_175 = torch._C._nn.linear( + x_174, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_174 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = tmp_5 + x_176 + tmp_5 = x_176 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_177, + (224,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_178 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_41 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_179 = torch._C._nn.gelu(x_178, approximate="none") + x_178 = None + x_180 = torch.nn.functional.dropout(x_179, 0.0, False, False) + x_179 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_177 + x_182 + x_177 = x_182 = None + layer_norm_42 = torch.nn.functional.layer_norm( + tmp_7, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_84 = torch._C._nn.linear( + layer_norm_42, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_42 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_46 = linear_84.reshape(1, 197, 3, 7, 64) + linear_84 = None + qkv_15 = reshape_46.permute(2, 0, 3, 1, 4) + reshape_46 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_19 = unbind_15[0] + k_19 = unbind_15[1] + v_19 = unbind_15[2] + unbind_15 = None + x_184 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, attn_mask=None, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_25 = x_184.transpose(1, 2) + x_184 = None + x_185 = transpose_25.reshape(1, 197, 448) + transpose_25 = None + x_186 = torch._C._nn.linear( + x_185, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_185 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_187 = torch.nn.functional.dropout(x_186, 0.0, False, False) + x_186 = None + x_188 = tmp_7 + x_187 + tmp_7 = x_187 = None + layer_norm_43 = torch.nn.functional.layer_norm( + x_188, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_189 = torch._C._nn.linear( + layer_norm_43, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_43 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_190 = torch._C._nn.gelu(x_189, approximate="none") + x_189 = None + x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) + x_190 = None + x_192 = torch._C._nn.linear( + x_191, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_191 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_188 + x_193 + x_188 = x_193 = None + layer_norm_44 = torch.nn.functional.layer_norm( + x_194, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_88 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_44 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_48 = linear_88.reshape(1, 197, 3, 7, 64) + linear_88 = None + qkv_16 = reshape_48.permute(2, 0, 3, 1, 4) + reshape_48 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_20 = unbind_16[0] + k_20 = unbind_16[1] + v_20 = unbind_16[2] + unbind_16 = None + x_195 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, attn_mask=None, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_26 = x_195.transpose(1, 2) + x_195 = None + x_196 = transpose_26.reshape(1, 197, 448) + transpose_26 = None + x_197 = torch._C._nn.linear( + x_196, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_196 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_198 = torch.nn.functional.dropout(x_197, 0.0, False, False) + x_197 = None + x_199 = x_194 + x_198 + x_194 = x_198 = None + layer_norm_45 = torch.nn.functional.layer_norm( + x_199, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_200 = torch._C._nn.linear( + layer_norm_45, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_45 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_201 = torch._C._nn.gelu(x_200, approximate="none") + x_200 = None + x_202 = torch.nn.functional.dropout(x_201, 0.0, False, False) + x_201 = None + x_203 = torch._C._nn.linear( + x_202, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_202 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_205 = x_199 + x_204 + x_199 = x_204 = None + layer_norm_46 = torch.nn.functional.layer_norm( + x_205, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_92 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_46 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_50 = linear_92.reshape(1, 197, 3, 7, 64) + linear_92 = None + qkv_17 = reshape_50.permute(2, 0, 3, 1, 4) + reshape_50 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_21 = unbind_17[0] + k_21 = unbind_17[1] + v_21 = unbind_17[2] + unbind_17 = None + x_206 = torch._C._nn.scaled_dot_product_attention( + q_21, k_21, v_21, attn_mask=None, dropout_p=0.0 + ) + q_21 = k_21 = v_21 = None + transpose_27 = x_206.transpose(1, 2) + x_206 = None + x_207 = transpose_27.reshape(1, 197, 448) + transpose_27 = None + x_208 = torch._C._nn.linear( + x_207, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_207 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_209 = torch.nn.functional.dropout(x_208, 0.0, False, False) + x_208 = None + x_210 = x_205 + x_209 + x_205 = x_209 = None + layer_norm_47 = torch.nn.functional.layer_norm( + x_210, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_211 = torch._C._nn.linear( + layer_norm_47, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_47 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_212 = torch._C._nn.gelu(x_211, approximate="none") + x_211 = None + x_213 = torch.nn.functional.dropout(x_212, 0.0, False, False) + x_212 = None + x_214 = torch._C._nn.linear( + x_213, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_213 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = x_210 + x_215 + x_210 = x_215 = None + layer_norm_48 = torch.nn.functional.layer_norm( + x_216, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_96 = torch._C._nn.linear( + layer_norm_48, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_48 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_52 = linear_96.reshape(1, 197, 3, 7, 64) + linear_96 = None + qkv_18 = reshape_52.permute(2, 0, 3, 1, 4) + reshape_52 = None + unbind_18 = qkv_18.unbind(0) + qkv_18 = None + q_22 = unbind_18[0] + k_22 = unbind_18[1] + v_22 = unbind_18[2] + unbind_18 = None + x_217 = torch._C._nn.scaled_dot_product_attention( + q_22, k_22, v_22, attn_mask=None, dropout_p=0.0 + ) + q_22 = k_22 = v_22 = None + transpose_28 = x_217.transpose(1, 2) + x_217 = None + x_218 = transpose_28.reshape(1, 197, 448) + transpose_28 = None + x_219 = torch._C._nn.linear( + x_218, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_218 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_220 = torch.nn.functional.dropout(x_219, 0.0, False, False) + x_219 = None + x_221 = x_216 + x_220 + x_216 = x_220 = None + layer_norm_49 = torch.nn.functional.layer_norm( + x_221, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_222 = torch._C._nn.linear( + layer_norm_49, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_49 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_223 = torch._C._nn.gelu(x_222, approximate="none") + x_222 = None + x_224 = torch.nn.functional.dropout(x_223, 0.0, False, False) + x_223 = None + x_225 = torch._C._nn.linear( + x_224, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_224 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_226 = torch.nn.functional.dropout(x_225, 0.0, False, False) + x_225 = None + x_227 = x_221 + x_226 + x_221 = x_226 = None + layer_norm_50 = torch.nn.functional.layer_norm( + x_227, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_100 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_50 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_54 = linear_100.reshape(1, 197, 3, 7, 64) + linear_100 = None + qkv_19 = reshape_54.permute(2, 0, 3, 1, 4) + reshape_54 = None + unbind_19 = qkv_19.unbind(0) + qkv_19 = None + q_23 = unbind_19[0] + k_23 = unbind_19[1] + v_23 = unbind_19[2] + unbind_19 = None + x_228 = torch._C._nn.scaled_dot_product_attention( + q_23, k_23, v_23, attn_mask=None, dropout_p=0.0 + ) + q_23 = k_23 = v_23 = None + transpose_29 = x_228.transpose(1, 2) + x_228 = None + x_229 = transpose_29.reshape(1, 197, 448) + transpose_29 = None + x_230 = torch._C._nn.linear( + x_229, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_229 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_231 = torch.nn.functional.dropout(x_230, 0.0, False, False) + x_230 = None + x_232 = x_227 + x_231 + x_227 = x_231 = None + layer_norm_51 = torch.nn.functional.layer_norm( + x_232, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_233 = torch._C._nn.linear( + layer_norm_51, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_51 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_234 = torch._C._nn.gelu(x_233, approximate="none") + x_233 = None + x_235 = torch.nn.functional.dropout(x_234, 0.0, False, False) + x_234 = None + x_236 = torch._C._nn.linear( + x_235, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_235 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_237 = torch.nn.functional.dropout(x_236, 0.0, False, False) + x_236 = None + x_238 = x_232 + x_237 + x_232 = x_237 = None + layer_norm_52 = torch.nn.functional.layer_norm( + x_238, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = (None) + linear_104 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_52 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_56 = linear_104.reshape(1, 197, 3, 7, 64) + linear_104 = None + qkv_20 = reshape_56.permute(2, 0, 3, 1, 4) + reshape_56 = None + unbind_20 = qkv_20.unbind(0) + qkv_20 = None + q_24 = unbind_20[0] + k_24 = unbind_20[1] + v_24 = unbind_20[2] + unbind_20 = None + x_239 = torch._C._nn.scaled_dot_product_attention( + q_24, k_24, v_24, attn_mask=None, dropout_p=0.0 + ) + q_24 = k_24 = v_24 = None + transpose_30 = x_239.transpose(1, 2) + x_239 = None + x_240 = transpose_30.reshape(1, 197, 448) + transpose_30 = None + x_241 = torch._C._nn.linear( + x_240, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_240 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = (None) + x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) + x_241 = None + x_243 = x_238 + x_242 + x_238 = x_242 = None + layer_norm_53 = torch.nn.functional.layer_norm( + x_243, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = (None) + x_244 = torch._C._nn.linear( + layer_norm_53, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_53 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_245 = torch._C._nn.gelu(x_244, approximate="none") + x_244 = None + x_246 = torch.nn.functional.dropout(x_245, 0.0, False, False) + x_245 = None + x_247 = torch._C._nn.linear( + x_246, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_246 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_248 = torch.nn.functional.dropout(x_247, 0.0, False, False) + x_247 = None + x_249 = x_243 + x_248 + x_243 = x_248 = None + getitem_103 = x_183[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_25 = torch.nn.functional.layer_norm( + getitem_103, + (224,), + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_103 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_26 = torch._C._nn.gelu(input_25, approximate="none") + input_25 = None + input_27 = torch._C._nn.linear( + input_26, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_26 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_104 = x_249[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_28 = torch.nn.functional.layer_norm( + getitem_104, + (448,), + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_104 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_29 = torch._C._nn.gelu(input_28, approximate="none") + input_28 = None + input_30 = torch._C._nn.linear( + input_29, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_29 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_105 = x_249[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_8 = torch.cat((input_27, getitem_105), dim=1) + input_27 = getitem_105 = None + getitem_106 = tmp_8[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_56 = torch.nn.functional.layer_norm( + tmp_8, + (448,), + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_8 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_107 = layer_norm_56[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_110 = torch._C._nn.linear( + getitem_107, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_107 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_58 = linear_110.reshape(1, 1, 7, 64) + linear_110 = None + q_25 = reshape_58.permute(0, 2, 1, 3) + reshape_58 = None + linear_111 = torch._C._nn.linear( + layer_norm_56, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_59 = linear_111.reshape(1, 197, 7, 64) + linear_111 = None + k_25 = reshape_59.permute(0, 2, 1, 3) + reshape_59 = None + linear_112 = torch._C._nn.linear( + layer_norm_56, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_56 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_60 = linear_112.reshape(1, 197, 7, 64) + linear_112 = None + v_25 = reshape_60.permute(0, 2, 1, 3) + reshape_60 = None + transpose_31 = k_25.transpose(-2, -1) + k_25 = None + matmul_8 = q_25 @ transpose_31 + q_25 = transpose_31 = None + attn_12 = matmul_8 * 0.125 + matmul_8 = None + attn_13 = attn_12.softmax(dim=-1) + attn_12 = None + attn_14 = torch.nn.functional.dropout(attn_13, 0.0, False, False) + attn_13 = None + matmul_9 = attn_14 @ v_25 + attn_14 = v_25 = None + transpose_32 = matmul_9.transpose(1, 2) + matmul_9 = None + x_250 = transpose_32.reshape(1, 1, 448) + transpose_32 = None + x_251 = torch._C._nn.linear( + x_250, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_250 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_252 = torch.nn.functional.dropout(x_251, 0.0, False, False) + x_251 = None + x_253 = getitem_106 + x_252 + getitem_106 = x_252 = None + getitem_108 = x_253[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_253 = None + input_31 = torch.nn.functional.layer_norm( + getitem_108, + (448,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_108 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_32 = torch._C._nn.gelu(input_31, approximate="none") + input_31 = None + input_33 = torch._C._nn.linear( + input_32, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_32 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_109 = x_183[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_9 = torch.cat((input_33, getitem_109), dim=1) + input_33 = getitem_109 = None + getitem_110 = x_183[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_183 = None + tmp_10 = torch.cat((input_30, getitem_110), dim=1) + input_30 = getitem_110 = None + getitem_111 = tmp_10[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_58 = torch.nn.functional.layer_norm( + tmp_10, + (224,), + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_10 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_112 = layer_norm_58[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_115 = torch._C._nn.linear( + getitem_112, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_112 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_62 = linear_115.reshape(1, 1, 7, 32) + linear_115 = None + q_26 = reshape_62.permute(0, 2, 1, 3) + reshape_62 = None + linear_116 = torch._C._nn.linear( + layer_norm_58, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_63 = linear_116.reshape(1, 401, 7, 32) + linear_116 = None + k_26 = reshape_63.permute(0, 2, 1, 3) + reshape_63 = None + linear_117 = torch._C._nn.linear( + layer_norm_58, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_58 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_64 = linear_117.reshape(1, 401, 7, 32) + linear_117 = None + v_26 = reshape_64.permute(0, 2, 1, 3) + reshape_64 = None + transpose_33 = k_26.transpose(-2, -1) + k_26 = None + matmul_10 = q_26 @ transpose_33 + q_26 = transpose_33 = None + attn_15 = matmul_10 * 0.1767766952966369 + matmul_10 = None + attn_16 = attn_15.softmax(dim=-1) + attn_15 = None + attn_17 = torch.nn.functional.dropout(attn_16, 0.0, False, False) + attn_16 = None + matmul_11 = attn_17 @ v_26 + attn_17 = v_26 = None + transpose_34 = matmul_11.transpose(1, 2) + matmul_11 = None + x_254 = transpose_34.reshape(1, 1, 224) + transpose_34 = None + x_255 = torch._C._nn.linear( + x_254, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_254 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_257 = getitem_111 + x_256 + getitem_111 = x_256 = None + getitem_113 = x_257[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_257 = None + input_34 = torch.nn.functional.layer_norm( + getitem_113, + (224,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_113 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_35 = torch._C._nn.gelu(input_34, approximate="none") + input_34 = None + input_36 = torch._C._nn.linear( + input_35, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_35 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_114 = x_249[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_249 = None + tmp_11 = torch.cat((input_36, getitem_114), dim=1) + input_36 = getitem_114 = None + x_258 = torch.nn.functional.layer_norm( + tmp_9, + (224,), + l_self_modules_norm_modules_0_parameters_weight_, + l_self_modules_norm_modules_0_parameters_bias_, + 1e-06, + ) + tmp_9 = ( + l_self_modules_norm_modules_0_parameters_weight_ + ) = l_self_modules_norm_modules_0_parameters_bias_ = None + x_259 = torch.nn.functional.layer_norm( + tmp_11, + (448,), + l_self_modules_norm_modules_1_parameters_weight_, + l_self_modules_norm_modules_1_parameters_bias_, + 1e-06, + ) + tmp_11 = ( + l_self_modules_norm_modules_1_parameters_weight_ + ) = l_self_modules_norm_modules_1_parameters_bias_ = None + x_260 = x_258[(slice(None, None, None), 0)] + x_258 = None + x_261 = x_259[(slice(None, None, None), 0)] + x_259 = None + dropout_77 = torch.nn.functional.dropout(x_260, 0.0, False, False) + x_260 = None + dropout_78 = torch.nn.functional.dropout(x_261, 0.0, False, False) + x_261 = None + linear_120 = torch._C._nn.linear( + dropout_77, + l_self_modules_head_modules_0_parameters_weight_, + l_self_modules_head_modules_0_parameters_bias_, + ) + dropout_77 = ( + l_self_modules_head_modules_0_parameters_weight_ + ) = l_self_modules_head_modules_0_parameters_bias_ = None + linear_121 = torch._C._nn.linear( + dropout_78, + l_self_modules_head_modules_1_parameters_weight_, + l_self_modules_head_modules_1_parameters_bias_, + ) + dropout_78 = ( + l_self_modules_head_modules_1_parameters_weight_ + ) = l_self_modules_head_modules_1_parameters_bias_ = None + stack = torch.stack([linear_120, linear_121], dim=0) + linear_120 = linear_121 = None + x_262 = torch.mean(stack, dim=0) + stack = None + return (x_262,) diff --git a/samples/timm/crossvit_18_240.in1k/weight_meta.py b/samples/timm/crossvit_18_240.in1k/weight_meta.py new file mode 100644 index 0000000000..f8c3efb498 --- /dev/null +++ b/samples/timm/crossvit_18_240.in1k/weight_meta.py @@ -0,0 +1,3768 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.227 + std = 1.288 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_" + shape = [224, 3, 12, 12] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.028 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_0_: + name = "L_self_parameters_cls_token_0_" + shape = [1, 1, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_0_: + name = "L_self_parameters_pos_embed_0_" + shape = [1, 401, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_" + shape = [448, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_1_: + name = "L_self_parameters_cls_token_1_" + shape = [1, 1, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_1_: + name = "L_self_parameters_pos_embed_1_" + shape = [1, 197, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [224, 672] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [224, 672] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [224, 672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_weight_: + name = "L_self_modules_norm_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_bias_: + name = "L_self_modules_norm_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_weight_: + name = "L_self_modules_norm_modules_1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_bias_: + name = "L_self_modules_norm_modules_1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_weight_: + name = "L_self_modules_head_modules_0_parameters_weight_" + shape = [1000, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_bias_: + name = "L_self_modules_head_modules_0_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_weight_: + name = "L_self_modules_head_modules_1_parameters_weight_" + shape = [1000, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_bias_: + name = "L_self_modules_head_modules_1_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/crossvit_18_dagger_240.in1k/graph_hash.txt b/samples/timm/crossvit_18_dagger_240.in1k/graph_hash.txt new file mode 100644 index 0000000000..e1d20be9a7 --- /dev/null +++ b/samples/timm/crossvit_18_dagger_240.in1k/graph_hash.txt @@ -0,0 +1 @@ +38cd8450bce14ad3d7db0b79ba42ef2cc211152c220e9615c9ecadec74e45cff \ No newline at end of file diff --git a/samples/timm/crossvit_18_dagger_240.in1k/graph_net.json b/samples/timm/crossvit_18_dagger_240.in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/crossvit_18_dagger_240.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/crossvit_18_dagger_240.in1k/input_meta.py b/samples/timm/crossvit_18_dagger_240.in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/crossvit_18_dagger_240.in1k/input_tensor_constraints.py b/samples/timm/crossvit_18_dagger_240.in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/crossvit_18_dagger_240.in1k/model.py b/samples/timm/crossvit_18_dagger_240.in1k/model.py new file mode 100644 index 0000000000..98891b26b3 --- /dev/null +++ b/samples/timm/crossvit_18_dagger_240.in1k/model.py @@ -0,0 +1,3091 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_0_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_0_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_1_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ + ) + l_self_parameters_cls_token_0_ = L_self_parameters_cls_token_0_ + l_self_parameters_pos_embed_0_ = L_self_parameters_pos_embed_0_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ + ) + l_self_parameters_cls_token_1_ = L_self_parameters_cls_token_1_ + l_self_parameters_pos_embed_1_ = L_self_parameters_pos_embed_1_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_norm_modules_0_parameters_weight_ = ( + L_self_modules_norm_modules_0_parameters_weight_ + ) + l_self_modules_norm_modules_0_parameters_bias_ = ( + L_self_modules_norm_modules_0_parameters_bias_ + ) + l_self_modules_norm_modules_1_parameters_weight_ = ( + L_self_modules_norm_modules_1_parameters_weight_ + ) + l_self_modules_norm_modules_1_parameters_bias_ = ( + L_self_modules_norm_modules_1_parameters_bias_ + ) + l_self_modules_head_modules_0_parameters_weight_ = ( + L_self_modules_head_modules_0_parameters_weight_ + ) + l_self_modules_head_modules_0_parameters_bias_ = ( + L_self_modules_head_modules_0_parameters_bias_ + ) + l_self_modules_head_modules_1_parameters_weight_ = ( + L_self_modules_head_modules_1_parameters_weight_ + ) + l_self_modules_head_modules_1_parameters_bias_ = ( + L_self_modules_head_modules_1_parameters_bias_ + ) + x = torch.nn.functional.interpolate( + l_x_, size=(240, 240), mode="bicubic", align_corners=False + ) + input_1 = torch.conv2d( + x, + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_, + (4, 4), + (3, 3), + (1, 1), + 1, + ) + x = l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ + ) = None + input_2 = torch.nn.functional.relu(input_1, inplace=True) + input_1 = None + input_3 = torch.conv2d( + input_2, + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_, + (3, 3), + (0, 0), + (1, 1), + 1, + ) + input_2 = l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ + ) = None + input_4 = torch.nn.functional.relu(input_3, inplace=True) + input_3 = None + input_5 = torch.conv2d( + input_4, + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + input_4 = l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ + ) = None + flatten = input_5.flatten(2) + input_5 = None + x_1 = flatten.transpose(1, 2) + flatten = None + cls_tokens = l_self_parameters_cls_token_0_.expand(1, -1, -1) + l_self_parameters_cls_token_0_ = None + x_ = torch.cat((cls_tokens, x_1), dim=1) + cls_tokens = x_1 = None + x__1 = x_ + l_self_parameters_pos_embed_0_ + x_ = l_self_parameters_pos_embed_0_ = None + x__2 = torch.nn.functional.dropout(x__1, 0.0, False, False) + x__1 = None + input_6 = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_, + (4, 4), + (3, 3), + (1, 1), + 1, + ) + l_x_ = l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ + ) = None + input_7 = torch.nn.functional.relu(input_6, inplace=True) + input_6 = None + input_8 = torch.conv2d( + input_7, + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_7 = l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ + ) = None + input_9 = torch.nn.functional.relu(input_8, inplace=True) + input_8 = None + input_10 = torch.conv2d( + input_9, + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_9 = l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ + ) = None + flatten_1 = input_10.flatten(2) + input_10 = None + x_2 = flatten_1.transpose(1, 2) + flatten_1 = None + cls_tokens_1 = l_self_parameters_cls_token_1_.expand(1, -1, -1) + l_self_parameters_cls_token_1_ = None + x__3 = torch.cat((cls_tokens_1, x_2), dim=1) + cls_tokens_1 = x_2 = None + x__4 = x__3 + l_self_parameters_pos_embed_1_ + x__3 = l_self_parameters_pos_embed_1_ = None + x__5 = torch.nn.functional.dropout(x__4, 0.0, False, False) + x__4 = None + layer_norm = torch.nn.functional.layer_norm( + x__2, + (224,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear = torch._C._nn.linear( + layer_norm, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 401, 3, 7, 32) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_3 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_2 = x_3.transpose(1, 2) + x_3 = None + x_4 = transpose_2.reshape(1, 401, 224) + transpose_2 = None + x_5 = torch._C._nn.linear( + x_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_6 = torch.nn.functional.dropout(x_5, 0.0, False, False) + x_5 = None + x_7 = x__2 + x_6 + x__2 = x_6 = None + layer_norm_1 = torch.nn.functional.layer_norm( + x_7, + (224,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_8 = torch._C._nn.linear( + layer_norm_1, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_1 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_9 = torch._C._nn.gelu(x_8, approximate="none") + x_8 = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = torch._C._nn.linear( + x_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_7 + x_12 + x_7 = x_12 = None + layer_norm_2 = torch.nn.functional.layer_norm( + x__5, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_2, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_2 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_2 = linear_4.reshape(1, 197, 3, 7, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_14 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_3 = x_14.transpose(1, 2) + x_14 = None + x_15 = transpose_3.reshape(1, 197, 448) + transpose_3 = None + x_16 = torch._C._nn.linear( + x_15, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_15 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = x__5 + x_17 + x__5 = x_17 = None + layer_norm_3 = torch.nn.functional.layer_norm( + x_18, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_19 = torch._C._nn.linear( + layer_norm_3, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_3 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_20 = torch._C._nn.gelu(x_19, approximate="none") + x_19 = None + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = torch._C._nn.linear( + x_21, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_21 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_18 + x_23 + x_18 = x_23 = None + layer_norm_4 = torch.nn.functional.layer_norm( + x_24, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_8.reshape(1, 197, 3, 7, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_25 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_4 = x_25.transpose(1, 2) + x_25 = None + x_26 = transpose_4.reshape(1, 197, 448) + transpose_4 = None + x_27 = torch._C._nn.linear( + x_26, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_26 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_28 = torch.nn.functional.dropout(x_27, 0.0, False, False) + x_27 = None + x_29 = x_24 + x_28 + x_24 = x_28 = None + layer_norm_5 = torch.nn.functional.layer_norm( + x_29, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_30 = torch._C._nn.linear( + layer_norm_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_31 = torch._C._nn.gelu(x_30, approximate="none") + x_30 = None + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = torch._C._nn.linear( + x_32, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_32 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = x_29 + x_34 + x_29 = x_34 = None + layer_norm_6 = torch.nn.functional.layer_norm( + x_35, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_6, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_6 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_6 = linear_12.reshape(1, 197, 3, 7, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_36 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_5 = x_36.transpose(1, 2) + x_36 = None + x_37 = transpose_5.reshape(1, 197, 448) + transpose_5 = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_37 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_39 = torch.nn.functional.dropout(x_38, 0.0, False, False) + x_38 = None + x_40 = x_35 + x_39 + x_35 = x_39 = None + layer_norm_7 = torch.nn.functional.layer_norm( + x_40, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_41 = torch._C._nn.linear( + layer_norm_7, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_7 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_42 = torch._C._nn.gelu(x_41, approximate="none") + x_41 = None + x_43 = torch.nn.functional.dropout(x_42, 0.0, False, False) + x_42 = None + x_44 = torch._C._nn.linear( + x_43, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_43 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = x_40 + x_45 + x_40 = x_45 = None + layer_norm_8 = torch.nn.functional.layer_norm( + x_46, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_8, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_8 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_8 = linear_16.reshape(1, 197, 3, 7, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_47 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_6 = x_47.transpose(1, 2) + x_47 = None + x_48 = transpose_6.reshape(1, 197, 448) + transpose_6 = None + x_49 = torch._C._nn.linear( + x_48, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_48 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_50 = torch.nn.functional.dropout(x_49, 0.0, False, False) + x_49 = None + x_51 = x_46 + x_50 + x_46 = x_50 = None + layer_norm_9 = torch.nn.functional.layer_norm( + x_51, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_52 = torch._C._nn.linear( + layer_norm_9, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_9 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_53 = torch._C._nn.gelu(x_52, approximate="none") + x_52 = None + x_54 = torch.nn.functional.dropout(x_53, 0.0, False, False) + x_53 = None + x_55 = torch._C._nn.linear( + x_54, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_54 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_56 = torch.nn.functional.dropout(x_55, 0.0, False, False) + x_55 = None + x_57 = x_51 + x_56 + x_51 = x_56 = None + layer_norm_10 = torch.nn.functional.layer_norm( + x_57, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_20 = torch._C._nn.linear( + layer_norm_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_10 = linear_20.reshape(1, 197, 3, 7, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_7 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_7.reshape(1, 197, 448) + transpose_7 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + x_62 = x_57 + x_61 + x_57 = x_61 = None + layer_norm_11 = torch.nn.functional.layer_norm( + x_62, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_63 = torch._C._nn.linear( + layer_norm_11, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_11 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_64 = torch._C._nn.gelu(x_63, approximate="none") + x_63 = None + x_65 = torch.nn.functional.dropout(x_64, 0.0, False, False) + x_64 = None + x_66 = torch._C._nn.linear( + x_65, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_65 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_67 = torch.nn.functional.dropout(x_66, 0.0, False, False) + x_66 = None + x_68 = x_62 + x_67 + x_62 = x_67 = None + layer_norm_12 = torch.nn.functional.layer_norm( + x_68, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = (None) + linear_24 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_12 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_12 = linear_24.reshape(1, 197, 3, 7, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_69 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_8 = x_69.transpose(1, 2) + x_69 = None + x_70 = transpose_8.reshape(1, 197, 448) + transpose_8 = None + x_71 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_70 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = (None) + x_72 = torch.nn.functional.dropout(x_71, 0.0, False, False) + x_71 = None + x_73 = x_68 + x_72 + x_68 = x_72 = None + layer_norm_13 = torch.nn.functional.layer_norm( + x_73, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = (None) + x_74 = torch._C._nn.linear( + layer_norm_13, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_13 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_75 = torch._C._nn.gelu(x_74, approximate="none") + x_74 = None + x_76 = torch.nn.functional.dropout(x_75, 0.0, False, False) + x_75 = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_76 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_78 = torch.nn.functional.dropout(x_77, 0.0, False, False) + x_77 = None + x_79 = x_73 + x_78 + x_73 = x_78 = None + getitem_37 = x_13[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_11 = torch.nn.functional.layer_norm( + getitem_37, + (224,), + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_37 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_12 = torch._C._nn.gelu(input_11, approximate="none") + input_11 = None + input_13 = torch._C._nn.linear( + input_12, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_12 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_38 = x_79[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_14 = torch.nn.functional.layer_norm( + getitem_38, + (448,), + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_38 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_15 = torch._C._nn.gelu(input_14, approximate="none") + input_14 = None + input_16 = torch._C._nn.linear( + input_15, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_15 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_39 = x_79[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp = torch.cat((input_13, getitem_39), dim=1) + input_13 = getitem_39 = None + getitem_40 = tmp[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_16 = torch.nn.functional.layer_norm( + tmp, + (448,), + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_41 = layer_norm_16[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_30 = torch._C._nn.linear( + getitem_41, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_41 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_14 = linear_30.reshape(1, 1, 7, 64) + linear_30 = None + q_7 = reshape_14.permute(0, 2, 1, 3) + reshape_14 = None + linear_31 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_15 = linear_31.reshape(1, 197, 7, 64) + linear_31 = None + k_7 = reshape_15.permute(0, 2, 1, 3) + reshape_15 = None + linear_32 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_16 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_16 = linear_32.reshape(1, 197, 7, 64) + linear_32 = None + v_7 = reshape_16.permute(0, 2, 1, 3) + reshape_16 = None + transpose_9 = k_7.transpose(-2, -1) + k_7 = None + matmul = q_7 @ transpose_9 + q_7 = transpose_9 = None + attn = matmul * 0.125 + matmul = None + attn_1 = attn.softmax(dim=-1) + attn = None + attn_2 = torch.nn.functional.dropout(attn_1, 0.0, False, False) + attn_1 = None + matmul_1 = attn_2 @ v_7 + attn_2 = v_7 = None + transpose_10 = matmul_1.transpose(1, 2) + matmul_1 = None + x_80 = transpose_10.reshape(1, 1, 448) + transpose_10 = None + x_81 = torch._C._nn.linear( + x_80, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_80 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_82 = torch.nn.functional.dropout(x_81, 0.0, False, False) + x_81 = None + x_83 = getitem_40 + x_82 + getitem_40 = x_82 = None + getitem_42 = x_83[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_83 = None + input_17 = torch.nn.functional.layer_norm( + getitem_42, + (448,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_42 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_18 = torch._C._nn.gelu(input_17, approximate="none") + input_17 = None + input_19 = torch._C._nn.linear( + input_18, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_18 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_43 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_1 = torch.cat((input_19, getitem_43), dim=1) + input_19 = getitem_43 = None + getitem_44 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_13 = None + tmp_2 = torch.cat((input_16, getitem_44), dim=1) + input_16 = getitem_44 = None + getitem_45 = tmp_2[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_18 = torch.nn.functional.layer_norm( + tmp_2, + (224,), + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_2 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_46 = layer_norm_18[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_35 = torch._C._nn.linear( + getitem_46, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_46 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_18 = linear_35.reshape(1, 1, 7, 32) + linear_35 = None + q_8 = reshape_18.permute(0, 2, 1, 3) + reshape_18 = None + linear_36 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_19 = linear_36.reshape(1, 401, 7, 32) + linear_36 = None + k_8 = reshape_19.permute(0, 2, 1, 3) + reshape_19 = None + linear_37 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_18 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_20 = linear_37.reshape(1, 401, 7, 32) + linear_37 = None + v_8 = reshape_20.permute(0, 2, 1, 3) + reshape_20 = None + transpose_11 = k_8.transpose(-2, -1) + k_8 = None + matmul_2 = q_8 @ transpose_11 + q_8 = transpose_11 = None + attn_3 = matmul_2 * 0.1767766952966369 + matmul_2 = None + attn_4 = attn_3.softmax(dim=-1) + attn_3 = None + attn_5 = torch.nn.functional.dropout(attn_4, 0.0, False, False) + attn_4 = None + matmul_3 = attn_5 @ v_8 + attn_5 = v_8 = None + transpose_12 = matmul_3.transpose(1, 2) + matmul_3 = None + x_84 = transpose_12.reshape(1, 1, 224) + transpose_12 = None + x_85 = torch._C._nn.linear( + x_84, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_84 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_86 = torch.nn.functional.dropout(x_85, 0.0, False, False) + x_85 = None + x_87 = getitem_45 + x_86 + getitem_45 = x_86 = None + getitem_47 = x_87[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_87 = None + input_20 = torch.nn.functional.layer_norm( + getitem_47, + (224,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_47 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_21 = torch._C._nn.gelu(input_20, approximate="none") + input_20 = None + input_22 = torch._C._nn.linear( + input_21, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_21 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_48 = x_79[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_79 = None + tmp_3 = torch.cat((input_22, getitem_48), dim=1) + input_22 = getitem_48 = None + layer_norm_20 = torch.nn.functional.layer_norm( + tmp_1, + (224,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_20 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_22 = linear_40.reshape(1, 401, 3, 7, 32) + linear_40 = None + qkv_7 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_9 = unbind_7[0] + k_9 = unbind_7[1] + v_9 = unbind_7[2] + unbind_7 = None + x_88 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_13 = x_88.transpose(1, 2) + x_88 = None + x_89 = transpose_13.reshape(1, 401, 224) + transpose_13 = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_89 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = tmp_1 + x_91 + tmp_1 = x_91 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_92, + (224,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_93 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_21 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_94 = torch._C._nn.gelu(x_93, approximate="none") + x_93 = None + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = torch._C._nn.linear( + x_95, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_95 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_97 = torch.nn.functional.dropout(x_96, 0.0, False, False) + x_96 = None + x_98 = x_92 + x_97 + x_92 = x_97 = None + layer_norm_22 = torch.nn.functional.layer_norm( + tmp_3, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_22, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_22 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_24 = linear_44.reshape(1, 197, 3, 7, 64) + linear_44 = None + qkv_8 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_10 = unbind_8[0] + k_10 = unbind_8[1] + v_10 = unbind_8[2] + unbind_8 = None + x_99 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_14 = x_99.transpose(1, 2) + x_99 = None + x_100 = transpose_14.reshape(1, 197, 448) + transpose_14 = None + x_101 = torch._C._nn.linear( + x_100, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_100 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = tmp_3 + x_102 + tmp_3 = x_102 = None + layer_norm_23 = torch.nn.functional.layer_norm( + x_103, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_104 = torch._C._nn.linear( + layer_norm_23, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_23 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_105 = torch._C._nn.gelu(x_104, approximate="none") + x_104 = None + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch._C._nn.linear( + x_106, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_106 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_103 + x_108 + x_103 = x_108 = None + layer_norm_24 = torch.nn.functional.layer_norm( + x_109, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_24 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_26 = linear_48.reshape(1, 197, 3, 7, 64) + linear_48 = None + qkv_9 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_11 = unbind_9[0] + k_11 = unbind_9[1] + v_11 = unbind_9[2] + unbind_9 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_15 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_15.reshape(1, 197, 448) + transpose_15 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_109 + x_113 + x_109 = x_113 = None + layer_norm_25 = torch.nn.functional.layer_norm( + x_114, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_115 = torch._C._nn.linear( + layer_norm_25, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_25 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_116 = torch._C._nn.gelu(x_115, approximate="none") + x_115 = None + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = torch._C._nn.linear( + x_117, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_117 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_119 = torch.nn.functional.dropout(x_118, 0.0, False, False) + x_118 = None + x_120 = x_114 + x_119 + x_114 = x_119 = None + layer_norm_26 = torch.nn.functional.layer_norm( + x_120, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_52 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_26 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_28 = linear_52.reshape(1, 197, 3, 7, 64) + linear_52 = None + qkv_10 = reshape_28.permute(2, 0, 3, 1, 4) + reshape_28 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_12 = unbind_10[0] + k_12 = unbind_10[1] + v_12 = unbind_10[2] + unbind_10 = None + x_121 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_16 = x_121.transpose(1, 2) + x_121 = None + x_122 = transpose_16.reshape(1, 197, 448) + transpose_16 = None + x_123 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_122 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_124 = torch.nn.functional.dropout(x_123, 0.0, False, False) + x_123 = None + x_125 = x_120 + x_124 + x_120 = x_124 = None + layer_norm_27 = torch.nn.functional.layer_norm( + x_125, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_126 = torch._C._nn.linear( + layer_norm_27, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_27 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_127 = torch._C._nn.gelu(x_126, approximate="none") + x_126 = None + x_128 = torch.nn.functional.dropout(x_127, 0.0, False, False) + x_127 = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_128 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_130 = torch.nn.functional.dropout(x_129, 0.0, False, False) + x_129 = None + x_131 = x_125 + x_130 + x_125 = x_130 = None + layer_norm_28 = torch.nn.functional.layer_norm( + x_131, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_28 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_30 = linear_56.reshape(1, 197, 3, 7, 64) + linear_56 = None + qkv_11 = reshape_30.permute(2, 0, 3, 1, 4) + reshape_30 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_13 = unbind_11[0] + k_13 = unbind_11[1] + v_13 = unbind_11[2] + unbind_11 = None + x_132 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_17 = x_132.transpose(1, 2) + x_132 = None + x_133 = transpose_17.reshape(1, 197, 448) + transpose_17 = None + x_134 = torch._C._nn.linear( + x_133, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_133 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_135 = torch.nn.functional.dropout(x_134, 0.0, False, False) + x_134 = None + x_136 = x_131 + x_135 + x_131 = x_135 = None + layer_norm_29 = torch.nn.functional.layer_norm( + x_136, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_137 = torch._C._nn.linear( + layer_norm_29, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_29 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_138 = torch._C._nn.gelu(x_137, approximate="none") + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = torch._C._nn.linear( + x_139, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_139 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_141 = torch.nn.functional.dropout(x_140, 0.0, False, False) + x_140 = None + x_142 = x_136 + x_141 + x_136 = x_141 = None + layer_norm_30 = torch.nn.functional.layer_norm( + x_142, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_60 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_30 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_32 = linear_60.reshape(1, 197, 3, 7, 64) + linear_60 = None + qkv_12 = reshape_32.permute(2, 0, 3, 1, 4) + reshape_32 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_14 = unbind_12[0] + k_14 = unbind_12[1] + v_14 = unbind_12[2] + unbind_12 = None + x_143 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_18 = x_143.transpose(1, 2) + x_143 = None + x_144 = transpose_18.reshape(1, 197, 448) + transpose_18 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_144 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = x_142 + x_146 + x_142 = x_146 = None + layer_norm_31 = torch.nn.functional.layer_norm( + x_147, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_148 = torch._C._nn.linear( + layer_norm_31, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_31 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_149 = torch._C._nn.gelu(x_148, approximate="none") + x_148 = None + x_150 = torch.nn.functional.dropout(x_149, 0.0, False, False) + x_149 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + x_153 = x_147 + x_152 + x_147 = x_152 = None + layer_norm_32 = torch.nn.functional.layer_norm( + x_153, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_32 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_34 = linear_64.reshape(1, 197, 3, 7, 64) + linear_64 = None + qkv_13 = reshape_34.permute(2, 0, 3, 1, 4) + reshape_34 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_15 = unbind_13[0] + k_15 = unbind_13[1] + v_15 = unbind_13[2] + unbind_13 = None + x_154 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_19 = x_154.transpose(1, 2) + x_154 = None + x_155 = transpose_19.reshape(1, 197, 448) + transpose_19 = None + x_156 = torch._C._nn.linear( + x_155, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_155 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = (None) + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = x_153 + x_157 + x_153 = x_157 = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_158, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = (None) + x_159 = torch._C._nn.linear( + layer_norm_33, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_33 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_160 = torch._C._nn.gelu(x_159, approximate="none") + x_159 = None + x_161 = torch.nn.functional.dropout(x_160, 0.0, False, False) + x_160 = None + x_162 = torch._C._nn.linear( + x_161, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_161 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_163 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + x_164 = x_158 + x_163 + x_158 = x_163 = None + getitem_70 = x_98[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_23 = torch.nn.functional.layer_norm( + getitem_70, + (224,), + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_70 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_24 = torch._C._nn.gelu(input_23, approximate="none") + input_23 = None + input_25 = torch._C._nn.linear( + input_24, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_24 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_71 = x_164[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_26 = torch.nn.functional.layer_norm( + getitem_71, + (448,), + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_71 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_27 = torch._C._nn.gelu(input_26, approximate="none") + input_26 = None + input_28 = torch._C._nn.linear( + input_27, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_27 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_72 = x_164[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_4 = torch.cat((input_25, getitem_72), dim=1) + input_25 = getitem_72 = None + getitem_73 = tmp_4[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_36 = torch.nn.functional.layer_norm( + tmp_4, + (448,), + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_4 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_74 = layer_norm_36[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_70 = torch._C._nn.linear( + getitem_74, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_74 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_36 = linear_70.reshape(1, 1, 7, 64) + linear_70 = None + q_16 = reshape_36.permute(0, 2, 1, 3) + reshape_36 = None + linear_71 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_37 = linear_71.reshape(1, 197, 7, 64) + linear_71 = None + k_16 = reshape_37.permute(0, 2, 1, 3) + reshape_37 = None + linear_72 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_36 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_38 = linear_72.reshape(1, 197, 7, 64) + linear_72 = None + v_16 = reshape_38.permute(0, 2, 1, 3) + reshape_38 = None + transpose_20 = k_16.transpose(-2, -1) + k_16 = None + matmul_4 = q_16 @ transpose_20 + q_16 = transpose_20 = None + attn_6 = matmul_4 * 0.125 + matmul_4 = None + attn_7 = attn_6.softmax(dim=-1) + attn_6 = None + attn_8 = torch.nn.functional.dropout(attn_7, 0.0, False, False) + attn_7 = None + matmul_5 = attn_8 @ v_16 + attn_8 = v_16 = None + transpose_21 = matmul_5.transpose(1, 2) + matmul_5 = None + x_165 = transpose_21.reshape(1, 1, 448) + transpose_21 = None + x_166 = torch._C._nn.linear( + x_165, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_165 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_167 = torch.nn.functional.dropout(x_166, 0.0, False, False) + x_166 = None + x_168 = getitem_73 + x_167 + getitem_73 = x_167 = None + getitem_75 = x_168[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_168 = None + input_29 = torch.nn.functional.layer_norm( + getitem_75, + (448,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_75 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_30 = torch._C._nn.gelu(input_29, approximate="none") + input_29 = None + input_31 = torch._C._nn.linear( + input_30, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_30 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_76 = x_98[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_5 = torch.cat((input_31, getitem_76), dim=1) + input_31 = getitem_76 = None + getitem_77 = x_98[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_98 = None + tmp_6 = torch.cat((input_28, getitem_77), dim=1) + input_28 = getitem_77 = None + getitem_78 = tmp_6[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_38 = torch.nn.functional.layer_norm( + tmp_6, + (224,), + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_6 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_79 = layer_norm_38[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_75 = torch._C._nn.linear( + getitem_79, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_79 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_40 = linear_75.reshape(1, 1, 7, 32) + linear_75 = None + q_17 = reshape_40.permute(0, 2, 1, 3) + reshape_40 = None + linear_76 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_41 = linear_76.reshape(1, 401, 7, 32) + linear_76 = None + k_17 = reshape_41.permute(0, 2, 1, 3) + reshape_41 = None + linear_77 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_38 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_42 = linear_77.reshape(1, 401, 7, 32) + linear_77 = None + v_17 = reshape_42.permute(0, 2, 1, 3) + reshape_42 = None + transpose_22 = k_17.transpose(-2, -1) + k_17 = None + matmul_6 = q_17 @ transpose_22 + q_17 = transpose_22 = None + attn_9 = matmul_6 * 0.1767766952966369 + matmul_6 = None + attn_10 = attn_9.softmax(dim=-1) + attn_9 = None + attn_11 = torch.nn.functional.dropout(attn_10, 0.0, False, False) + attn_10 = None + matmul_7 = attn_11 @ v_17 + attn_11 = v_17 = None + transpose_23 = matmul_7.transpose(1, 2) + matmul_7 = None + x_169 = transpose_23.reshape(1, 1, 224) + transpose_23 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_169 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = getitem_78 + x_171 + getitem_78 = x_171 = None + getitem_80 = x_172[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_172 = None + input_32 = torch.nn.functional.layer_norm( + getitem_80, + (224,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_80 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_33 = torch._C._nn.gelu(input_32, approximate="none") + input_32 = None + input_34 = torch._C._nn.linear( + input_33, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_33 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_81 = x_164[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_164 = None + tmp_7 = torch.cat((input_34, getitem_81), dim=1) + input_34 = getitem_81 = None + layer_norm_40 = torch.nn.functional.layer_norm( + tmp_5, + (224,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_40 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_44 = linear_80.reshape(1, 401, 3, 7, 32) + linear_80 = None + qkv_14 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_18 = unbind_14[0] + k_18 = unbind_14[1] + v_18 = unbind_14[2] + unbind_14 = None + x_173 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_24 = x_173.transpose(1, 2) + x_173 = None + x_174 = transpose_24.reshape(1, 401, 224) + transpose_24 = None + x_175 = torch._C._nn.linear( + x_174, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_174 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = tmp_5 + x_176 + tmp_5 = x_176 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_177, + (224,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_178 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_41 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_179 = torch._C._nn.gelu(x_178, approximate="none") + x_178 = None + x_180 = torch.nn.functional.dropout(x_179, 0.0, False, False) + x_179 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_177 + x_182 + x_177 = x_182 = None + layer_norm_42 = torch.nn.functional.layer_norm( + tmp_7, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_84 = torch._C._nn.linear( + layer_norm_42, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_42 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_46 = linear_84.reshape(1, 197, 3, 7, 64) + linear_84 = None + qkv_15 = reshape_46.permute(2, 0, 3, 1, 4) + reshape_46 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_19 = unbind_15[0] + k_19 = unbind_15[1] + v_19 = unbind_15[2] + unbind_15 = None + x_184 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, attn_mask=None, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_25 = x_184.transpose(1, 2) + x_184 = None + x_185 = transpose_25.reshape(1, 197, 448) + transpose_25 = None + x_186 = torch._C._nn.linear( + x_185, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_185 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_187 = torch.nn.functional.dropout(x_186, 0.0, False, False) + x_186 = None + x_188 = tmp_7 + x_187 + tmp_7 = x_187 = None + layer_norm_43 = torch.nn.functional.layer_norm( + x_188, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_189 = torch._C._nn.linear( + layer_norm_43, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_43 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_190 = torch._C._nn.gelu(x_189, approximate="none") + x_189 = None + x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) + x_190 = None + x_192 = torch._C._nn.linear( + x_191, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_191 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_193 = torch.nn.functional.dropout(x_192, 0.0, False, False) + x_192 = None + x_194 = x_188 + x_193 + x_188 = x_193 = None + layer_norm_44 = torch.nn.functional.layer_norm( + x_194, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_88 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_44 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_48 = linear_88.reshape(1, 197, 3, 7, 64) + linear_88 = None + qkv_16 = reshape_48.permute(2, 0, 3, 1, 4) + reshape_48 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_20 = unbind_16[0] + k_20 = unbind_16[1] + v_20 = unbind_16[2] + unbind_16 = None + x_195 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, attn_mask=None, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_26 = x_195.transpose(1, 2) + x_195 = None + x_196 = transpose_26.reshape(1, 197, 448) + transpose_26 = None + x_197 = torch._C._nn.linear( + x_196, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_196 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_198 = torch.nn.functional.dropout(x_197, 0.0, False, False) + x_197 = None + x_199 = x_194 + x_198 + x_194 = x_198 = None + layer_norm_45 = torch.nn.functional.layer_norm( + x_199, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_200 = torch._C._nn.linear( + layer_norm_45, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_45 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_201 = torch._C._nn.gelu(x_200, approximate="none") + x_200 = None + x_202 = torch.nn.functional.dropout(x_201, 0.0, False, False) + x_201 = None + x_203 = torch._C._nn.linear( + x_202, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_202 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + x_205 = x_199 + x_204 + x_199 = x_204 = None + layer_norm_46 = torch.nn.functional.layer_norm( + x_205, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_92 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_46 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_50 = linear_92.reshape(1, 197, 3, 7, 64) + linear_92 = None + qkv_17 = reshape_50.permute(2, 0, 3, 1, 4) + reshape_50 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_21 = unbind_17[0] + k_21 = unbind_17[1] + v_21 = unbind_17[2] + unbind_17 = None + x_206 = torch._C._nn.scaled_dot_product_attention( + q_21, k_21, v_21, attn_mask=None, dropout_p=0.0 + ) + q_21 = k_21 = v_21 = None + transpose_27 = x_206.transpose(1, 2) + x_206 = None + x_207 = transpose_27.reshape(1, 197, 448) + transpose_27 = None + x_208 = torch._C._nn.linear( + x_207, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_207 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_209 = torch.nn.functional.dropout(x_208, 0.0, False, False) + x_208 = None + x_210 = x_205 + x_209 + x_205 = x_209 = None + layer_norm_47 = torch.nn.functional.layer_norm( + x_210, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_211 = torch._C._nn.linear( + layer_norm_47, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_47 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_212 = torch._C._nn.gelu(x_211, approximate="none") + x_211 = None + x_213 = torch.nn.functional.dropout(x_212, 0.0, False, False) + x_212 = None + x_214 = torch._C._nn.linear( + x_213, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_213 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_215 = torch.nn.functional.dropout(x_214, 0.0, False, False) + x_214 = None + x_216 = x_210 + x_215 + x_210 = x_215 = None + layer_norm_48 = torch.nn.functional.layer_norm( + x_216, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_96 = torch._C._nn.linear( + layer_norm_48, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_48 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_52 = linear_96.reshape(1, 197, 3, 7, 64) + linear_96 = None + qkv_18 = reshape_52.permute(2, 0, 3, 1, 4) + reshape_52 = None + unbind_18 = qkv_18.unbind(0) + qkv_18 = None + q_22 = unbind_18[0] + k_22 = unbind_18[1] + v_22 = unbind_18[2] + unbind_18 = None + x_217 = torch._C._nn.scaled_dot_product_attention( + q_22, k_22, v_22, attn_mask=None, dropout_p=0.0 + ) + q_22 = k_22 = v_22 = None + transpose_28 = x_217.transpose(1, 2) + x_217 = None + x_218 = transpose_28.reshape(1, 197, 448) + transpose_28 = None + x_219 = torch._C._nn.linear( + x_218, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_218 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_220 = torch.nn.functional.dropout(x_219, 0.0, False, False) + x_219 = None + x_221 = x_216 + x_220 + x_216 = x_220 = None + layer_norm_49 = torch.nn.functional.layer_norm( + x_221, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_222 = torch._C._nn.linear( + layer_norm_49, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_49 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_223 = torch._C._nn.gelu(x_222, approximate="none") + x_222 = None + x_224 = torch.nn.functional.dropout(x_223, 0.0, False, False) + x_223 = None + x_225 = torch._C._nn.linear( + x_224, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_224 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_226 = torch.nn.functional.dropout(x_225, 0.0, False, False) + x_225 = None + x_227 = x_221 + x_226 + x_221 = x_226 = None + layer_norm_50 = torch.nn.functional.layer_norm( + x_227, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_100 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_50 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_54 = linear_100.reshape(1, 197, 3, 7, 64) + linear_100 = None + qkv_19 = reshape_54.permute(2, 0, 3, 1, 4) + reshape_54 = None + unbind_19 = qkv_19.unbind(0) + qkv_19 = None + q_23 = unbind_19[0] + k_23 = unbind_19[1] + v_23 = unbind_19[2] + unbind_19 = None + x_228 = torch._C._nn.scaled_dot_product_attention( + q_23, k_23, v_23, attn_mask=None, dropout_p=0.0 + ) + q_23 = k_23 = v_23 = None + transpose_29 = x_228.transpose(1, 2) + x_228 = None + x_229 = transpose_29.reshape(1, 197, 448) + transpose_29 = None + x_230 = torch._C._nn.linear( + x_229, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_229 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_231 = torch.nn.functional.dropout(x_230, 0.0, False, False) + x_230 = None + x_232 = x_227 + x_231 + x_227 = x_231 = None + layer_norm_51 = torch.nn.functional.layer_norm( + x_232, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_233 = torch._C._nn.linear( + layer_norm_51, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_51 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_234 = torch._C._nn.gelu(x_233, approximate="none") + x_233 = None + x_235 = torch.nn.functional.dropout(x_234, 0.0, False, False) + x_234 = None + x_236 = torch._C._nn.linear( + x_235, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_235 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_237 = torch.nn.functional.dropout(x_236, 0.0, False, False) + x_236 = None + x_238 = x_232 + x_237 + x_232 = x_237 = None + layer_norm_52 = torch.nn.functional.layer_norm( + x_238, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = (None) + linear_104 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_52 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_56 = linear_104.reshape(1, 197, 3, 7, 64) + linear_104 = None + qkv_20 = reshape_56.permute(2, 0, 3, 1, 4) + reshape_56 = None + unbind_20 = qkv_20.unbind(0) + qkv_20 = None + q_24 = unbind_20[0] + k_24 = unbind_20[1] + v_24 = unbind_20[2] + unbind_20 = None + x_239 = torch._C._nn.scaled_dot_product_attention( + q_24, k_24, v_24, attn_mask=None, dropout_p=0.0 + ) + q_24 = k_24 = v_24 = None + transpose_30 = x_239.transpose(1, 2) + x_239 = None + x_240 = transpose_30.reshape(1, 197, 448) + transpose_30 = None + x_241 = torch._C._nn.linear( + x_240, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_240 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = (None) + x_242 = torch.nn.functional.dropout(x_241, 0.0, False, False) + x_241 = None + x_243 = x_238 + x_242 + x_238 = x_242 = None + layer_norm_53 = torch.nn.functional.layer_norm( + x_243, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = (None) + x_244 = torch._C._nn.linear( + layer_norm_53, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_53 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_245 = torch._C._nn.gelu(x_244, approximate="none") + x_244 = None + x_246 = torch.nn.functional.dropout(x_245, 0.0, False, False) + x_245 = None + x_247 = torch._C._nn.linear( + x_246, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_246 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_248 = torch.nn.functional.dropout(x_247, 0.0, False, False) + x_247 = None + x_249 = x_243 + x_248 + x_243 = x_248 = None + getitem_103 = x_183[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_35 = torch.nn.functional.layer_norm( + getitem_103, + (224,), + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_103 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_36 = torch._C._nn.gelu(input_35, approximate="none") + input_35 = None + input_37 = torch._C._nn.linear( + input_36, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_36 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_104 = x_249[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_38 = torch.nn.functional.layer_norm( + getitem_104, + (448,), + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_104 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_39 = torch._C._nn.gelu(input_38, approximate="none") + input_38 = None + input_40 = torch._C._nn.linear( + input_39, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_39 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_105 = x_249[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_8 = torch.cat((input_37, getitem_105), dim=1) + input_37 = getitem_105 = None + getitem_106 = tmp_8[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_56 = torch.nn.functional.layer_norm( + tmp_8, + (448,), + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_8 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_107 = layer_norm_56[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_110 = torch._C._nn.linear( + getitem_107, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_107 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_58 = linear_110.reshape(1, 1, 7, 64) + linear_110 = None + q_25 = reshape_58.permute(0, 2, 1, 3) + reshape_58 = None + linear_111 = torch._C._nn.linear( + layer_norm_56, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_59 = linear_111.reshape(1, 197, 7, 64) + linear_111 = None + k_25 = reshape_59.permute(0, 2, 1, 3) + reshape_59 = None + linear_112 = torch._C._nn.linear( + layer_norm_56, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_56 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_60 = linear_112.reshape(1, 197, 7, 64) + linear_112 = None + v_25 = reshape_60.permute(0, 2, 1, 3) + reshape_60 = None + transpose_31 = k_25.transpose(-2, -1) + k_25 = None + matmul_8 = q_25 @ transpose_31 + q_25 = transpose_31 = None + attn_12 = matmul_8 * 0.125 + matmul_8 = None + attn_13 = attn_12.softmax(dim=-1) + attn_12 = None + attn_14 = torch.nn.functional.dropout(attn_13, 0.0, False, False) + attn_13 = None + matmul_9 = attn_14 @ v_25 + attn_14 = v_25 = None + transpose_32 = matmul_9.transpose(1, 2) + matmul_9 = None + x_250 = transpose_32.reshape(1, 1, 448) + transpose_32 = None + x_251 = torch._C._nn.linear( + x_250, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_250 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_252 = torch.nn.functional.dropout(x_251, 0.0, False, False) + x_251 = None + x_253 = getitem_106 + x_252 + getitem_106 = x_252 = None + getitem_108 = x_253[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_253 = None + input_41 = torch.nn.functional.layer_norm( + getitem_108, + (448,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_108 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_42 = torch._C._nn.gelu(input_41, approximate="none") + input_41 = None + input_43 = torch._C._nn.linear( + input_42, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_42 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_109 = x_183[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_9 = torch.cat((input_43, getitem_109), dim=1) + input_43 = getitem_109 = None + getitem_110 = x_183[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_183 = None + tmp_10 = torch.cat((input_40, getitem_110), dim=1) + input_40 = getitem_110 = None + getitem_111 = tmp_10[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_58 = torch.nn.functional.layer_norm( + tmp_10, + (224,), + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_10 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_112 = layer_norm_58[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_115 = torch._C._nn.linear( + getitem_112, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_112 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_62 = linear_115.reshape(1, 1, 7, 32) + linear_115 = None + q_26 = reshape_62.permute(0, 2, 1, 3) + reshape_62 = None + linear_116 = torch._C._nn.linear( + layer_norm_58, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_63 = linear_116.reshape(1, 401, 7, 32) + linear_116 = None + k_26 = reshape_63.permute(0, 2, 1, 3) + reshape_63 = None + linear_117 = torch._C._nn.linear( + layer_norm_58, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_58 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_64 = linear_117.reshape(1, 401, 7, 32) + linear_117 = None + v_26 = reshape_64.permute(0, 2, 1, 3) + reshape_64 = None + transpose_33 = k_26.transpose(-2, -1) + k_26 = None + matmul_10 = q_26 @ transpose_33 + q_26 = transpose_33 = None + attn_15 = matmul_10 * 0.1767766952966369 + matmul_10 = None + attn_16 = attn_15.softmax(dim=-1) + attn_15 = None + attn_17 = torch.nn.functional.dropout(attn_16, 0.0, False, False) + attn_16 = None + matmul_11 = attn_17 @ v_26 + attn_17 = v_26 = None + transpose_34 = matmul_11.transpose(1, 2) + matmul_11 = None + x_254 = transpose_34.reshape(1, 1, 224) + transpose_34 = None + x_255 = torch._C._nn.linear( + x_254, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_254 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + x_257 = getitem_111 + x_256 + getitem_111 = x_256 = None + getitem_113 = x_257[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_257 = None + input_44 = torch.nn.functional.layer_norm( + getitem_113, + (224,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_113 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_45 = torch._C._nn.gelu(input_44, approximate="none") + input_44 = None + input_46 = torch._C._nn.linear( + input_45, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_45 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_114 = x_249[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_249 = None + tmp_11 = torch.cat((input_46, getitem_114), dim=1) + input_46 = getitem_114 = None + x_258 = torch.nn.functional.layer_norm( + tmp_9, + (224,), + l_self_modules_norm_modules_0_parameters_weight_, + l_self_modules_norm_modules_0_parameters_bias_, + 1e-06, + ) + tmp_9 = ( + l_self_modules_norm_modules_0_parameters_weight_ + ) = l_self_modules_norm_modules_0_parameters_bias_ = None + x_259 = torch.nn.functional.layer_norm( + tmp_11, + (448,), + l_self_modules_norm_modules_1_parameters_weight_, + l_self_modules_norm_modules_1_parameters_bias_, + 1e-06, + ) + tmp_11 = ( + l_self_modules_norm_modules_1_parameters_weight_ + ) = l_self_modules_norm_modules_1_parameters_bias_ = None + x_260 = x_258[(slice(None, None, None), 0)] + x_258 = None + x_261 = x_259[(slice(None, None, None), 0)] + x_259 = None + dropout_77 = torch.nn.functional.dropout(x_260, 0.0, False, False) + x_260 = None + dropout_78 = torch.nn.functional.dropout(x_261, 0.0, False, False) + x_261 = None + linear_120 = torch._C._nn.linear( + dropout_77, + l_self_modules_head_modules_0_parameters_weight_, + l_self_modules_head_modules_0_parameters_bias_, + ) + dropout_77 = ( + l_self_modules_head_modules_0_parameters_weight_ + ) = l_self_modules_head_modules_0_parameters_bias_ = None + linear_121 = torch._C._nn.linear( + dropout_78, + l_self_modules_head_modules_1_parameters_weight_, + l_self_modules_head_modules_1_parameters_bias_, + ) + dropout_78 = ( + l_self_modules_head_modules_1_parameters_weight_ + ) = l_self_modules_head_modules_1_parameters_bias_ = None + stack = torch.stack([linear_120, linear_121], dim=0) + linear_120 = linear_121 = None + x_262 = torch.mean(stack, dim=0) + stack = None + return (x_262,) diff --git a/samples/timm/crossvit_18_dagger_240.in1k/weight_meta.py b/samples/timm/crossvit_18_dagger_240.in1k/weight_meta.py new file mode 100644 index 0000000000..564b4082d4 --- /dev/null +++ b/samples/timm/crossvit_18_dagger_240.in1k/weight_meta.py @@ -0,0 +1,3872 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.228 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_" + ) + shape = [56, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_" + ) + shape = [56] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_" + ) + shape = [112, 56, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_" + ) + shape = [112] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_" + ) + shape = [224, 112, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_" + ) + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_0_: + name = "L_self_parameters_cls_token_0_" + shape = [1, 1, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_0_: + name = "L_self_parameters_pos_embed_0_" + shape = [1, 401, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_" + ) + shape = [112, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_" + ) + shape = [112] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.011 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_" + ) + shape = [224, 112, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_" + ) + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_" + ) + shape = [448, 224, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.013 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_" + ) + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.012 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_1_: + name = "L_self_parameters_cls_token_1_" + shape = [1, 1, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_1_: + name = "L_self_parameters_pos_embed_1_" + shape = [1, 197, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [224, 672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [224, 672] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [224, 672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_weight_: + name = "L_self_modules_norm_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_bias_: + name = "L_self_modules_norm_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_weight_: + name = "L_self_modules_norm_modules_1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_bias_: + name = "L_self_modules_norm_modules_1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_weight_: + name = "L_self_modules_head_modules_0_parameters_weight_" + shape = [1000, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_bias_: + name = "L_self_modules_head_modules_0_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_weight_: + name = "L_self_modules_head_modules_1_parameters_weight_" + shape = [1000, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_bias_: + name = "L_self_modules_head_modules_1_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/crossvit_18_dagger_408.in1k/graph_hash.txt b/samples/timm/crossvit_18_dagger_408.in1k/graph_hash.txt new file mode 100644 index 0000000000..98580fb365 --- /dev/null +++ b/samples/timm/crossvit_18_dagger_408.in1k/graph_hash.txt @@ -0,0 +1 @@ +f22e958a02b95aa6cfc33f02b15734dc4dbdacecfe968467dedfb62f0b896cde \ No newline at end of file diff --git a/samples/timm/crossvit_18_dagger_408.in1k/graph_net.json b/samples/timm/crossvit_18_dagger_408.in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/crossvit_18_dagger_408.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/crossvit_18_dagger_408.in1k/input_meta.py b/samples/timm/crossvit_18_dagger_408.in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/crossvit_18_dagger_408.in1k/input_tensor_constraints.py b/samples/timm/crossvit_18_dagger_408.in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/crossvit_18_dagger_408.in1k/model.py b/samples/timm/crossvit_18_dagger_408.in1k/model.py new file mode 100644 index 0000000000..ea63f99b24 --- /dev/null +++ b/samples/timm/crossvit_18_dagger_408.in1k/model.py @@ -0,0 +1,3096 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_0_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_0_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_1_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ + ) + l_self_parameters_cls_token_0_ = L_self_parameters_cls_token_0_ + l_self_parameters_pos_embed_0_ = L_self_parameters_pos_embed_0_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ + ) + l_self_parameters_cls_token_1_ = L_self_parameters_cls_token_1_ + l_self_parameters_pos_embed_1_ = L_self_parameters_pos_embed_1_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_norm_modules_0_parameters_weight_ = ( + L_self_modules_norm_modules_0_parameters_weight_ + ) + l_self_modules_norm_modules_0_parameters_bias_ = ( + L_self_modules_norm_modules_0_parameters_bias_ + ) + l_self_modules_norm_modules_1_parameters_weight_ = ( + L_self_modules_norm_modules_1_parameters_weight_ + ) + l_self_modules_norm_modules_1_parameters_bias_ = ( + L_self_modules_norm_modules_1_parameters_bias_ + ) + l_self_modules_head_modules_0_parameters_weight_ = ( + L_self_modules_head_modules_0_parameters_weight_ + ) + l_self_modules_head_modules_0_parameters_bias_ = ( + L_self_modules_head_modules_0_parameters_bias_ + ) + l_self_modules_head_modules_1_parameters_weight_ = ( + L_self_modules_head_modules_1_parameters_weight_ + ) + l_self_modules_head_modules_1_parameters_bias_ = ( + L_self_modules_head_modules_1_parameters_bias_ + ) + x = torch.nn.functional.interpolate( + l_x_, size=(408, 408), mode="bicubic", align_corners=False + ) + input_1 = torch.conv2d( + x, + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_, + (4, 4), + (3, 3), + (1, 1), + 1, + ) + x = l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ + ) = None + input_2 = torch.nn.functional.relu(input_1, inplace=True) + input_1 = None + input_3 = torch.conv2d( + input_2, + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_, + (3, 3), + (0, 0), + (1, 1), + 1, + ) + input_2 = l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ + ) = None + input_4 = torch.nn.functional.relu(input_3, inplace=True) + input_3 = None + input_5 = torch.conv2d( + input_4, + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + input_4 = l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ + ) = None + flatten = input_5.flatten(2) + input_5 = None + x_1 = flatten.transpose(1, 2) + flatten = None + cls_tokens = l_self_parameters_cls_token_0_.expand(1, -1, -1) + l_self_parameters_cls_token_0_ = None + x_ = torch.cat((cls_tokens, x_1), dim=1) + cls_tokens = x_1 = None + x__1 = x_ + l_self_parameters_pos_embed_0_ + x_ = l_self_parameters_pos_embed_0_ = None + x__2 = torch.nn.functional.dropout(x__1, 0.0, False, False) + x__1 = None + x_2 = torch.nn.functional.interpolate( + l_x_, size=(384, 384), mode="bicubic", align_corners=False + ) + l_x_ = None + input_6 = torch.conv2d( + x_2, + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_, + (4, 4), + (3, 3), + (1, 1), + 1, + ) + x_2 = l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ + ) = None + input_7 = torch.nn.functional.relu(input_6, inplace=True) + input_6 = None + input_8 = torch.conv2d( + input_7, + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_7 = l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ + ) = None + input_9 = torch.nn.functional.relu(input_8, inplace=True) + input_8 = None + input_10 = torch.conv2d( + input_9, + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_9 = l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ + ) = None + flatten_1 = input_10.flatten(2) + input_10 = None + x_3 = flatten_1.transpose(1, 2) + flatten_1 = None + cls_tokens_1 = l_self_parameters_cls_token_1_.expand(1, -1, -1) + l_self_parameters_cls_token_1_ = None + x__3 = torch.cat((cls_tokens_1, x_3), dim=1) + cls_tokens_1 = x_3 = None + x__4 = x__3 + l_self_parameters_pos_embed_1_ + x__3 = l_self_parameters_pos_embed_1_ = None + x__5 = torch.nn.functional.dropout(x__4, 0.0, False, False) + x__4 = None + layer_norm = torch.nn.functional.layer_norm( + x__2, + (224,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear = torch._C._nn.linear( + layer_norm, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 1157, 3, 7, 32) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_4 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_2 = x_4.transpose(1, 2) + x_4 = None + x_5 = transpose_2.reshape(1, 1157, 224) + transpose_2 = None + x_6 = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_7 = torch.nn.functional.dropout(x_6, 0.0, False, False) + x_6 = None + x_8 = x__2 + x_7 + x__2 = x_7 = None + layer_norm_1 = torch.nn.functional.layer_norm( + x_8, + (224,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_9 = torch._C._nn.linear( + layer_norm_1, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_1 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_10 = torch._C._nn.gelu(x_9, approximate="none") + x_9 = None + x_11 = torch.nn.functional.dropout(x_10, 0.0, False, False) + x_10 = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_11 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_13 = torch.nn.functional.dropout(x_12, 0.0, False, False) + x_12 = None + x_14 = x_8 + x_13 + x_8 = x_13 = None + layer_norm_2 = torch.nn.functional.layer_norm( + x__5, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_2, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_2 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_2 = linear_4.reshape(1, 577, 3, 7, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_15 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_3 = x_15.transpose(1, 2) + x_15 = None + x_16 = transpose_3.reshape(1, 577, 448) + transpose_3 = None + x_17 = torch._C._nn.linear( + x_16, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_16 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_18 = torch.nn.functional.dropout(x_17, 0.0, False, False) + x_17 = None + x_19 = x__5 + x_18 + x__5 = x_18 = None + layer_norm_3 = torch.nn.functional.layer_norm( + x_19, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_20 = torch._C._nn.linear( + layer_norm_3, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_3 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_21 = torch._C._nn.gelu(x_20, approximate="none") + x_20 = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = torch._C._nn.linear( + x_22, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_22 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_24 = torch.nn.functional.dropout(x_23, 0.0, False, False) + x_23 = None + x_25 = x_19 + x_24 + x_19 = x_24 = None + layer_norm_4 = torch.nn.functional.layer_norm( + x_25, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_8.reshape(1, 577, 3, 7, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_26 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_4 = x_26.transpose(1, 2) + x_26 = None + x_27 = transpose_4.reshape(1, 577, 448) + transpose_4 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_27 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + x_30 = x_25 + x_29 + x_25 = x_29 = None + layer_norm_5 = torch.nn.functional.layer_norm( + x_30, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_31 = torch._C._nn.linear( + layer_norm_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_32 = torch._C._nn.gelu(x_31, approximate="none") + x_31 = None + x_33 = torch.nn.functional.dropout(x_32, 0.0, False, False) + x_32 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_33 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + x_36 = x_30 + x_35 + x_30 = x_35 = None + layer_norm_6 = torch.nn.functional.layer_norm( + x_36, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_6, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_6 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_6 = linear_12.reshape(1, 577, 3, 7, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_37 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_5 = x_37.transpose(1, 2) + x_37 = None + x_38 = transpose_5.reshape(1, 577, 448) + transpose_5 = None + x_39 = torch._C._nn.linear( + x_38, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_38 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = x_36 + x_40 + x_36 = x_40 = None + layer_norm_7 = torch.nn.functional.layer_norm( + x_41, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_42 = torch._C._nn.linear( + layer_norm_7, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_7 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_43 = torch._C._nn.gelu(x_42, approximate="none") + x_42 = None + x_44 = torch.nn.functional.dropout(x_43, 0.0, False, False) + x_43 = None + x_45 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_44 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_46 = torch.nn.functional.dropout(x_45, 0.0, False, False) + x_45 = None + x_47 = x_41 + x_46 + x_41 = x_46 = None + layer_norm_8 = torch.nn.functional.layer_norm( + x_47, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_8, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_8 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_8 = linear_16.reshape(1, 577, 3, 7, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_48 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_6 = x_48.transpose(1, 2) + x_48 = None + x_49 = transpose_6.reshape(1, 577, 448) + transpose_6 = None + x_50 = torch._C._nn.linear( + x_49, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_49 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_51 = torch.nn.functional.dropout(x_50, 0.0, False, False) + x_50 = None + x_52 = x_47 + x_51 + x_47 = x_51 = None + layer_norm_9 = torch.nn.functional.layer_norm( + x_52, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_53 = torch._C._nn.linear( + layer_norm_9, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_9 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_54 = torch._C._nn.gelu(x_53, approximate="none") + x_53 = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = torch._C._nn.linear( + x_55, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_55 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_57 = torch.nn.functional.dropout(x_56, 0.0, False, False) + x_56 = None + x_58 = x_52 + x_57 + x_52 = x_57 = None + layer_norm_10 = torch.nn.functional.layer_norm( + x_58, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_20 = torch._C._nn.linear( + layer_norm_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_10 = linear_20.reshape(1, 577, 3, 7, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_59 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_7 = x_59.transpose(1, 2) + x_59 = None + x_60 = transpose_7.reshape(1, 577, 448) + transpose_7 = None + x_61 = torch._C._nn.linear( + x_60, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_60 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + x_63 = x_58 + x_62 + x_58 = x_62 = None + layer_norm_11 = torch.nn.functional.layer_norm( + x_63, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_64 = torch._C._nn.linear( + layer_norm_11, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_11 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_63 + x_68 + x_63 = x_68 = None + layer_norm_12 = torch.nn.functional.layer_norm( + x_69, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = (None) + linear_24 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_12 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_12 = linear_24.reshape(1, 577, 3, 7, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_70 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_8 = x_70.transpose(1, 2) + x_70 = None + x_71 = transpose_8.reshape(1, 577, 448) + transpose_8 = None + x_72 = torch._C._nn.linear( + x_71, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_71 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = (None) + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = x_69 + x_73 + x_69 = x_73 = None + layer_norm_13 = torch.nn.functional.layer_norm( + x_74, + (448,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = (None) + x_75 = torch._C._nn.linear( + layer_norm_13, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_13 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_76 = torch._C._nn.gelu(x_75, approximate="none") + x_75 = None + x_77 = torch.nn.functional.dropout(x_76, 0.0, False, False) + x_76 = None + x_78 = torch._C._nn.linear( + x_77, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_77 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = x_74 + x_79 + x_74 = x_79 = None + getitem_37 = x_14[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_11 = torch.nn.functional.layer_norm( + getitem_37, + (224,), + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_37 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_12 = torch._C._nn.gelu(input_11, approximate="none") + input_11 = None + input_13 = torch._C._nn.linear( + input_12, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_12 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_38 = x_80[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_14 = torch.nn.functional.layer_norm( + getitem_38, + (448,), + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_38 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_15 = torch._C._nn.gelu(input_14, approximate="none") + input_14 = None + input_16 = torch._C._nn.linear( + input_15, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_15 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_39 = x_80[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp = torch.cat((input_13, getitem_39), dim=1) + input_13 = getitem_39 = None + getitem_40 = tmp[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_16 = torch.nn.functional.layer_norm( + tmp, + (448,), + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_41 = layer_norm_16[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_30 = torch._C._nn.linear( + getitem_41, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_41 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_14 = linear_30.reshape(1, 1, 7, 64) + linear_30 = None + q_7 = reshape_14.permute(0, 2, 1, 3) + reshape_14 = None + linear_31 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_15 = linear_31.reshape(1, 577, 7, 64) + linear_31 = None + k_7 = reshape_15.permute(0, 2, 1, 3) + reshape_15 = None + linear_32 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_16 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_16 = linear_32.reshape(1, 577, 7, 64) + linear_32 = None + v_7 = reshape_16.permute(0, 2, 1, 3) + reshape_16 = None + transpose_9 = k_7.transpose(-2, -1) + k_7 = None + matmul = q_7 @ transpose_9 + q_7 = transpose_9 = None + attn = matmul * 0.125 + matmul = None + attn_1 = attn.softmax(dim=-1) + attn = None + attn_2 = torch.nn.functional.dropout(attn_1, 0.0, False, False) + attn_1 = None + matmul_1 = attn_2 @ v_7 + attn_2 = v_7 = None + transpose_10 = matmul_1.transpose(1, 2) + matmul_1 = None + x_81 = transpose_10.reshape(1, 1, 448) + transpose_10 = None + x_82 = torch._C._nn.linear( + x_81, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_81 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_83 = torch.nn.functional.dropout(x_82, 0.0, False, False) + x_82 = None + x_84 = getitem_40 + x_83 + getitem_40 = x_83 = None + getitem_42 = x_84[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_84 = None + input_17 = torch.nn.functional.layer_norm( + getitem_42, + (448,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_42 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_18 = torch._C._nn.gelu(input_17, approximate="none") + input_17 = None + input_19 = torch._C._nn.linear( + input_18, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_18 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_43 = x_14[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_1 = torch.cat((input_19, getitem_43), dim=1) + input_19 = getitem_43 = None + getitem_44 = x_14[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_14 = None + tmp_2 = torch.cat((input_16, getitem_44), dim=1) + input_16 = getitem_44 = None + getitem_45 = tmp_2[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_18 = torch.nn.functional.layer_norm( + tmp_2, + (224,), + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_2 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_46 = layer_norm_18[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_35 = torch._C._nn.linear( + getitem_46, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_46 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_18 = linear_35.reshape(1, 1, 7, 32) + linear_35 = None + q_8 = reshape_18.permute(0, 2, 1, 3) + reshape_18 = None + linear_36 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_19 = linear_36.reshape(1, 1157, 7, 32) + linear_36 = None + k_8 = reshape_19.permute(0, 2, 1, 3) + reshape_19 = None + linear_37 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_18 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_20 = linear_37.reshape(1, 1157, 7, 32) + linear_37 = None + v_8 = reshape_20.permute(0, 2, 1, 3) + reshape_20 = None + transpose_11 = k_8.transpose(-2, -1) + k_8 = None + matmul_2 = q_8 @ transpose_11 + q_8 = transpose_11 = None + attn_3 = matmul_2 * 0.1767766952966369 + matmul_2 = None + attn_4 = attn_3.softmax(dim=-1) + attn_3 = None + attn_5 = torch.nn.functional.dropout(attn_4, 0.0, False, False) + attn_4 = None + matmul_3 = attn_5 @ v_8 + attn_5 = v_8 = None + transpose_12 = matmul_3.transpose(1, 2) + matmul_3 = None + x_85 = transpose_12.reshape(1, 1, 224) + transpose_12 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_85 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + x_88 = getitem_45 + x_87 + getitem_45 = x_87 = None + getitem_47 = x_88[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_88 = None + input_20 = torch.nn.functional.layer_norm( + getitem_47, + (224,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_47 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_21 = torch._C._nn.gelu(input_20, approximate="none") + input_20 = None + input_22 = torch._C._nn.linear( + input_21, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_21 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_48 = x_80[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_80 = None + tmp_3 = torch.cat((input_22, getitem_48), dim=1) + input_22 = getitem_48 = None + layer_norm_20 = torch.nn.functional.layer_norm( + tmp_1, + (224,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_20 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_22 = linear_40.reshape(1, 1157, 3, 7, 32) + linear_40 = None + qkv_7 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_9 = unbind_7[0] + k_9 = unbind_7[1] + v_9 = unbind_7[2] + unbind_7 = None + x_89 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_13 = x_89.transpose(1, 2) + x_89 = None + x_90 = transpose_13.reshape(1, 1157, 224) + transpose_13 = None + x_91 = torch._C._nn.linear( + x_90, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_90 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = tmp_1 + x_92 + tmp_1 = x_92 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_93, + (224,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_94 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_21 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_95 = torch._C._nn.gelu(x_94, approximate="none") + x_94 = None + x_96 = torch.nn.functional.dropout(x_95, 0.0, False, False) + x_95 = None + x_97 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_96 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_98 = torch.nn.functional.dropout(x_97, 0.0, False, False) + x_97 = None + x_99 = x_93 + x_98 + x_93 = x_98 = None + layer_norm_22 = torch.nn.functional.layer_norm( + tmp_3, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_22, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_22 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_24 = linear_44.reshape(1, 577, 3, 7, 64) + linear_44 = None + qkv_8 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_10 = unbind_8[0] + k_10 = unbind_8[1] + v_10 = unbind_8[2] + unbind_8 = None + x_100 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_14 = x_100.transpose(1, 2) + x_100 = None + x_101 = transpose_14.reshape(1, 577, 448) + transpose_14 = None + x_102 = torch._C._nn.linear( + x_101, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_101 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_103 = torch.nn.functional.dropout(x_102, 0.0, False, False) + x_102 = None + x_104 = tmp_3 + x_103 + tmp_3 = x_103 = None + layer_norm_23 = torch.nn.functional.layer_norm( + x_104, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_105 = torch._C._nn.linear( + layer_norm_23, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_23 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_106 = torch._C._nn.gelu(x_105, approximate="none") + x_105 = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + x_108 = torch._C._nn.linear( + x_107, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_107 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_109 = torch.nn.functional.dropout(x_108, 0.0, False, False) + x_108 = None + x_110 = x_104 + x_109 + x_104 = x_109 = None + layer_norm_24 = torch.nn.functional.layer_norm( + x_110, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_24 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_26 = linear_48.reshape(1, 577, 3, 7, 64) + linear_48 = None + qkv_9 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_11 = unbind_9[0] + k_11 = unbind_9[1] + v_11 = unbind_9[2] + unbind_9 = None + x_111 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_15 = x_111.transpose(1, 2) + x_111 = None + x_112 = transpose_15.reshape(1, 577, 448) + transpose_15 = None + x_113 = torch._C._nn.linear( + x_112, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_112 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_115 = x_110 + x_114 + x_110 = x_114 = None + layer_norm_25 = torch.nn.functional.layer_norm( + x_115, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_116 = torch._C._nn.linear( + layer_norm_25, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_25 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + x_121 = x_115 + x_120 + x_115 = x_120 = None + layer_norm_26 = torch.nn.functional.layer_norm( + x_121, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_52 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_26 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_28 = linear_52.reshape(1, 577, 3, 7, 64) + linear_52 = None + qkv_10 = reshape_28.permute(2, 0, 3, 1, 4) + reshape_28 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_12 = unbind_10[0] + k_12 = unbind_10[1] + v_12 = unbind_10[2] + unbind_10 = None + x_122 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_16 = x_122.transpose(1, 2) + x_122 = None + x_123 = transpose_16.reshape(1, 577, 448) + transpose_16 = None + x_124 = torch._C._nn.linear( + x_123, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_123 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_126 = x_121 + x_125 + x_121 = x_125 = None + layer_norm_27 = torch.nn.functional.layer_norm( + x_126, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_127 = torch._C._nn.linear( + layer_norm_27, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_27 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_128 = torch._C._nn.gelu(x_127, approximate="none") + x_127 = None + x_129 = torch.nn.functional.dropout(x_128, 0.0, False, False) + x_128 = None + x_130 = torch._C._nn.linear( + x_129, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_129 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = x_126 + x_131 + x_126 = x_131 = None + layer_norm_28 = torch.nn.functional.layer_norm( + x_132, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_28 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_30 = linear_56.reshape(1, 577, 3, 7, 64) + linear_56 = None + qkv_11 = reshape_30.permute(2, 0, 3, 1, 4) + reshape_30 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_13 = unbind_11[0] + k_13 = unbind_11[1] + v_13 = unbind_11[2] + unbind_11 = None + x_133 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_17 = x_133.transpose(1, 2) + x_133 = None + x_134 = transpose_17.reshape(1, 577, 448) + transpose_17 = None + x_135 = torch._C._nn.linear( + x_134, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_134 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + x_137 = x_132 + x_136 + x_132 = x_136 = None + layer_norm_29 = torch.nn.functional.layer_norm( + x_137, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_138 = torch._C._nn.linear( + layer_norm_29, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_29 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_139 = torch._C._nn.gelu(x_138, approximate="none") + x_138 = None + x_140 = torch.nn.functional.dropout(x_139, 0.0, False, False) + x_139 = None + x_141 = torch._C._nn.linear( + x_140, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_140 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = x_137 + x_142 + x_137 = x_142 = None + layer_norm_30 = torch.nn.functional.layer_norm( + x_143, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_60 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_30 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_32 = linear_60.reshape(1, 577, 3, 7, 64) + linear_60 = None + qkv_12 = reshape_32.permute(2, 0, 3, 1, 4) + reshape_32 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_14 = unbind_12[0] + k_14 = unbind_12[1] + v_14 = unbind_12[2] + unbind_12 = None + x_144 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_18 = x_144.transpose(1, 2) + x_144 = None + x_145 = transpose_18.reshape(1, 577, 448) + transpose_18 = None + x_146 = torch._C._nn.linear( + x_145, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_145 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_147 = torch.nn.functional.dropout(x_146, 0.0, False, False) + x_146 = None + x_148 = x_143 + x_147 + x_143 = x_147 = None + layer_norm_31 = torch.nn.functional.layer_norm( + x_148, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_149 = torch._C._nn.linear( + layer_norm_31, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_31 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_150 = torch._C._nn.gelu(x_149, approximate="none") + x_149 = None + x_151 = torch.nn.functional.dropout(x_150, 0.0, False, False) + x_150 = None + x_152 = torch._C._nn.linear( + x_151, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_151 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_153 = torch.nn.functional.dropout(x_152, 0.0, False, False) + x_152 = None + x_154 = x_148 + x_153 + x_148 = x_153 = None + layer_norm_32 = torch.nn.functional.layer_norm( + x_154, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_32 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_34 = linear_64.reshape(1, 577, 3, 7, 64) + linear_64 = None + qkv_13 = reshape_34.permute(2, 0, 3, 1, 4) + reshape_34 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_15 = unbind_13[0] + k_15 = unbind_13[1] + v_15 = unbind_13[2] + unbind_13 = None + x_155 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_19 = x_155.transpose(1, 2) + x_155 = None + x_156 = transpose_19.reshape(1, 577, 448) + transpose_19 = None + x_157 = torch._C._nn.linear( + x_156, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_156 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = (None) + x_158 = torch.nn.functional.dropout(x_157, 0.0, False, False) + x_157 = None + x_159 = x_154 + x_158 + x_154 = x_158 = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_159, + (448,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = (None) + x_160 = torch._C._nn.linear( + layer_norm_33, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_33 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_161 = torch._C._nn.gelu(x_160, approximate="none") + x_160 = None + x_162 = torch.nn.functional.dropout(x_161, 0.0, False, False) + x_161 = None + x_163 = torch._C._nn.linear( + x_162, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_162 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_164 = torch.nn.functional.dropout(x_163, 0.0, False, False) + x_163 = None + x_165 = x_159 + x_164 + x_159 = x_164 = None + getitem_70 = x_99[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_23 = torch.nn.functional.layer_norm( + getitem_70, + (224,), + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_70 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_24 = torch._C._nn.gelu(input_23, approximate="none") + input_23 = None + input_25 = torch._C._nn.linear( + input_24, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_24 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_71 = x_165[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_26 = torch.nn.functional.layer_norm( + getitem_71, + (448,), + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_71 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_27 = torch._C._nn.gelu(input_26, approximate="none") + input_26 = None + input_28 = torch._C._nn.linear( + input_27, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_27 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_72 = x_165[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_4 = torch.cat((input_25, getitem_72), dim=1) + input_25 = getitem_72 = None + getitem_73 = tmp_4[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_36 = torch.nn.functional.layer_norm( + tmp_4, + (448,), + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_4 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_74 = layer_norm_36[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_70 = torch._C._nn.linear( + getitem_74, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_74 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_36 = linear_70.reshape(1, 1, 7, 64) + linear_70 = None + q_16 = reshape_36.permute(0, 2, 1, 3) + reshape_36 = None + linear_71 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_37 = linear_71.reshape(1, 577, 7, 64) + linear_71 = None + k_16 = reshape_37.permute(0, 2, 1, 3) + reshape_37 = None + linear_72 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_36 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_38 = linear_72.reshape(1, 577, 7, 64) + linear_72 = None + v_16 = reshape_38.permute(0, 2, 1, 3) + reshape_38 = None + transpose_20 = k_16.transpose(-2, -1) + k_16 = None + matmul_4 = q_16 @ transpose_20 + q_16 = transpose_20 = None + attn_6 = matmul_4 * 0.125 + matmul_4 = None + attn_7 = attn_6.softmax(dim=-1) + attn_6 = None + attn_8 = torch.nn.functional.dropout(attn_7, 0.0, False, False) + attn_7 = None + matmul_5 = attn_8 @ v_16 + attn_8 = v_16 = None + transpose_21 = matmul_5.transpose(1, 2) + matmul_5 = None + x_166 = transpose_21.reshape(1, 1, 448) + transpose_21 = None + x_167 = torch._C._nn.linear( + x_166, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_166 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_168 = torch.nn.functional.dropout(x_167, 0.0, False, False) + x_167 = None + x_169 = getitem_73 + x_168 + getitem_73 = x_168 = None + getitem_75 = x_169[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_169 = None + input_29 = torch.nn.functional.layer_norm( + getitem_75, + (448,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_75 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_30 = torch._C._nn.gelu(input_29, approximate="none") + input_29 = None + input_31 = torch._C._nn.linear( + input_30, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_30 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_76 = x_99[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_5 = torch.cat((input_31, getitem_76), dim=1) + input_31 = getitem_76 = None + getitem_77 = x_99[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_99 = None + tmp_6 = torch.cat((input_28, getitem_77), dim=1) + input_28 = getitem_77 = None + getitem_78 = tmp_6[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_38 = torch.nn.functional.layer_norm( + tmp_6, + (224,), + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_6 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_79 = layer_norm_38[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_75 = torch._C._nn.linear( + getitem_79, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_79 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_40 = linear_75.reshape(1, 1, 7, 32) + linear_75 = None + q_17 = reshape_40.permute(0, 2, 1, 3) + reshape_40 = None + linear_76 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_41 = linear_76.reshape(1, 1157, 7, 32) + linear_76 = None + k_17 = reshape_41.permute(0, 2, 1, 3) + reshape_41 = None + linear_77 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_38 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_42 = linear_77.reshape(1, 1157, 7, 32) + linear_77 = None + v_17 = reshape_42.permute(0, 2, 1, 3) + reshape_42 = None + transpose_22 = k_17.transpose(-2, -1) + k_17 = None + matmul_6 = q_17 @ transpose_22 + q_17 = transpose_22 = None + attn_9 = matmul_6 * 0.1767766952966369 + matmul_6 = None + attn_10 = attn_9.softmax(dim=-1) + attn_9 = None + attn_11 = torch.nn.functional.dropout(attn_10, 0.0, False, False) + attn_10 = None + matmul_7 = attn_11 @ v_17 + attn_11 = v_17 = None + transpose_23 = matmul_7.transpose(1, 2) + matmul_7 = None + x_170 = transpose_23.reshape(1, 1, 224) + transpose_23 = None + x_171 = torch._C._nn.linear( + x_170, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_170 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_172 = torch.nn.functional.dropout(x_171, 0.0, False, False) + x_171 = None + x_173 = getitem_78 + x_172 + getitem_78 = x_172 = None + getitem_80 = x_173[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_173 = None + input_32 = torch.nn.functional.layer_norm( + getitem_80, + (224,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_80 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_33 = torch._C._nn.gelu(input_32, approximate="none") + input_32 = None + input_34 = torch._C._nn.linear( + input_33, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_33 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_81 = x_165[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_165 = None + tmp_7 = torch.cat((input_34, getitem_81), dim=1) + input_34 = getitem_81 = None + layer_norm_40 = torch.nn.functional.layer_norm( + tmp_5, + (224,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_40 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_44 = linear_80.reshape(1, 1157, 3, 7, 32) + linear_80 = None + qkv_14 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_18 = unbind_14[0] + k_18 = unbind_14[1] + v_18 = unbind_14[2] + unbind_14 = None + x_174 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_24 = x_174.transpose(1, 2) + x_174 = None + x_175 = transpose_24.reshape(1, 1157, 224) + transpose_24 = None + x_176 = torch._C._nn.linear( + x_175, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_175 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_177 = torch.nn.functional.dropout(x_176, 0.0, False, False) + x_176 = None + x_178 = tmp_5 + x_177 + tmp_5 = x_177 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_178, + (224,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_179 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_41 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_180 = torch._C._nn.gelu(x_179, approximate="none") + x_179 = None + x_181 = torch.nn.functional.dropout(x_180, 0.0, False, False) + x_180 = None + x_182 = torch._C._nn.linear( + x_181, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_181 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_183 = torch.nn.functional.dropout(x_182, 0.0, False, False) + x_182 = None + x_184 = x_178 + x_183 + x_178 = x_183 = None + layer_norm_42 = torch.nn.functional.layer_norm( + tmp_7, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_84 = torch._C._nn.linear( + layer_norm_42, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_42 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_46 = linear_84.reshape(1, 577, 3, 7, 64) + linear_84 = None + qkv_15 = reshape_46.permute(2, 0, 3, 1, 4) + reshape_46 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_19 = unbind_15[0] + k_19 = unbind_15[1] + v_19 = unbind_15[2] + unbind_15 = None + x_185 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, attn_mask=None, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_25 = x_185.transpose(1, 2) + x_185 = None + x_186 = transpose_25.reshape(1, 577, 448) + transpose_25 = None + x_187 = torch._C._nn.linear( + x_186, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_186 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_188 = torch.nn.functional.dropout(x_187, 0.0, False, False) + x_187 = None + x_189 = tmp_7 + x_188 + tmp_7 = x_188 = None + layer_norm_43 = torch.nn.functional.layer_norm( + x_189, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_190 = torch._C._nn.linear( + layer_norm_43, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_43 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_191 = torch._C._nn.gelu(x_190, approximate="none") + x_190 = None + x_192 = torch.nn.functional.dropout(x_191, 0.0, False, False) + x_191 = None + x_193 = torch._C._nn.linear( + x_192, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_192 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_194 = torch.nn.functional.dropout(x_193, 0.0, False, False) + x_193 = None + x_195 = x_189 + x_194 + x_189 = x_194 = None + layer_norm_44 = torch.nn.functional.layer_norm( + x_195, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_88 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_44 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_48 = linear_88.reshape(1, 577, 3, 7, 64) + linear_88 = None + qkv_16 = reshape_48.permute(2, 0, 3, 1, 4) + reshape_48 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_20 = unbind_16[0] + k_20 = unbind_16[1] + v_20 = unbind_16[2] + unbind_16 = None + x_196 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, attn_mask=None, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_26 = x_196.transpose(1, 2) + x_196 = None + x_197 = transpose_26.reshape(1, 577, 448) + transpose_26 = None + x_198 = torch._C._nn.linear( + x_197, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_197 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_199 = torch.nn.functional.dropout(x_198, 0.0, False, False) + x_198 = None + x_200 = x_195 + x_199 + x_195 = x_199 = None + layer_norm_45 = torch.nn.functional.layer_norm( + x_200, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_201 = torch._C._nn.linear( + layer_norm_45, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_45 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_202 = torch._C._nn.gelu(x_201, approximate="none") + x_201 = None + x_203 = torch.nn.functional.dropout(x_202, 0.0, False, False) + x_202 = None + x_204 = torch._C._nn.linear( + x_203, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_203 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_205 = torch.nn.functional.dropout(x_204, 0.0, False, False) + x_204 = None + x_206 = x_200 + x_205 + x_200 = x_205 = None + layer_norm_46 = torch.nn.functional.layer_norm( + x_206, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_92 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_46 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_50 = linear_92.reshape(1, 577, 3, 7, 64) + linear_92 = None + qkv_17 = reshape_50.permute(2, 0, 3, 1, 4) + reshape_50 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_21 = unbind_17[0] + k_21 = unbind_17[1] + v_21 = unbind_17[2] + unbind_17 = None + x_207 = torch._C._nn.scaled_dot_product_attention( + q_21, k_21, v_21, attn_mask=None, dropout_p=0.0 + ) + q_21 = k_21 = v_21 = None + transpose_27 = x_207.transpose(1, 2) + x_207 = None + x_208 = transpose_27.reshape(1, 577, 448) + transpose_27 = None + x_209 = torch._C._nn.linear( + x_208, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_208 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_210 = torch.nn.functional.dropout(x_209, 0.0, False, False) + x_209 = None + x_211 = x_206 + x_210 + x_206 = x_210 = None + layer_norm_47 = torch.nn.functional.layer_norm( + x_211, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_212 = torch._C._nn.linear( + layer_norm_47, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_47 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_213 = torch._C._nn.gelu(x_212, approximate="none") + x_212 = None + x_214 = torch.nn.functional.dropout(x_213, 0.0, False, False) + x_213 = None + x_215 = torch._C._nn.linear( + x_214, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_214 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_216 = torch.nn.functional.dropout(x_215, 0.0, False, False) + x_215 = None + x_217 = x_211 + x_216 + x_211 = x_216 = None + layer_norm_48 = torch.nn.functional.layer_norm( + x_217, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_96 = torch._C._nn.linear( + layer_norm_48, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_48 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_52 = linear_96.reshape(1, 577, 3, 7, 64) + linear_96 = None + qkv_18 = reshape_52.permute(2, 0, 3, 1, 4) + reshape_52 = None + unbind_18 = qkv_18.unbind(0) + qkv_18 = None + q_22 = unbind_18[0] + k_22 = unbind_18[1] + v_22 = unbind_18[2] + unbind_18 = None + x_218 = torch._C._nn.scaled_dot_product_attention( + q_22, k_22, v_22, attn_mask=None, dropout_p=0.0 + ) + q_22 = k_22 = v_22 = None + transpose_28 = x_218.transpose(1, 2) + x_218 = None + x_219 = transpose_28.reshape(1, 577, 448) + transpose_28 = None + x_220 = torch._C._nn.linear( + x_219, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_219 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_221 = torch.nn.functional.dropout(x_220, 0.0, False, False) + x_220 = None + x_222 = x_217 + x_221 + x_217 = x_221 = None + layer_norm_49 = torch.nn.functional.layer_norm( + x_222, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_223 = torch._C._nn.linear( + layer_norm_49, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_49 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_224 = torch._C._nn.gelu(x_223, approximate="none") + x_223 = None + x_225 = torch.nn.functional.dropout(x_224, 0.0, False, False) + x_224 = None + x_226 = torch._C._nn.linear( + x_225, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_225 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_227 = torch.nn.functional.dropout(x_226, 0.0, False, False) + x_226 = None + x_228 = x_222 + x_227 + x_222 = x_227 = None + layer_norm_50 = torch.nn.functional.layer_norm( + x_228, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_ = (None) + linear_100 = torch._C._nn.linear( + layer_norm_50, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_50 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_54 = linear_100.reshape(1, 577, 3, 7, 64) + linear_100 = None + qkv_19 = reshape_54.permute(2, 0, 3, 1, 4) + reshape_54 = None + unbind_19 = qkv_19.unbind(0) + qkv_19 = None + q_23 = unbind_19[0] + k_23 = unbind_19[1] + v_23 = unbind_19[2] + unbind_19 = None + x_229 = torch._C._nn.scaled_dot_product_attention( + q_23, k_23, v_23, attn_mask=None, dropout_p=0.0 + ) + q_23 = k_23 = v_23 = None + transpose_29 = x_229.transpose(1, 2) + x_229 = None + x_230 = transpose_29.reshape(1, 577, 448) + transpose_29 = None + x_231 = torch._C._nn.linear( + x_230, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_230 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_ = (None) + x_232 = torch.nn.functional.dropout(x_231, 0.0, False, False) + x_231 = None + x_233 = x_228 + x_232 + x_228 = x_232 = None + layer_norm_51 = torch.nn.functional.layer_norm( + x_233, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_ = (None) + x_234 = torch._C._nn.linear( + layer_norm_51, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_51 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_235 = torch._C._nn.gelu(x_234, approximate="none") + x_234 = None + x_236 = torch.nn.functional.dropout(x_235, 0.0, False, False) + x_235 = None + x_237 = torch._C._nn.linear( + x_236, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_236 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_238 = torch.nn.functional.dropout(x_237, 0.0, False, False) + x_237 = None + x_239 = x_233 + x_238 + x_233 = x_238 = None + layer_norm_52 = torch.nn.functional.layer_norm( + x_239, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_ = (None) + linear_104 = torch._C._nn.linear( + layer_norm_52, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_52 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_56 = linear_104.reshape(1, 577, 3, 7, 64) + linear_104 = None + qkv_20 = reshape_56.permute(2, 0, 3, 1, 4) + reshape_56 = None + unbind_20 = qkv_20.unbind(0) + qkv_20 = None + q_24 = unbind_20[0] + k_24 = unbind_20[1] + v_24 = unbind_20[2] + unbind_20 = None + x_240 = torch._C._nn.scaled_dot_product_attention( + q_24, k_24, v_24, attn_mask=None, dropout_p=0.0 + ) + q_24 = k_24 = v_24 = None + transpose_30 = x_240.transpose(1, 2) + x_240 = None + x_241 = transpose_30.reshape(1, 577, 448) + transpose_30 = None + x_242 = torch._C._nn.linear( + x_241, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_241 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_ = (None) + x_243 = torch.nn.functional.dropout(x_242, 0.0, False, False) + x_242 = None + x_244 = x_239 + x_243 + x_239 = x_243 = None + layer_norm_53 = torch.nn.functional.layer_norm( + x_244, + (448,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_ = (None) + x_245 = torch._C._nn.linear( + layer_norm_53, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_53 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_246 = torch._C._nn.gelu(x_245, approximate="none") + x_245 = None + x_247 = torch.nn.functional.dropout(x_246, 0.0, False, False) + x_246 = None + x_248 = torch._C._nn.linear( + x_247, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_247 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_249 = torch.nn.functional.dropout(x_248, 0.0, False, False) + x_248 = None + x_250 = x_244 + x_249 + x_244 = x_249 = None + getitem_103 = x_184[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_35 = torch.nn.functional.layer_norm( + getitem_103, + (224,), + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_103 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_36 = torch._C._nn.gelu(input_35, approximate="none") + input_35 = None + input_37 = torch._C._nn.linear( + input_36, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_36 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_104 = x_250[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_38 = torch.nn.functional.layer_norm( + getitem_104, + (448,), + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_104 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_39 = torch._C._nn.gelu(input_38, approximate="none") + input_38 = None + input_40 = torch._C._nn.linear( + input_39, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_39 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_105 = x_250[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_8 = torch.cat((input_37, getitem_105), dim=1) + input_37 = getitem_105 = None + getitem_106 = tmp_8[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_56 = torch.nn.functional.layer_norm( + tmp_8, + (448,), + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_8 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_107 = layer_norm_56[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_110 = torch._C._nn.linear( + getitem_107, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_107 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_58 = linear_110.reshape(1, 1, 7, 64) + linear_110 = None + q_25 = reshape_58.permute(0, 2, 1, 3) + reshape_58 = None + linear_111 = torch._C._nn.linear( + layer_norm_56, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_59 = linear_111.reshape(1, 577, 7, 64) + linear_111 = None + k_25 = reshape_59.permute(0, 2, 1, 3) + reshape_59 = None + linear_112 = torch._C._nn.linear( + layer_norm_56, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_56 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_60 = linear_112.reshape(1, 577, 7, 64) + linear_112 = None + v_25 = reshape_60.permute(0, 2, 1, 3) + reshape_60 = None + transpose_31 = k_25.transpose(-2, -1) + k_25 = None + matmul_8 = q_25 @ transpose_31 + q_25 = transpose_31 = None + attn_12 = matmul_8 * 0.125 + matmul_8 = None + attn_13 = attn_12.softmax(dim=-1) + attn_12 = None + attn_14 = torch.nn.functional.dropout(attn_13, 0.0, False, False) + attn_13 = None + matmul_9 = attn_14 @ v_25 + attn_14 = v_25 = None + transpose_32 = matmul_9.transpose(1, 2) + matmul_9 = None + x_251 = transpose_32.reshape(1, 1, 448) + transpose_32 = None + x_252 = torch._C._nn.linear( + x_251, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_251 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_253 = torch.nn.functional.dropout(x_252, 0.0, False, False) + x_252 = None + x_254 = getitem_106 + x_253 + getitem_106 = x_253 = None + getitem_108 = x_254[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_254 = None + input_41 = torch.nn.functional.layer_norm( + getitem_108, + (448,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_108 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_42 = torch._C._nn.gelu(input_41, approximate="none") + input_41 = None + input_43 = torch._C._nn.linear( + input_42, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_42 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_109 = x_184[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_9 = torch.cat((input_43, getitem_109), dim=1) + input_43 = getitem_109 = None + getitem_110 = x_184[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_184 = None + tmp_10 = torch.cat((input_40, getitem_110), dim=1) + input_40 = getitem_110 = None + getitem_111 = tmp_10[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_58 = torch.nn.functional.layer_norm( + tmp_10, + (224,), + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_10 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_112 = layer_norm_58[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_115 = torch._C._nn.linear( + getitem_112, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_112 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_62 = linear_115.reshape(1, 1, 7, 32) + linear_115 = None + q_26 = reshape_62.permute(0, 2, 1, 3) + reshape_62 = None + linear_116 = torch._C._nn.linear( + layer_norm_58, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_63 = linear_116.reshape(1, 1157, 7, 32) + linear_116 = None + k_26 = reshape_63.permute(0, 2, 1, 3) + reshape_63 = None + linear_117 = torch._C._nn.linear( + layer_norm_58, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_58 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_64 = linear_117.reshape(1, 1157, 7, 32) + linear_117 = None + v_26 = reshape_64.permute(0, 2, 1, 3) + reshape_64 = None + transpose_33 = k_26.transpose(-2, -1) + k_26 = None + matmul_10 = q_26 @ transpose_33 + q_26 = transpose_33 = None + attn_15 = matmul_10 * 0.1767766952966369 + matmul_10 = None + attn_16 = attn_15.softmax(dim=-1) + attn_15 = None + attn_17 = torch.nn.functional.dropout(attn_16, 0.0, False, False) + attn_16 = None + matmul_11 = attn_17 @ v_26 + attn_17 = v_26 = None + transpose_34 = matmul_11.transpose(1, 2) + matmul_11 = None + x_255 = transpose_34.reshape(1, 1, 224) + transpose_34 = None + x_256 = torch._C._nn.linear( + x_255, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_255 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_257 = torch.nn.functional.dropout(x_256, 0.0, False, False) + x_256 = None + x_258 = getitem_111 + x_257 + getitem_111 = x_257 = None + getitem_113 = x_258[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_258 = None + input_44 = torch.nn.functional.layer_norm( + getitem_113, + (224,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_113 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_45 = torch._C._nn.gelu(input_44, approximate="none") + input_44 = None + input_46 = torch._C._nn.linear( + input_45, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_45 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_114 = x_250[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_250 = None + tmp_11 = torch.cat((input_46, getitem_114), dim=1) + input_46 = getitem_114 = None + x_259 = torch.nn.functional.layer_norm( + tmp_9, + (224,), + l_self_modules_norm_modules_0_parameters_weight_, + l_self_modules_norm_modules_0_parameters_bias_, + 1e-06, + ) + tmp_9 = ( + l_self_modules_norm_modules_0_parameters_weight_ + ) = l_self_modules_norm_modules_0_parameters_bias_ = None + x_260 = torch.nn.functional.layer_norm( + tmp_11, + (448,), + l_self_modules_norm_modules_1_parameters_weight_, + l_self_modules_norm_modules_1_parameters_bias_, + 1e-06, + ) + tmp_11 = ( + l_self_modules_norm_modules_1_parameters_weight_ + ) = l_self_modules_norm_modules_1_parameters_bias_ = None + x_261 = x_259[(slice(None, None, None), 0)] + x_259 = None + x_262 = x_260[(slice(None, None, None), 0)] + x_260 = None + dropout_77 = torch.nn.functional.dropout(x_261, 0.0, False, False) + x_261 = None + dropout_78 = torch.nn.functional.dropout(x_262, 0.0, False, False) + x_262 = None + linear_120 = torch._C._nn.linear( + dropout_77, + l_self_modules_head_modules_0_parameters_weight_, + l_self_modules_head_modules_0_parameters_bias_, + ) + dropout_77 = ( + l_self_modules_head_modules_0_parameters_weight_ + ) = l_self_modules_head_modules_0_parameters_bias_ = None + linear_121 = torch._C._nn.linear( + dropout_78, + l_self_modules_head_modules_1_parameters_weight_, + l_self_modules_head_modules_1_parameters_bias_, + ) + dropout_78 = ( + l_self_modules_head_modules_1_parameters_weight_ + ) = l_self_modules_head_modules_1_parameters_bias_ = None + stack = torch.stack([linear_120, linear_121], dim=0) + linear_120 = linear_121 = None + x_263 = torch.mean(stack, dim=0) + stack = None + return (x_263,) diff --git a/samples/timm/crossvit_18_dagger_408.in1k/weight_meta.py b/samples/timm/crossvit_18_dagger_408.in1k/weight_meta.py new file mode 100644 index 0000000000..1551a691e4 --- /dev/null +++ b/samples/timm/crossvit_18_dagger_408.in1k/weight_meta.py @@ -0,0 +1,3882 @@ +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.224 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_" + ) + shape = [56, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_" + ) + shape = [56] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.045 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_" + ) + shape = [112, 56, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_" + ) + shape = [112] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.025 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_" + ) + shape = [224, 112, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_" + ) + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_0_: + name = "L_self_parameters_cls_token_0_" + shape = [1, 1, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_0_: + name = "L_self_parameters_pos_embed_0_" + shape = [1, 1157, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_" + ) + shape = [112, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_" + ) + shape = [112] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_" + ) + shape = [224, 112, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_" + ) + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_" + ) + shape = [448, 224, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.013 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_" + ) + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.013 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_1_: + name = "L_self_parameters_cls_token_1_" + shape = [1, 1, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_1_: + name = "L_self_parameters_pos_embed_1_" + shape = [1, 577, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [224, 672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [224, 672] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [672, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [224, 672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_norm2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1344, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [448, 1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [448, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [224, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [448, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_weight_: + name = "L_self_modules_norm_modules_0_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_bias_: + name = "L_self_modules_norm_modules_0_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_weight_: + name = "L_self_modules_norm_modules_1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_bias_: + name = "L_self_modules_norm_modules_1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_weight_: + name = "L_self_modules_head_modules_0_parameters_weight_" + shape = [1000, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_bias_: + name = "L_self_modules_head_modules_0_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_weight_: + name = "L_self_modules_head_modules_1_parameters_weight_" + shape = [1000, 448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_bias_: + name = "L_self_modules_head_modules_1_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/crossvit_9_240.in1k/graph_hash.txt b/samples/timm/crossvit_9_240.in1k/graph_hash.txt new file mode 100644 index 0000000000..132be6b212 --- /dev/null +++ b/samples/timm/crossvit_9_240.in1k/graph_hash.txt @@ -0,0 +1 @@ +7f4aa4bf0e6727cd9b1cbdaecbb0c382d588738468f7dd687ec41bcc79f65ffa \ No newline at end of file diff --git a/samples/timm/crossvit_9_240.in1k/graph_net.json b/samples/timm/crossvit_9_240.in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/crossvit_9_240.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/crossvit_9_240.in1k/input_meta.py b/samples/timm/crossvit_9_240.in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/crossvit_9_240.in1k/input_tensor_constraints.py b/samples/timm/crossvit_9_240.in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/crossvit_9_240.in1k/model.py b/samples/timm/crossvit_9_240.in1k/model.py new file mode 100644 index 0000000000..f84c380c77 --- /dev/null +++ b/samples/timm/crossvit_9_240.in1k/model.py @@ -0,0 +1,2169 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_0_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_0_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_1_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ + ) + l_self_parameters_cls_token_0_ = L_self_parameters_cls_token_0_ + l_self_parameters_pos_embed_0_ = L_self_parameters_pos_embed_0_ + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ + ) + l_self_parameters_cls_token_1_ = L_self_parameters_cls_token_1_ + l_self_parameters_pos_embed_1_ = L_self_parameters_pos_embed_1_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_norm_modules_0_parameters_weight_ = ( + L_self_modules_norm_modules_0_parameters_weight_ + ) + l_self_modules_norm_modules_0_parameters_bias_ = ( + L_self_modules_norm_modules_0_parameters_bias_ + ) + l_self_modules_norm_modules_1_parameters_weight_ = ( + L_self_modules_norm_modules_1_parameters_weight_ + ) + l_self_modules_norm_modules_1_parameters_bias_ = ( + L_self_modules_norm_modules_1_parameters_bias_ + ) + l_self_modules_head_modules_0_parameters_weight_ = ( + L_self_modules_head_modules_0_parameters_weight_ + ) + l_self_modules_head_modules_0_parameters_bias_ = ( + L_self_modules_head_modules_0_parameters_bias_ + ) + l_self_modules_head_modules_1_parameters_weight_ = ( + L_self_modules_head_modules_1_parameters_weight_ + ) + l_self_modules_head_modules_1_parameters_bias_ = ( + L_self_modules_head_modules_1_parameters_bias_ + ) + x = torch.nn.functional.interpolate( + l_x_, size=(240, 240), mode="bicubic", align_corners=False + ) + conv2d = torch.conv2d( + x, + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_, + (12, 12), + (0, 0), + (1, 1), + 1, + ) + x = ( + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ = None + flatten = conv2d.flatten(2) + conv2d = None + x_1 = flatten.transpose(1, 2) + flatten = None + cls_tokens = l_self_parameters_cls_token_0_.expand(1, -1, -1) + l_self_parameters_cls_token_0_ = None + x_ = torch.cat((cls_tokens, x_1), dim=1) + cls_tokens = x_1 = None + x__1 = x_ + l_self_parameters_pos_embed_0_ + x_ = l_self_parameters_pos_embed_0_ = None + x__2 = torch.nn.functional.dropout(x__1, 0.0, False, False) + x__1 = None + conv2d_1 = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ = None + flatten_1 = conv2d_1.flatten(2) + conv2d_1 = None + x_2 = flatten_1.transpose(1, 2) + flatten_1 = None + cls_tokens_1 = l_self_parameters_cls_token_1_.expand(1, -1, -1) + l_self_parameters_cls_token_1_ = None + x__3 = torch.cat((cls_tokens_1, x_2), dim=1) + cls_tokens_1 = x_2 = None + x__4 = x__3 + l_self_parameters_pos_embed_1_ + x__3 = l_self_parameters_pos_embed_1_ = None + x__5 = torch.nn.functional.dropout(x__4, 0.0, False, False) + x__4 = None + layer_norm = torch.nn.functional.layer_norm( + x__2, + (128,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear = torch._C._nn.linear( + layer_norm, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 401, 3, 4, 32) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_3 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_2 = x_3.transpose(1, 2) + x_3 = None + x_4 = transpose_2.reshape(1, 401, 128) + transpose_2 = None + x_5 = torch._C._nn.linear( + x_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_6 = torch.nn.functional.dropout(x_5, 0.0, False, False) + x_5 = None + x_7 = x__2 + x_6 + x__2 = x_6 = None + layer_norm_1 = torch.nn.functional.layer_norm( + x_7, + (128,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_8 = torch._C._nn.linear( + layer_norm_1, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_1 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_9 = torch._C._nn.gelu(x_8, approximate="none") + x_8 = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = torch._C._nn.linear( + x_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_7 + x_12 + x_7 = x_12 = None + layer_norm_2 = torch.nn.functional.layer_norm( + x__5, + (256,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_2, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_2 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_2 = linear_4.reshape(1, 197, 3, 4, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_14 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_3 = x_14.transpose(1, 2) + x_14 = None + x_15 = transpose_3.reshape(1, 197, 256) + transpose_3 = None + x_16 = torch._C._nn.linear( + x_15, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_15 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = x__5 + x_17 + x__5 = x_17 = None + layer_norm_3 = torch.nn.functional.layer_norm( + x_18, + (256,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_19 = torch._C._nn.linear( + layer_norm_3, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_3 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_20 = torch._C._nn.gelu(x_19, approximate="none") + x_19 = None + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = torch._C._nn.linear( + x_21, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_21 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_18 + x_23 + x_18 = x_23 = None + layer_norm_4 = torch.nn.functional.layer_norm( + x_24, + (256,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_8.reshape(1, 197, 3, 4, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_25 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_4 = x_25.transpose(1, 2) + x_25 = None + x_26 = transpose_4.reshape(1, 197, 256) + transpose_4 = None + x_27 = torch._C._nn.linear( + x_26, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_26 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_28 = torch.nn.functional.dropout(x_27, 0.0, False, False) + x_27 = None + x_29 = x_24 + x_28 + x_24 = x_28 = None + layer_norm_5 = torch.nn.functional.layer_norm( + x_29, + (256,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_30 = torch._C._nn.linear( + layer_norm_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_31 = torch._C._nn.gelu(x_30, approximate="none") + x_30 = None + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = torch._C._nn.linear( + x_32, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_32 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = x_29 + x_34 + x_29 = x_34 = None + layer_norm_6 = torch.nn.functional.layer_norm( + x_35, + (256,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_6, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_6 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_6 = linear_12.reshape(1, 197, 3, 4, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_36 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_5 = x_36.transpose(1, 2) + x_36 = None + x_37 = transpose_5.reshape(1, 197, 256) + transpose_5 = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_37 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_39 = torch.nn.functional.dropout(x_38, 0.0, False, False) + x_38 = None + x_40 = x_35 + x_39 + x_35 = x_39 = None + layer_norm_7 = torch.nn.functional.layer_norm( + x_40, + (256,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_41 = torch._C._nn.linear( + layer_norm_7, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_7 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_42 = torch._C._nn.gelu(x_41, approximate="none") + x_41 = None + x_43 = torch.nn.functional.dropout(x_42, 0.0, False, False) + x_42 = None + x_44 = torch._C._nn.linear( + x_43, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_43 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = x_40 + x_45 + x_40 = x_45 = None + getitem_28 = x_13[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_1 = torch.nn.functional.layer_norm( + getitem_28, + (128,), + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_28 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_2 = torch._C._nn.gelu(input_1, approximate="none") + input_1 = None + input_3 = torch._C._nn.linear( + input_2, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_2 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_29 = x_46[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_4 = torch.nn.functional.layer_norm( + getitem_29, + (256,), + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_29 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_5 = torch._C._nn.gelu(input_4, approximate="none") + input_4 = None + input_6 = torch._C._nn.linear( + input_5, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_5 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_30 = x_46[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp = torch.cat((input_3, getitem_30), dim=1) + input_3 = getitem_30 = None + getitem_31 = tmp[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_10 = torch.nn.functional.layer_norm( + tmp, + (256,), + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_32 = layer_norm_10[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_18 = torch._C._nn.linear( + getitem_32, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_32 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_8 = linear_18.reshape(1, 1, 4, 64) + linear_18 = None + q_4 = reshape_8.permute(0, 2, 1, 3) + reshape_8 = None + linear_19 = torch._C._nn.linear( + layer_norm_10, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_9 = linear_19.reshape(1, 197, 4, 64) + linear_19 = None + k_4 = reshape_9.permute(0, 2, 1, 3) + reshape_9 = None + linear_20 = torch._C._nn.linear( + layer_norm_10, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_10 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_10 = linear_20.reshape(1, 197, 4, 64) + linear_20 = None + v_4 = reshape_10.permute(0, 2, 1, 3) + reshape_10 = None + transpose_6 = k_4.transpose(-2, -1) + k_4 = None + matmul = q_4 @ transpose_6 + q_4 = transpose_6 = None + attn = matmul * 0.125 + matmul = None + attn_1 = attn.softmax(dim=-1) + attn = None + attn_2 = torch.nn.functional.dropout(attn_1, 0.0, False, False) + attn_1 = None + matmul_1 = attn_2 @ v_4 + attn_2 = v_4 = None + transpose_7 = matmul_1.transpose(1, 2) + matmul_1 = None + x_47 = transpose_7.reshape(1, 1, 256) + transpose_7 = None + x_48 = torch._C._nn.linear( + x_47, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_47 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_49 = torch.nn.functional.dropout(x_48, 0.0, False, False) + x_48 = None + x_50 = getitem_31 + x_49 + getitem_31 = x_49 = None + getitem_33 = x_50[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_50 = None + input_7 = torch.nn.functional.layer_norm( + getitem_33, + (256,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_33 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_8 = torch._C._nn.gelu(input_7, approximate="none") + input_7 = None + input_9 = torch._C._nn.linear( + input_8, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_8 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_34 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_1 = torch.cat((input_9, getitem_34), dim=1) + input_9 = getitem_34 = None + getitem_35 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_13 = None + tmp_2 = torch.cat((input_6, getitem_35), dim=1) + input_6 = getitem_35 = None + getitem_36 = tmp_2[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_12 = torch.nn.functional.layer_norm( + tmp_2, + (128,), + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_2 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_37 = layer_norm_12[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_23 = torch._C._nn.linear( + getitem_37, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_37 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_12 = linear_23.reshape(1, 1, 4, 32) + linear_23 = None + q_5 = reshape_12.permute(0, 2, 1, 3) + reshape_12 = None + linear_24 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_13 = linear_24.reshape(1, 401, 4, 32) + linear_24 = None + k_5 = reshape_13.permute(0, 2, 1, 3) + reshape_13 = None + linear_25 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_12 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_14 = linear_25.reshape(1, 401, 4, 32) + linear_25 = None + v_5 = reshape_14.permute(0, 2, 1, 3) + reshape_14 = None + transpose_8 = k_5.transpose(-2, -1) + k_5 = None + matmul_2 = q_5 @ transpose_8 + q_5 = transpose_8 = None + attn_3 = matmul_2 * 0.1767766952966369 + matmul_2 = None + attn_4 = attn_3.softmax(dim=-1) + attn_3 = None + attn_5 = torch.nn.functional.dropout(attn_4, 0.0, False, False) + attn_4 = None + matmul_3 = attn_5 @ v_5 + attn_5 = v_5 = None + transpose_9 = matmul_3.transpose(1, 2) + matmul_3 = None + x_51 = transpose_9.reshape(1, 1, 128) + transpose_9 = None + x_52 = torch._C._nn.linear( + x_51, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_51 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = getitem_36 + x_53 + getitem_36 = x_53 = None + getitem_38 = x_54[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_54 = None + input_10 = torch.nn.functional.layer_norm( + getitem_38, + (128,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_38 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_11 = torch._C._nn.gelu(input_10, approximate="none") + input_10 = None + input_12 = torch._C._nn.linear( + input_11, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_11 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_39 = x_46[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_46 = None + tmp_3 = torch.cat((input_12, getitem_39), dim=1) + input_12 = getitem_39 = None + layer_norm_14 = torch.nn.functional.layer_norm( + tmp_1, + (128,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_28 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_14 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_16 = linear_28.reshape(1, 401, 3, 4, 32) + linear_28 = None + qkv_4 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_6 = unbind_4[0] + k_6 = unbind_4[1] + v_6 = unbind_4[2] + unbind_4 = None + x_55 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_10 = x_55.transpose(1, 2) + x_55 = None + x_56 = transpose_10.reshape(1, 401, 128) + transpose_10 = None + x_57 = torch._C._nn.linear( + x_56, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_56 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_58 = torch.nn.functional.dropout(x_57, 0.0, False, False) + x_57 = None + x_59 = tmp_1 + x_58 + tmp_1 = x_58 = None + layer_norm_15 = torch.nn.functional.layer_norm( + x_59, + (128,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_60 = torch._C._nn.linear( + layer_norm_15, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_15 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_61 = torch._C._nn.gelu(x_60, approximate="none") + x_60 = None + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + x_63 = torch._C._nn.linear( + x_62, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_62 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_64 = torch.nn.functional.dropout(x_63, 0.0, False, False) + x_63 = None + x_65 = x_59 + x_64 + x_59 = x_64 = None + layer_norm_16 = torch.nn.functional.layer_norm( + tmp_3, + (256,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_32 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_16 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_18 = linear_32.reshape(1, 197, 3, 4, 64) + linear_32 = None + qkv_5 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_7 = unbind_5[0] + k_7 = unbind_5[1] + v_7 = unbind_5[2] + unbind_5 = None + x_66 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_11 = x_66.transpose(1, 2) + x_66 = None + x_67 = transpose_11.reshape(1, 197, 256) + transpose_11 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_67 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = tmp_3 + x_69 + tmp_3 = x_69 = None + layer_norm_17 = torch.nn.functional.layer_norm( + x_70, + (256,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_71 = torch._C._nn.linear( + layer_norm_17, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_17 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_72 = torch._C._nn.gelu(x_71, approximate="none") + x_71 = None + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = torch._C._nn.linear( + x_73, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_73 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_75 = torch.nn.functional.dropout(x_74, 0.0, False, False) + x_74 = None + x_76 = x_70 + x_75 + x_70 = x_75 = None + layer_norm_18 = torch.nn.functional.layer_norm( + x_76, + (256,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_18 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_20 = linear_36.reshape(1, 197, 3, 4, 64) + linear_36 = None + qkv_6 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_8 = unbind_6[0] + k_8 = unbind_6[1] + v_8 = unbind_6[2] + unbind_6 = None + x_77 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_12 = x_77.transpose(1, 2) + x_77 = None + x_78 = transpose_12.reshape(1, 197, 256) + transpose_12 = None + x_79 = torch._C._nn.linear( + x_78, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_78 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = x_76 + x_80 + x_76 = x_80 = None + layer_norm_19 = torch.nn.functional.layer_norm( + x_81, + (256,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_82 = torch._C._nn.linear( + layer_norm_19, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_19 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_83 = torch._C._nn.gelu(x_82, approximate="none") + x_82 = None + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = torch._C._nn.linear( + x_84, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_84 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_86 = torch.nn.functional.dropout(x_85, 0.0, False, False) + x_85 = None + x_87 = x_81 + x_86 + x_81 = x_86 = None + layer_norm_20 = torch.nn.functional.layer_norm( + x_87, + (256,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_20 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_22 = linear_40.reshape(1, 197, 3, 4, 64) + linear_40 = None + qkv_7 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_9 = unbind_7[0] + k_9 = unbind_7[1] + v_9 = unbind_7[2] + unbind_7 = None + x_88 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_13 = x_88.transpose(1, 2) + x_88 = None + x_89 = transpose_13.reshape(1, 197, 256) + transpose_13 = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_89 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = x_87 + x_91 + x_87 = x_91 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_92, + (256,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_93 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_21 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_94 = torch._C._nn.gelu(x_93, approximate="none") + x_93 = None + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = torch._C._nn.linear( + x_95, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_95 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_97 = torch.nn.functional.dropout(x_96, 0.0, False, False) + x_96 = None + x_98 = x_92 + x_97 + x_92 = x_97 = None + getitem_52 = x_65[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_13 = torch.nn.functional.layer_norm( + getitem_52, + (128,), + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_52 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_14 = torch._C._nn.gelu(input_13, approximate="none") + input_13 = None + input_15 = torch._C._nn.linear( + input_14, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_14 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_53 = x_98[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_16 = torch.nn.functional.layer_norm( + getitem_53, + (256,), + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_53 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_17 = torch._C._nn.gelu(input_16, approximate="none") + input_16 = None + input_18 = torch._C._nn.linear( + input_17, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_17 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_54 = x_98[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_4 = torch.cat((input_15, getitem_54), dim=1) + input_15 = getitem_54 = None + getitem_55 = tmp_4[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_24 = torch.nn.functional.layer_norm( + tmp_4, + (256,), + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_4 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_56 = layer_norm_24[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_46 = torch._C._nn.linear( + getitem_56, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_56 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_24 = linear_46.reshape(1, 1, 4, 64) + linear_46 = None + q_10 = reshape_24.permute(0, 2, 1, 3) + reshape_24 = None + linear_47 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_25 = linear_47.reshape(1, 197, 4, 64) + linear_47 = None + k_10 = reshape_25.permute(0, 2, 1, 3) + reshape_25 = None + linear_48 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_24 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_26 = linear_48.reshape(1, 197, 4, 64) + linear_48 = None + v_10 = reshape_26.permute(0, 2, 1, 3) + reshape_26 = None + transpose_14 = k_10.transpose(-2, -1) + k_10 = None + matmul_4 = q_10 @ transpose_14 + q_10 = transpose_14 = None + attn_6 = matmul_4 * 0.125 + matmul_4 = None + attn_7 = attn_6.softmax(dim=-1) + attn_6 = None + attn_8 = torch.nn.functional.dropout(attn_7, 0.0, False, False) + attn_7 = None + matmul_5 = attn_8 @ v_10 + attn_8 = v_10 = None + transpose_15 = matmul_5.transpose(1, 2) + matmul_5 = None + x_99 = transpose_15.reshape(1, 1, 256) + transpose_15 = None + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_99 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_101 = torch.nn.functional.dropout(x_100, 0.0, False, False) + x_100 = None + x_102 = getitem_55 + x_101 + getitem_55 = x_101 = None + getitem_57 = x_102[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_102 = None + input_19 = torch.nn.functional.layer_norm( + getitem_57, + (256,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_57 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_20 = torch._C._nn.gelu(input_19, approximate="none") + input_19 = None + input_21 = torch._C._nn.linear( + input_20, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_20 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_58 = x_65[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_5 = torch.cat((input_21, getitem_58), dim=1) + input_21 = getitem_58 = None + getitem_59 = x_65[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_65 = None + tmp_6 = torch.cat((input_18, getitem_59), dim=1) + input_18 = getitem_59 = None + getitem_60 = tmp_6[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_26 = torch.nn.functional.layer_norm( + tmp_6, + (128,), + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_6 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_61 = layer_norm_26[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_51 = torch._C._nn.linear( + getitem_61, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_61 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_28 = linear_51.reshape(1, 1, 4, 32) + linear_51 = None + q_11 = reshape_28.permute(0, 2, 1, 3) + reshape_28 = None + linear_52 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_29 = linear_52.reshape(1, 401, 4, 32) + linear_52 = None + k_11 = reshape_29.permute(0, 2, 1, 3) + reshape_29 = None + linear_53 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_26 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_30 = linear_53.reshape(1, 401, 4, 32) + linear_53 = None + v_11 = reshape_30.permute(0, 2, 1, 3) + reshape_30 = None + transpose_16 = k_11.transpose(-2, -1) + k_11 = None + matmul_6 = q_11 @ transpose_16 + q_11 = transpose_16 = None + attn_9 = matmul_6 * 0.1767766952966369 + matmul_6 = None + attn_10 = attn_9.softmax(dim=-1) + attn_9 = None + attn_11 = torch.nn.functional.dropout(attn_10, 0.0, False, False) + attn_10 = None + matmul_7 = attn_11 @ v_11 + attn_11 = v_11 = None + transpose_17 = matmul_7.transpose(1, 2) + matmul_7 = None + x_103 = transpose_17.reshape(1, 1, 128) + transpose_17 = None + x_104 = torch._C._nn.linear( + x_103, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_103 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = getitem_60 + x_105 + getitem_60 = x_105 = None + getitem_62 = x_106[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_106 = None + input_22 = torch.nn.functional.layer_norm( + getitem_62, + (128,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_62 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_23 = torch._C._nn.gelu(input_22, approximate="none") + input_22 = None + input_24 = torch._C._nn.linear( + input_23, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_23 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_63 = x_98[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_98 = None + tmp_7 = torch.cat((input_24, getitem_63), dim=1) + input_24 = getitem_63 = None + layer_norm_28 = torch.nn.functional.layer_norm( + tmp_5, + (128,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_28 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_32 = linear_56.reshape(1, 401, 3, 4, 32) + linear_56 = None + qkv_8 = reshape_32.permute(2, 0, 3, 1, 4) + reshape_32 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_12 = unbind_8[0] + k_12 = unbind_8[1] + v_12 = unbind_8[2] + unbind_8 = None + x_107 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_18 = x_107.transpose(1, 2) + x_107 = None + x_108 = transpose_18.reshape(1, 401, 128) + transpose_18 = None + x_109 = torch._C._nn.linear( + x_108, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_108 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_110 = torch.nn.functional.dropout(x_109, 0.0, False, False) + x_109 = None + x_111 = tmp_5 + x_110 + tmp_5 = x_110 = None + layer_norm_29 = torch.nn.functional.layer_norm( + x_111, + (128,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_112 = torch._C._nn.linear( + layer_norm_29, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_29 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112, approximate="none") + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_115 = torch._C._nn.linear( + x_114, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_114 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + x_117 = x_111 + x_116 + x_111 = x_116 = None + layer_norm_30 = torch.nn.functional.layer_norm( + tmp_7, + (256,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_60 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_30 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_34 = linear_60.reshape(1, 197, 3, 4, 64) + linear_60 = None + qkv_9 = reshape_34.permute(2, 0, 3, 1, 4) + reshape_34 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_13 = unbind_9[0] + k_13 = unbind_9[1] + v_13 = unbind_9[2] + unbind_9 = None + x_118 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_19 = x_118.transpose(1, 2) + x_118 = None + x_119 = transpose_19.reshape(1, 197, 256) + transpose_19 = None + x_120 = torch._C._nn.linear( + x_119, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_119 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_121 = torch.nn.functional.dropout(x_120, 0.0, False, False) + x_120 = None + x_122 = tmp_7 + x_121 + tmp_7 = x_121 = None + layer_norm_31 = torch.nn.functional.layer_norm( + x_122, + (256,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_123 = torch._C._nn.linear( + layer_norm_31, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_31 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_124 = torch._C._nn.gelu(x_123, approximate="none") + x_123 = None + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_125 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = x_122 + x_127 + x_122 = x_127 = None + layer_norm_32 = torch.nn.functional.layer_norm( + x_128, + (256,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_32 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_36 = linear_64.reshape(1, 197, 3, 4, 64) + linear_64 = None + qkv_10 = reshape_36.permute(2, 0, 3, 1, 4) + reshape_36 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_14 = unbind_10[0] + k_14 = unbind_10[1] + v_14 = unbind_10[2] + unbind_10 = None + x_129 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_20 = x_129.transpose(1, 2) + x_129 = None + x_130 = transpose_20.reshape(1, 197, 256) + transpose_20 = None + x_131 = torch._C._nn.linear( + x_130, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_130 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_132 = torch.nn.functional.dropout(x_131, 0.0, False, False) + x_131 = None + x_133 = x_128 + x_132 + x_128 = x_132 = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_133, + (256,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_134 = torch._C._nn.linear( + layer_norm_33, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_33 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_135 = torch._C._nn.gelu(x_134, approximate="none") + x_134 = None + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_133 + x_138 + x_133 = x_138 = None + layer_norm_34 = torch.nn.functional.layer_norm( + x_139, + (256,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_68 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_34 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_38 = linear_68.reshape(1, 197, 3, 4, 64) + linear_68 = None + qkv_11 = reshape_38.permute(2, 0, 3, 1, 4) + reshape_38 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_15 = unbind_11[0] + k_15 = unbind_11[1] + v_15 = unbind_11[2] + unbind_11 = None + x_140 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_21 = x_140.transpose(1, 2) + x_140 = None + x_141 = transpose_21.reshape(1, 197, 256) + transpose_21 = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_141 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_143 = torch.nn.functional.dropout(x_142, 0.0, False, False) + x_142 = None + x_144 = x_139 + x_143 + x_139 = x_143 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_144, + (256,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_145 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_35 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_146 = torch._C._nn.gelu(x_145, approximate="none") + x_145 = None + x_147 = torch.nn.functional.dropout(x_146, 0.0, False, False) + x_146 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_144 + x_149 + x_144 = x_149 = None + getitem_76 = x_117[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_25 = torch.nn.functional.layer_norm( + getitem_76, + (128,), + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_76 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_26 = torch._C._nn.gelu(input_25, approximate="none") + input_25 = None + input_27 = torch._C._nn.linear( + input_26, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_26 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_77 = x_150[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_28 = torch.nn.functional.layer_norm( + getitem_77, + (256,), + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_77 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_29 = torch._C._nn.gelu(input_28, approximate="none") + input_28 = None + input_30 = torch._C._nn.linear( + input_29, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_29 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_78 = x_150[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_8 = torch.cat((input_27, getitem_78), dim=1) + input_27 = getitem_78 = None + getitem_79 = tmp_8[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_38 = torch.nn.functional.layer_norm( + tmp_8, + (256,), + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_8 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_80 = layer_norm_38[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_74 = torch._C._nn.linear( + getitem_80, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_80 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_40 = linear_74.reshape(1, 1, 4, 64) + linear_74 = None + q_16 = reshape_40.permute(0, 2, 1, 3) + reshape_40 = None + linear_75 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_41 = linear_75.reshape(1, 197, 4, 64) + linear_75 = None + k_16 = reshape_41.permute(0, 2, 1, 3) + reshape_41 = None + linear_76 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_38 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_42 = linear_76.reshape(1, 197, 4, 64) + linear_76 = None + v_16 = reshape_42.permute(0, 2, 1, 3) + reshape_42 = None + transpose_22 = k_16.transpose(-2, -1) + k_16 = None + matmul_8 = q_16 @ transpose_22 + q_16 = transpose_22 = None + attn_12 = matmul_8 * 0.125 + matmul_8 = None + attn_13 = attn_12.softmax(dim=-1) + attn_12 = None + attn_14 = torch.nn.functional.dropout(attn_13, 0.0, False, False) + attn_13 = None + matmul_9 = attn_14 @ v_16 + attn_14 = v_16 = None + transpose_23 = matmul_9.transpose(1, 2) + matmul_9 = None + x_151 = transpose_23.reshape(1, 1, 256) + transpose_23 = None + x_152 = torch._C._nn.linear( + x_151, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_151 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_153 = torch.nn.functional.dropout(x_152, 0.0, False, False) + x_152 = None + x_154 = getitem_79 + x_153 + getitem_79 = x_153 = None + getitem_81 = x_154[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_154 = None + input_31 = torch.nn.functional.layer_norm( + getitem_81, + (256,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_81 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_32 = torch._C._nn.gelu(input_31, approximate="none") + input_31 = None + input_33 = torch._C._nn.linear( + input_32, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_32 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_82 = x_117[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_9 = torch.cat((input_33, getitem_82), dim=1) + input_33 = getitem_82 = None + getitem_83 = x_117[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_117 = None + tmp_10 = torch.cat((input_30, getitem_83), dim=1) + input_30 = getitem_83 = None + getitem_84 = tmp_10[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_40 = torch.nn.functional.layer_norm( + tmp_10, + (128,), + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_10 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_85 = layer_norm_40[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_79 = torch._C._nn.linear( + getitem_85, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_85 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_44 = linear_79.reshape(1, 1, 4, 32) + linear_79 = None + q_17 = reshape_44.permute(0, 2, 1, 3) + reshape_44 = None + linear_80 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_45 = linear_80.reshape(1, 401, 4, 32) + linear_80 = None + k_17 = reshape_45.permute(0, 2, 1, 3) + reshape_45 = None + linear_81 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_40 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_46 = linear_81.reshape(1, 401, 4, 32) + linear_81 = None + v_17 = reshape_46.permute(0, 2, 1, 3) + reshape_46 = None + transpose_24 = k_17.transpose(-2, -1) + k_17 = None + matmul_10 = q_17 @ transpose_24 + q_17 = transpose_24 = None + attn_15 = matmul_10 * 0.1767766952966369 + matmul_10 = None + attn_16 = attn_15.softmax(dim=-1) + attn_15 = None + attn_17 = torch.nn.functional.dropout(attn_16, 0.0, False, False) + attn_16 = None + matmul_11 = attn_17 @ v_17 + attn_17 = v_17 = None + transpose_25 = matmul_11.transpose(1, 2) + matmul_11 = None + x_155 = transpose_25.reshape(1, 1, 128) + transpose_25 = None + x_156 = torch._C._nn.linear( + x_155, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_155 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = getitem_84 + x_157 + getitem_84 = x_157 = None + getitem_86 = x_158[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_158 = None + input_34 = torch.nn.functional.layer_norm( + getitem_86, + (128,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_86 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_35 = torch._C._nn.gelu(input_34, approximate="none") + input_34 = None + input_36 = torch._C._nn.linear( + input_35, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_35 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_87 = x_150[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_150 = None + tmp_11 = torch.cat((input_36, getitem_87), dim=1) + input_36 = getitem_87 = None + x_159 = torch.nn.functional.layer_norm( + tmp_9, + (128,), + l_self_modules_norm_modules_0_parameters_weight_, + l_self_modules_norm_modules_0_parameters_bias_, + 1e-06, + ) + tmp_9 = ( + l_self_modules_norm_modules_0_parameters_weight_ + ) = l_self_modules_norm_modules_0_parameters_bias_ = None + x_160 = torch.nn.functional.layer_norm( + tmp_11, + (256,), + l_self_modules_norm_modules_1_parameters_weight_, + l_self_modules_norm_modules_1_parameters_bias_, + 1e-06, + ) + tmp_11 = ( + l_self_modules_norm_modules_1_parameters_weight_ + ) = l_self_modules_norm_modules_1_parameters_bias_ = None + x_161 = x_159[(slice(None, None, None), 0)] + x_159 = None + x_162 = x_160[(slice(None, None, None), 0)] + x_160 = None + dropout_50 = torch.nn.functional.dropout(x_161, 0.0, False, False) + x_161 = None + dropout_51 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + linear_84 = torch._C._nn.linear( + dropout_50, + l_self_modules_head_modules_0_parameters_weight_, + l_self_modules_head_modules_0_parameters_bias_, + ) + dropout_50 = ( + l_self_modules_head_modules_0_parameters_weight_ + ) = l_self_modules_head_modules_0_parameters_bias_ = None + linear_85 = torch._C._nn.linear( + dropout_51, + l_self_modules_head_modules_1_parameters_weight_, + l_self_modules_head_modules_1_parameters_bias_, + ) + dropout_51 = ( + l_self_modules_head_modules_1_parameters_weight_ + ) = l_self_modules_head_modules_1_parameters_bias_ = None + stack = torch.stack([linear_84, linear_85], dim=0) + linear_84 = linear_85 = None + x_163 = torch.mean(stack, dim=0) + stack = None + return (x_163,) diff --git a/samples/timm/crossvit_9_240.in1k/weight_meta.py b/samples/timm/crossvit_9_240.in1k/weight_meta.py new file mode 100644 index 0000000000..86db4db2eb --- /dev/null +++ b/samples/timm/crossvit_9_240.in1k/weight_meta.py @@ -0,0 +1,2688 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.228 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_" + shape = [128, 3, 12, 12] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.028 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.005 + std = 0.027 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_0_: + name = "L_self_parameters_cls_token_0_" + shape = [1, 1, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_0_: + name = "L_self_parameters_pos_embed_0_" + shape = [1, 401, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_" + shape = [256, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_1_: + name = "L_self_parameters_cls_token_1_" + shape = [1, 1, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_1_: + name = "L_self_parameters_pos_embed_1_" + shape = [1, 197, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [256, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [128, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [128, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [256, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [256, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [128, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [128, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [256, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [256, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [128, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [128, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [256, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_weight_: + name = "L_self_modules_norm_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_bias_: + name = "L_self_modules_norm_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_weight_: + name = "L_self_modules_norm_modules_1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_bias_: + name = "L_self_modules_norm_modules_1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_weight_: + name = "L_self_modules_head_modules_0_parameters_weight_" + shape = [1000, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_bias_: + name = "L_self_modules_head_modules_0_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_weight_: + name = "L_self_modules_head_modules_1_parameters_weight_" + shape = [1000, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_bias_: + name = "L_self_modules_head_modules_1_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/crossvit_9_dagger_240.in1k/graph_hash.txt b/samples/timm/crossvit_9_dagger_240.in1k/graph_hash.txt new file mode 100644 index 0000000000..c6b98d5f1b --- /dev/null +++ b/samples/timm/crossvit_9_dagger_240.in1k/graph_hash.txt @@ -0,0 +1 @@ +e01655a005966ed579a9eeb52b1521d441377c3157438eb7f4877ec9c13d775a \ No newline at end of file diff --git a/samples/timm/crossvit_9_dagger_240.in1k/graph_net.json b/samples/timm/crossvit_9_dagger_240.in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/crossvit_9_dagger_240.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/crossvit_9_dagger_240.in1k/input_meta.py b/samples/timm/crossvit_9_dagger_240.in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/crossvit_9_dagger_240.in1k/input_tensor_constraints.py b/samples/timm/crossvit_9_dagger_240.in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/crossvit_9_dagger_240.in1k/model.py b/samples/timm/crossvit_9_dagger_240.in1k/model.py new file mode 100644 index 0000000000..33c73d0aab --- /dev/null +++ b/samples/timm/crossvit_9_dagger_240.in1k/model.py @@ -0,0 +1,2245 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_0_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_0_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_1_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ = L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ + ) + l_self_parameters_cls_token_0_ = L_self_parameters_cls_token_0_ + l_self_parameters_pos_embed_0_ = L_self_parameters_pos_embed_0_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ = L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ + ) + l_self_parameters_cls_token_1_ = L_self_parameters_cls_token_1_ + l_self_parameters_pos_embed_1_ = L_self_parameters_pos_embed_1_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_norm_modules_0_parameters_weight_ = ( + L_self_modules_norm_modules_0_parameters_weight_ + ) + l_self_modules_norm_modules_0_parameters_bias_ = ( + L_self_modules_norm_modules_0_parameters_bias_ + ) + l_self_modules_norm_modules_1_parameters_weight_ = ( + L_self_modules_norm_modules_1_parameters_weight_ + ) + l_self_modules_norm_modules_1_parameters_bias_ = ( + L_self_modules_norm_modules_1_parameters_bias_ + ) + l_self_modules_head_modules_0_parameters_weight_ = ( + L_self_modules_head_modules_0_parameters_weight_ + ) + l_self_modules_head_modules_0_parameters_bias_ = ( + L_self_modules_head_modules_0_parameters_bias_ + ) + l_self_modules_head_modules_1_parameters_weight_ = ( + L_self_modules_head_modules_1_parameters_weight_ + ) + l_self_modules_head_modules_1_parameters_bias_ = ( + L_self_modules_head_modules_1_parameters_bias_ + ) + x = torch.nn.functional.interpolate( + l_x_, size=(240, 240), mode="bicubic", align_corners=False + ) + input_1 = torch.conv2d( + x, + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_, + (4, 4), + (3, 3), + (1, 1), + 1, + ) + x = l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_ + ) = None + input_2 = torch.nn.functional.relu(input_1, inplace=True) + input_1 = None + input_3 = torch.conv2d( + input_2, + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_, + (3, 3), + (0, 0), + (1, 1), + 1, + ) + input_2 = l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_ + ) = None + input_4 = torch.nn.functional.relu(input_3, inplace=True) + input_3 = None + input_5 = torch.conv2d( + input_4, + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + input_4 = l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_ = ( + l_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_ + ) = None + flatten = input_5.flatten(2) + input_5 = None + x_1 = flatten.transpose(1, 2) + flatten = None + cls_tokens = l_self_parameters_cls_token_0_.expand(1, -1, -1) + l_self_parameters_cls_token_0_ = None + x_ = torch.cat((cls_tokens, x_1), dim=1) + cls_tokens = x_1 = None + x__1 = x_ + l_self_parameters_pos_embed_0_ + x_ = l_self_parameters_pos_embed_0_ = None + x__2 = torch.nn.functional.dropout(x__1, 0.0, False, False) + x__1 = None + input_6 = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_, + (4, 4), + (3, 3), + (1, 1), + 1, + ) + l_x_ = l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_ + ) = None + input_7 = torch.nn.functional.relu(input_6, inplace=True) + input_6 = None + input_8 = torch.conv2d( + input_7, + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_7 = l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_ + ) = None + input_9 = torch.nn.functional.relu(input_8, inplace=True) + input_8 = None + input_10 = torch.conv2d( + input_9, + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_9 = l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_ + ) = None + flatten_1 = input_10.flatten(2) + input_10 = None + x_2 = flatten_1.transpose(1, 2) + flatten_1 = None + cls_tokens_1 = l_self_parameters_cls_token_1_.expand(1, -1, -1) + l_self_parameters_cls_token_1_ = None + x__3 = torch.cat((cls_tokens_1, x_2), dim=1) + cls_tokens_1 = x_2 = None + x__4 = x__3 + l_self_parameters_pos_embed_1_ + x__3 = l_self_parameters_pos_embed_1_ = None + x__5 = torch.nn.functional.dropout(x__4, 0.0, False, False) + x__4 = None + layer_norm = torch.nn.functional.layer_norm( + x__2, + (128,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear = torch._C._nn.linear( + layer_norm, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 401, 3, 4, 32) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_3 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_2 = x_3.transpose(1, 2) + x_3 = None + x_4 = transpose_2.reshape(1, 401, 128) + transpose_2 = None + x_5 = torch._C._nn.linear( + x_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_6 = torch.nn.functional.dropout(x_5, 0.0, False, False) + x_5 = None + x_7 = x__2 + x_6 + x__2 = x_6 = None + layer_norm_1 = torch.nn.functional.layer_norm( + x_7, + (128,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_8 = torch._C._nn.linear( + layer_norm_1, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_1 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_9 = torch._C._nn.gelu(x_8, approximate="none") + x_8 = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = torch._C._nn.linear( + x_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_7 + x_12 + x_7 = x_12 = None + layer_norm_2 = torch.nn.functional.layer_norm( + x__5, + (256,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_2, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_2 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_2 = linear_4.reshape(1, 197, 3, 4, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_14 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_3 = x_14.transpose(1, 2) + x_14 = None + x_15 = transpose_3.reshape(1, 197, 256) + transpose_3 = None + x_16 = torch._C._nn.linear( + x_15, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_15 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = x__5 + x_17 + x__5 = x_17 = None + layer_norm_3 = torch.nn.functional.layer_norm( + x_18, + (256,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_19 = torch._C._nn.linear( + layer_norm_3, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_3 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_20 = torch._C._nn.gelu(x_19, approximate="none") + x_19 = None + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = torch._C._nn.linear( + x_21, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_21 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_18 + x_23 + x_18 = x_23 = None + layer_norm_4 = torch.nn.functional.layer_norm( + x_24, + (256,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_8.reshape(1, 197, 3, 4, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_25 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_4 = x_25.transpose(1, 2) + x_25 = None + x_26 = transpose_4.reshape(1, 197, 256) + transpose_4 = None + x_27 = torch._C._nn.linear( + x_26, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_26 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_28 = torch.nn.functional.dropout(x_27, 0.0, False, False) + x_27 = None + x_29 = x_24 + x_28 + x_24 = x_28 = None + layer_norm_5 = torch.nn.functional.layer_norm( + x_29, + (256,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_30 = torch._C._nn.linear( + layer_norm_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_31 = torch._C._nn.gelu(x_30, approximate="none") + x_30 = None + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = torch._C._nn.linear( + x_32, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_32 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = x_29 + x_34 + x_29 = x_34 = None + layer_norm_6 = torch.nn.functional.layer_norm( + x_35, + (256,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_6, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_6 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_6 = linear_12.reshape(1, 197, 3, 4, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_36 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_5 = x_36.transpose(1, 2) + x_36 = None + x_37 = transpose_5.reshape(1, 197, 256) + transpose_5 = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_37 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_39 = torch.nn.functional.dropout(x_38, 0.0, False, False) + x_38 = None + x_40 = x_35 + x_39 + x_35 = x_39 = None + layer_norm_7 = torch.nn.functional.layer_norm( + x_40, + (256,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_41 = torch._C._nn.linear( + layer_norm_7, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_7 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_42 = torch._C._nn.gelu(x_41, approximate="none") + x_41 = None + x_43 = torch.nn.functional.dropout(x_42, 0.0, False, False) + x_42 = None + x_44 = torch._C._nn.linear( + x_43, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_43 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = x_40 + x_45 + x_40 = x_45 = None + getitem_28 = x_13[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_11 = torch.nn.functional.layer_norm( + getitem_28, + (128,), + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_28 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_12 = torch._C._nn.gelu(input_11, approximate="none") + input_11 = None + input_13 = torch._C._nn.linear( + input_12, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_12 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_29 = x_46[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_14 = torch.nn.functional.layer_norm( + getitem_29, + (256,), + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_29 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_15 = torch._C._nn.gelu(input_14, approximate="none") + input_14 = None + input_16 = torch._C._nn.linear( + input_15, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_15 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_30 = x_46[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp = torch.cat((input_13, getitem_30), dim=1) + input_13 = getitem_30 = None + getitem_31 = tmp[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_10 = torch.nn.functional.layer_norm( + tmp, + (256,), + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_32 = layer_norm_10[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_18 = torch._C._nn.linear( + getitem_32, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_32 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_8 = linear_18.reshape(1, 1, 4, 64) + linear_18 = None + q_4 = reshape_8.permute(0, 2, 1, 3) + reshape_8 = None + linear_19 = torch._C._nn.linear( + layer_norm_10, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_9 = linear_19.reshape(1, 197, 4, 64) + linear_19 = None + k_4 = reshape_9.permute(0, 2, 1, 3) + reshape_9 = None + linear_20 = torch._C._nn.linear( + layer_norm_10, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_10 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_10 = linear_20.reshape(1, 197, 4, 64) + linear_20 = None + v_4 = reshape_10.permute(0, 2, 1, 3) + reshape_10 = None + transpose_6 = k_4.transpose(-2, -1) + k_4 = None + matmul = q_4 @ transpose_6 + q_4 = transpose_6 = None + attn = matmul * 0.125 + matmul = None + attn_1 = attn.softmax(dim=-1) + attn = None + attn_2 = torch.nn.functional.dropout(attn_1, 0.0, False, False) + attn_1 = None + matmul_1 = attn_2 @ v_4 + attn_2 = v_4 = None + transpose_7 = matmul_1.transpose(1, 2) + matmul_1 = None + x_47 = transpose_7.reshape(1, 1, 256) + transpose_7 = None + x_48 = torch._C._nn.linear( + x_47, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_47 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_49 = torch.nn.functional.dropout(x_48, 0.0, False, False) + x_48 = None + x_50 = getitem_31 + x_49 + getitem_31 = x_49 = None + getitem_33 = x_50[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_50 = None + input_17 = torch.nn.functional.layer_norm( + getitem_33, + (256,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_33 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_18 = torch._C._nn.gelu(input_17, approximate="none") + input_17 = None + input_19 = torch._C._nn.linear( + input_18, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_18 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_34 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_1 = torch.cat((input_19, getitem_34), dim=1) + input_19 = getitem_34 = None + getitem_35 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_13 = None + tmp_2 = torch.cat((input_16, getitem_35), dim=1) + input_16 = getitem_35 = None + getitem_36 = tmp_2[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_12 = torch.nn.functional.layer_norm( + tmp_2, + (128,), + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_2 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_37 = layer_norm_12[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_23 = torch._C._nn.linear( + getitem_37, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_37 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_12 = linear_23.reshape(1, 1, 4, 32) + linear_23 = None + q_5 = reshape_12.permute(0, 2, 1, 3) + reshape_12 = None + linear_24 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_13 = linear_24.reshape(1, 401, 4, 32) + linear_24 = None + k_5 = reshape_13.permute(0, 2, 1, 3) + reshape_13 = None + linear_25 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_12 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_14 = linear_25.reshape(1, 401, 4, 32) + linear_25 = None + v_5 = reshape_14.permute(0, 2, 1, 3) + reshape_14 = None + transpose_8 = k_5.transpose(-2, -1) + k_5 = None + matmul_2 = q_5 @ transpose_8 + q_5 = transpose_8 = None + attn_3 = matmul_2 * 0.1767766952966369 + matmul_2 = None + attn_4 = attn_3.softmax(dim=-1) + attn_3 = None + attn_5 = torch.nn.functional.dropout(attn_4, 0.0, False, False) + attn_4 = None + matmul_3 = attn_5 @ v_5 + attn_5 = v_5 = None + transpose_9 = matmul_3.transpose(1, 2) + matmul_3 = None + x_51 = transpose_9.reshape(1, 1, 128) + transpose_9 = None + x_52 = torch._C._nn.linear( + x_51, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_51 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = getitem_36 + x_53 + getitem_36 = x_53 = None + getitem_38 = x_54[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_54 = None + input_20 = torch.nn.functional.layer_norm( + getitem_38, + (128,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_38 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_21 = torch._C._nn.gelu(input_20, approximate="none") + input_20 = None + input_22 = torch._C._nn.linear( + input_21, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_21 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_39 = x_46[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_46 = None + tmp_3 = torch.cat((input_22, getitem_39), dim=1) + input_22 = getitem_39 = None + layer_norm_14 = torch.nn.functional.layer_norm( + tmp_1, + (128,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_28 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_14 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_16 = linear_28.reshape(1, 401, 3, 4, 32) + linear_28 = None + qkv_4 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_6 = unbind_4[0] + k_6 = unbind_4[1] + v_6 = unbind_4[2] + unbind_4 = None + x_55 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_10 = x_55.transpose(1, 2) + x_55 = None + x_56 = transpose_10.reshape(1, 401, 128) + transpose_10 = None + x_57 = torch._C._nn.linear( + x_56, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_56 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_58 = torch.nn.functional.dropout(x_57, 0.0, False, False) + x_57 = None + x_59 = tmp_1 + x_58 + tmp_1 = x_58 = None + layer_norm_15 = torch.nn.functional.layer_norm( + x_59, + (128,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_60 = torch._C._nn.linear( + layer_norm_15, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_15 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_61 = torch._C._nn.gelu(x_60, approximate="none") + x_60 = None + x_62 = torch.nn.functional.dropout(x_61, 0.0, False, False) + x_61 = None + x_63 = torch._C._nn.linear( + x_62, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_62 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_64 = torch.nn.functional.dropout(x_63, 0.0, False, False) + x_63 = None + x_65 = x_59 + x_64 + x_59 = x_64 = None + layer_norm_16 = torch.nn.functional.layer_norm( + tmp_3, + (256,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_32 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_16 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_18 = linear_32.reshape(1, 197, 3, 4, 64) + linear_32 = None + qkv_5 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_7 = unbind_5[0] + k_7 = unbind_5[1] + v_7 = unbind_5[2] + unbind_5 = None + x_66 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_11 = x_66.transpose(1, 2) + x_66 = None + x_67 = transpose_11.reshape(1, 197, 256) + transpose_11 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_67 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = tmp_3 + x_69 + tmp_3 = x_69 = None + layer_norm_17 = torch.nn.functional.layer_norm( + x_70, + (256,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_71 = torch._C._nn.linear( + layer_norm_17, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_17 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_72 = torch._C._nn.gelu(x_71, approximate="none") + x_71 = None + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = torch._C._nn.linear( + x_73, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_73 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_75 = torch.nn.functional.dropout(x_74, 0.0, False, False) + x_74 = None + x_76 = x_70 + x_75 + x_70 = x_75 = None + layer_norm_18 = torch.nn.functional.layer_norm( + x_76, + (256,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_18 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_20 = linear_36.reshape(1, 197, 3, 4, 64) + linear_36 = None + qkv_6 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_8 = unbind_6[0] + k_8 = unbind_6[1] + v_8 = unbind_6[2] + unbind_6 = None + x_77 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_12 = x_77.transpose(1, 2) + x_77 = None + x_78 = transpose_12.reshape(1, 197, 256) + transpose_12 = None + x_79 = torch._C._nn.linear( + x_78, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_78 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = x_76 + x_80 + x_76 = x_80 = None + layer_norm_19 = torch.nn.functional.layer_norm( + x_81, + (256,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_82 = torch._C._nn.linear( + layer_norm_19, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_19 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_83 = torch._C._nn.gelu(x_82, approximate="none") + x_82 = None + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = torch._C._nn.linear( + x_84, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_84 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_86 = torch.nn.functional.dropout(x_85, 0.0, False, False) + x_85 = None + x_87 = x_81 + x_86 + x_81 = x_86 = None + layer_norm_20 = torch.nn.functional.layer_norm( + x_87, + (256,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_20 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_22 = linear_40.reshape(1, 197, 3, 4, 64) + linear_40 = None + qkv_7 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_9 = unbind_7[0] + k_9 = unbind_7[1] + v_9 = unbind_7[2] + unbind_7 = None + x_88 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_13 = x_88.transpose(1, 2) + x_88 = None + x_89 = transpose_13.reshape(1, 197, 256) + transpose_13 = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_89 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = x_87 + x_91 + x_87 = x_91 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_92, + (256,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_93 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_21 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_94 = torch._C._nn.gelu(x_93, approximate="none") + x_93 = None + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = torch._C._nn.linear( + x_95, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_95 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_97 = torch.nn.functional.dropout(x_96, 0.0, False, False) + x_96 = None + x_98 = x_92 + x_97 + x_92 = x_97 = None + getitem_52 = x_65[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_23 = torch.nn.functional.layer_norm( + getitem_52, + (128,), + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_52 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_24 = torch._C._nn.gelu(input_23, approximate="none") + input_23 = None + input_25 = torch._C._nn.linear( + input_24, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_24 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_53 = x_98[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_26 = torch.nn.functional.layer_norm( + getitem_53, + (256,), + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_53 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_27 = torch._C._nn.gelu(input_26, approximate="none") + input_26 = None + input_28 = torch._C._nn.linear( + input_27, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_27 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_54 = x_98[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_4 = torch.cat((input_25, getitem_54), dim=1) + input_25 = getitem_54 = None + getitem_55 = tmp_4[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_24 = torch.nn.functional.layer_norm( + tmp_4, + (256,), + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_4 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_56 = layer_norm_24[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_46 = torch._C._nn.linear( + getitem_56, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_56 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_24 = linear_46.reshape(1, 1, 4, 64) + linear_46 = None + q_10 = reshape_24.permute(0, 2, 1, 3) + reshape_24 = None + linear_47 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_25 = linear_47.reshape(1, 197, 4, 64) + linear_47 = None + k_10 = reshape_25.permute(0, 2, 1, 3) + reshape_25 = None + linear_48 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_24 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_26 = linear_48.reshape(1, 197, 4, 64) + linear_48 = None + v_10 = reshape_26.permute(0, 2, 1, 3) + reshape_26 = None + transpose_14 = k_10.transpose(-2, -1) + k_10 = None + matmul_4 = q_10 @ transpose_14 + q_10 = transpose_14 = None + attn_6 = matmul_4 * 0.125 + matmul_4 = None + attn_7 = attn_6.softmax(dim=-1) + attn_6 = None + attn_8 = torch.nn.functional.dropout(attn_7, 0.0, False, False) + attn_7 = None + matmul_5 = attn_8 @ v_10 + attn_8 = v_10 = None + transpose_15 = matmul_5.transpose(1, 2) + matmul_5 = None + x_99 = transpose_15.reshape(1, 1, 256) + transpose_15 = None + x_100 = torch._C._nn.linear( + x_99, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_99 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_101 = torch.nn.functional.dropout(x_100, 0.0, False, False) + x_100 = None + x_102 = getitem_55 + x_101 + getitem_55 = x_101 = None + getitem_57 = x_102[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_102 = None + input_29 = torch.nn.functional.layer_norm( + getitem_57, + (256,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_57 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_30 = torch._C._nn.gelu(input_29, approximate="none") + input_29 = None + input_31 = torch._C._nn.linear( + input_30, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_30 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_58 = x_65[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_5 = torch.cat((input_31, getitem_58), dim=1) + input_31 = getitem_58 = None + getitem_59 = x_65[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_65 = None + tmp_6 = torch.cat((input_28, getitem_59), dim=1) + input_28 = getitem_59 = None + getitem_60 = tmp_6[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_26 = torch.nn.functional.layer_norm( + tmp_6, + (128,), + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_6 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_61 = layer_norm_26[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_51 = torch._C._nn.linear( + getitem_61, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_61 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_28 = linear_51.reshape(1, 1, 4, 32) + linear_51 = None + q_11 = reshape_28.permute(0, 2, 1, 3) + reshape_28 = None + linear_52 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_29 = linear_52.reshape(1, 401, 4, 32) + linear_52 = None + k_11 = reshape_29.permute(0, 2, 1, 3) + reshape_29 = None + linear_53 = torch._C._nn.linear( + layer_norm_26, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_26 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_30 = linear_53.reshape(1, 401, 4, 32) + linear_53 = None + v_11 = reshape_30.permute(0, 2, 1, 3) + reshape_30 = None + transpose_16 = k_11.transpose(-2, -1) + k_11 = None + matmul_6 = q_11 @ transpose_16 + q_11 = transpose_16 = None + attn_9 = matmul_6 * 0.1767766952966369 + matmul_6 = None + attn_10 = attn_9.softmax(dim=-1) + attn_9 = None + attn_11 = torch.nn.functional.dropout(attn_10, 0.0, False, False) + attn_10 = None + matmul_7 = attn_11 @ v_11 + attn_11 = v_11 = None + transpose_17 = matmul_7.transpose(1, 2) + matmul_7 = None + x_103 = transpose_17.reshape(1, 1, 128) + transpose_17 = None + x_104 = torch._C._nn.linear( + x_103, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_103 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = getitem_60 + x_105 + getitem_60 = x_105 = None + getitem_62 = x_106[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_106 = None + input_32 = torch.nn.functional.layer_norm( + getitem_62, + (128,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_62 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_33 = torch._C._nn.gelu(input_32, approximate="none") + input_32 = None + input_34 = torch._C._nn.linear( + input_33, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_33 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_63 = x_98[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_98 = None + tmp_7 = torch.cat((input_34, getitem_63), dim=1) + input_34 = getitem_63 = None + layer_norm_28 = torch.nn.functional.layer_norm( + tmp_5, + (128,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_56 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_28 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_32 = linear_56.reshape(1, 401, 3, 4, 32) + linear_56 = None + qkv_8 = reshape_32.permute(2, 0, 3, 1, 4) + reshape_32 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_12 = unbind_8[0] + k_12 = unbind_8[1] + v_12 = unbind_8[2] + unbind_8 = None + x_107 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_18 = x_107.transpose(1, 2) + x_107 = None + x_108 = transpose_18.reshape(1, 401, 128) + transpose_18 = None + x_109 = torch._C._nn.linear( + x_108, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_108 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_110 = torch.nn.functional.dropout(x_109, 0.0, False, False) + x_109 = None + x_111 = tmp_5 + x_110 + tmp_5 = x_110 = None + layer_norm_29 = torch.nn.functional.layer_norm( + x_111, + (128,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_112 = torch._C._nn.linear( + layer_norm_29, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_29 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_113 = torch._C._nn.gelu(x_112, approximate="none") + x_112 = None + x_114 = torch.nn.functional.dropout(x_113, 0.0, False, False) + x_113 = None + x_115 = torch._C._nn.linear( + x_114, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_114 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_116 = torch.nn.functional.dropout(x_115, 0.0, False, False) + x_115 = None + x_117 = x_111 + x_116 + x_111 = x_116 = None + layer_norm_30 = torch.nn.functional.layer_norm( + tmp_7, + (256,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_60 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_30 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_34 = linear_60.reshape(1, 197, 3, 4, 64) + linear_60 = None + qkv_9 = reshape_34.permute(2, 0, 3, 1, 4) + reshape_34 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_13 = unbind_9[0] + k_13 = unbind_9[1] + v_13 = unbind_9[2] + unbind_9 = None + x_118 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_19 = x_118.transpose(1, 2) + x_118 = None + x_119 = transpose_19.reshape(1, 197, 256) + transpose_19 = None + x_120 = torch._C._nn.linear( + x_119, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_119 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_121 = torch.nn.functional.dropout(x_120, 0.0, False, False) + x_120 = None + x_122 = tmp_7 + x_121 + tmp_7 = x_121 = None + layer_norm_31 = torch.nn.functional.layer_norm( + x_122, + (256,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_123 = torch._C._nn.linear( + layer_norm_31, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_31 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_124 = torch._C._nn.gelu(x_123, approximate="none") + x_123 = None + x_125 = torch.nn.functional.dropout(x_124, 0.0, False, False) + x_124 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_125 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = x_122 + x_127 + x_122 = x_127 = None + layer_norm_32 = torch.nn.functional.layer_norm( + x_128, + (256,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_32 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_36 = linear_64.reshape(1, 197, 3, 4, 64) + linear_64 = None + qkv_10 = reshape_36.permute(2, 0, 3, 1, 4) + reshape_36 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_14 = unbind_10[0] + k_14 = unbind_10[1] + v_14 = unbind_10[2] + unbind_10 = None + x_129 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_20 = x_129.transpose(1, 2) + x_129 = None + x_130 = transpose_20.reshape(1, 197, 256) + transpose_20 = None + x_131 = torch._C._nn.linear( + x_130, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_130 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_132 = torch.nn.functional.dropout(x_131, 0.0, False, False) + x_131 = None + x_133 = x_128 + x_132 + x_128 = x_132 = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_133, + (256,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_134 = torch._C._nn.linear( + layer_norm_33, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_33 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_135 = torch._C._nn.gelu(x_134, approximate="none") + x_134 = None + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_133 + x_138 + x_133 = x_138 = None + layer_norm_34 = torch.nn.functional.layer_norm( + x_139, + (256,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_68 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_34 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_38 = linear_68.reshape(1, 197, 3, 4, 64) + linear_68 = None + qkv_11 = reshape_38.permute(2, 0, 3, 1, 4) + reshape_38 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_15 = unbind_11[0] + k_15 = unbind_11[1] + v_15 = unbind_11[2] + unbind_11 = None + x_140 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_21 = x_140.transpose(1, 2) + x_140 = None + x_141 = transpose_21.reshape(1, 197, 256) + transpose_21 = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_141 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_143 = torch.nn.functional.dropout(x_142, 0.0, False, False) + x_142 = None + x_144 = x_139 + x_143 + x_139 = x_143 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_144, + (256,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_145 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_35 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_146 = torch._C._nn.gelu(x_145, approximate="none") + x_145 = None + x_147 = torch.nn.functional.dropout(x_146, 0.0, False, False) + x_146 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_144 + x_149 + x_144 = x_149 = None + getitem_76 = x_117[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_35 = torch.nn.functional.layer_norm( + getitem_76, + (128,), + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_76 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_36 = torch._C._nn.gelu(input_35, approximate="none") + input_35 = None + input_37 = torch._C._nn.linear( + input_36, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_36 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_77 = x_150[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_38 = torch.nn.functional.layer_norm( + getitem_77, + (256,), + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_77 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_39 = torch._C._nn.gelu(input_38, approximate="none") + input_38 = None + input_40 = torch._C._nn.linear( + input_39, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_39 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_78 = x_150[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_8 = torch.cat((input_37, getitem_78), dim=1) + input_37 = getitem_78 = None + getitem_79 = tmp_8[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_38 = torch.nn.functional.layer_norm( + tmp_8, + (256,), + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_8 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_80 = layer_norm_38[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_74 = torch._C._nn.linear( + getitem_80, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_80 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_40 = linear_74.reshape(1, 1, 4, 64) + linear_74 = None + q_16 = reshape_40.permute(0, 2, 1, 3) + reshape_40 = None + linear_75 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_41 = linear_75.reshape(1, 197, 4, 64) + linear_75 = None + k_16 = reshape_41.permute(0, 2, 1, 3) + reshape_41 = None + linear_76 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_38 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_42 = linear_76.reshape(1, 197, 4, 64) + linear_76 = None + v_16 = reshape_42.permute(0, 2, 1, 3) + reshape_42 = None + transpose_22 = k_16.transpose(-2, -1) + k_16 = None + matmul_8 = q_16 @ transpose_22 + q_16 = transpose_22 = None + attn_12 = matmul_8 * 0.125 + matmul_8 = None + attn_13 = attn_12.softmax(dim=-1) + attn_12 = None + attn_14 = torch.nn.functional.dropout(attn_13, 0.0, False, False) + attn_13 = None + matmul_9 = attn_14 @ v_16 + attn_14 = v_16 = None + transpose_23 = matmul_9.transpose(1, 2) + matmul_9 = None + x_151 = transpose_23.reshape(1, 1, 256) + transpose_23 = None + x_152 = torch._C._nn.linear( + x_151, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_151 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_153 = torch.nn.functional.dropout(x_152, 0.0, False, False) + x_152 = None + x_154 = getitem_79 + x_153 + getitem_79 = x_153 = None + getitem_81 = x_154[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_154 = None + input_41 = torch.nn.functional.layer_norm( + getitem_81, + (256,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_81 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_42 = torch._C._nn.gelu(input_41, approximate="none") + input_41 = None + input_43 = torch._C._nn.linear( + input_42, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_42 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_82 = x_117[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_9 = torch.cat((input_43, getitem_82), dim=1) + input_43 = getitem_82 = None + getitem_83 = x_117[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_117 = None + tmp_10 = torch.cat((input_40, getitem_83), dim=1) + input_40 = getitem_83 = None + getitem_84 = tmp_10[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_40 = torch.nn.functional.layer_norm( + tmp_10, + (128,), + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_10 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_85 = layer_norm_40[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_79 = torch._C._nn.linear( + getitem_85, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_85 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_44 = linear_79.reshape(1, 1, 4, 32) + linear_79 = None + q_17 = reshape_44.permute(0, 2, 1, 3) + reshape_44 = None + linear_80 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_45 = linear_80.reshape(1, 401, 4, 32) + linear_80 = None + k_17 = reshape_45.permute(0, 2, 1, 3) + reshape_45 = None + linear_81 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_40 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_46 = linear_81.reshape(1, 401, 4, 32) + linear_81 = None + v_17 = reshape_46.permute(0, 2, 1, 3) + reshape_46 = None + transpose_24 = k_17.transpose(-2, -1) + k_17 = None + matmul_10 = q_17 @ transpose_24 + q_17 = transpose_24 = None + attn_15 = matmul_10 * 0.1767766952966369 + matmul_10 = None + attn_16 = attn_15.softmax(dim=-1) + attn_15 = None + attn_17 = torch.nn.functional.dropout(attn_16, 0.0, False, False) + attn_16 = None + matmul_11 = attn_17 @ v_17 + attn_17 = v_17 = None + transpose_25 = matmul_11.transpose(1, 2) + matmul_11 = None + x_155 = transpose_25.reshape(1, 1, 128) + transpose_25 = None + x_156 = torch._C._nn.linear( + x_155, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_155 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = getitem_84 + x_157 + getitem_84 = x_157 = None + getitem_86 = x_158[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_158 = None + input_44 = torch.nn.functional.layer_norm( + getitem_86, + (128,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_86 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_45 = torch._C._nn.gelu(input_44, approximate="none") + input_44 = None + input_46 = torch._C._nn.linear( + input_45, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_45 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_87 = x_150[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_150 = None + tmp_11 = torch.cat((input_46, getitem_87), dim=1) + input_46 = getitem_87 = None + x_159 = torch.nn.functional.layer_norm( + tmp_9, + (128,), + l_self_modules_norm_modules_0_parameters_weight_, + l_self_modules_norm_modules_0_parameters_bias_, + 1e-06, + ) + tmp_9 = ( + l_self_modules_norm_modules_0_parameters_weight_ + ) = l_self_modules_norm_modules_0_parameters_bias_ = None + x_160 = torch.nn.functional.layer_norm( + tmp_11, + (256,), + l_self_modules_norm_modules_1_parameters_weight_, + l_self_modules_norm_modules_1_parameters_bias_, + 1e-06, + ) + tmp_11 = ( + l_self_modules_norm_modules_1_parameters_weight_ + ) = l_self_modules_norm_modules_1_parameters_bias_ = None + x_161 = x_159[(slice(None, None, None), 0)] + x_159 = None + x_162 = x_160[(slice(None, None, None), 0)] + x_160 = None + dropout_50 = torch.nn.functional.dropout(x_161, 0.0, False, False) + x_161 = None + dropout_51 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + linear_84 = torch._C._nn.linear( + dropout_50, + l_self_modules_head_modules_0_parameters_weight_, + l_self_modules_head_modules_0_parameters_bias_, + ) + dropout_50 = ( + l_self_modules_head_modules_0_parameters_weight_ + ) = l_self_modules_head_modules_0_parameters_bias_ = None + linear_85 = torch._C._nn.linear( + dropout_51, + l_self_modules_head_modules_1_parameters_weight_, + l_self_modules_head_modules_1_parameters_bias_, + ) + dropout_51 = ( + l_self_modules_head_modules_1_parameters_weight_ + ) = l_self_modules_head_modules_1_parameters_bias_ = None + stack = torch.stack([linear_84, linear_85], dim=0) + linear_84 = linear_85 = None + x_163 = torch.mean(stack, dim=0) + stack = None + return (x_163,) diff --git a/samples/timm/crossvit_9_dagger_240.in1k/weight_meta.py b/samples/timm/crossvit_9_dagger_240.in1k/weight_meta.py new file mode 100644 index 0000000000..2441199f56 --- /dev/null +++ b/samples/timm/crossvit_9_dagger_240.in1k/weight_meta.py @@ -0,0 +1,2792 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.229 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_weight_" + ) + shape = [32, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_0_parameters_bias_" + ) + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_weight_" + ) + shape = [64, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_2_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_weight_" + ) + shape = [128, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_0_modules_proj_modules_4_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.023 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_0_: + name = "L_self_parameters_cls_token_0_" + shape = [1, 1, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_0_: + name = "L_self_parameters_pos_embed_0_" + shape = [1, 401, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_weight_" + ) + shape = [64, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_0_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.014 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_weight_" + ) + shape = [128, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_2_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_weight_" + ) + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.017 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_: + name = ( + "L_self_modules_patch_embed_modules_1_modules_proj_modules_4_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.017 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_1_: + name = "L_self_parameters_cls_token_1_" + shape = [1, 1, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_1_: + name = "L_self_parameters_pos_embed_1_" + shape = [1, 197, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [256, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [128, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [128, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [256, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [256, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [128, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [128, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [256, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [128, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [256, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [256, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [128, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [256, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [128, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [128, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [256, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_weight_: + name = "L_self_modules_norm_modules_0_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_bias_: + name = "L_self_modules_norm_modules_0_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_weight_: + name = "L_self_modules_norm_modules_1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_bias_: + name = "L_self_modules_norm_modules_1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_weight_: + name = "L_self_modules_head_modules_0_parameters_weight_" + shape = [1000, 128] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_bias_: + name = "L_self_modules_head_modules_0_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_weight_: + name = "L_self_modules_head_modules_1_parameters_weight_" + shape = [1000, 256] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_bias_: + name = "L_self_modules_head_modules_1_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/crossvit_base_240.in1k/graph_hash.txt b/samples/timm/crossvit_base_240.in1k/graph_hash.txt new file mode 100644 index 0000000000..66cadf2f24 --- /dev/null +++ b/samples/timm/crossvit_base_240.in1k/graph_hash.txt @@ -0,0 +1 @@ +e0627bcc7072a0e864ffc9d221b879ec6bdfca1250ff5d9551b6586a43c220dd \ No newline at end of file diff --git a/samples/timm/crossvit_base_240.in1k/graph_net.json b/samples/timm/crossvit_base_240.in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/crossvit_base_240.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/crossvit_base_240.in1k/input_meta.py b/samples/timm/crossvit_base_240.in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/crossvit_base_240.in1k/input_tensor_constraints.py b/samples/timm/crossvit_base_240.in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/crossvit_base_240.in1k/model.py b/samples/timm/crossvit_base_240.in1k/model.py new file mode 100644 index 0000000000..4ff3bd3304 --- /dev/null +++ b/samples/timm/crossvit_base_240.in1k/model.py @@ -0,0 +1,2451 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_0_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_0_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_1_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ + ) + l_self_parameters_cls_token_0_ = L_self_parameters_cls_token_0_ + l_self_parameters_pos_embed_0_ = L_self_parameters_pos_embed_0_ + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ + ) + l_self_parameters_cls_token_1_ = L_self_parameters_cls_token_1_ + l_self_parameters_pos_embed_1_ = L_self_parameters_pos_embed_1_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_norm_modules_0_parameters_weight_ = ( + L_self_modules_norm_modules_0_parameters_weight_ + ) + l_self_modules_norm_modules_0_parameters_bias_ = ( + L_self_modules_norm_modules_0_parameters_bias_ + ) + l_self_modules_norm_modules_1_parameters_weight_ = ( + L_self_modules_norm_modules_1_parameters_weight_ + ) + l_self_modules_norm_modules_1_parameters_bias_ = ( + L_self_modules_norm_modules_1_parameters_bias_ + ) + l_self_modules_head_modules_0_parameters_weight_ = ( + L_self_modules_head_modules_0_parameters_weight_ + ) + l_self_modules_head_modules_0_parameters_bias_ = ( + L_self_modules_head_modules_0_parameters_bias_ + ) + l_self_modules_head_modules_1_parameters_weight_ = ( + L_self_modules_head_modules_1_parameters_weight_ + ) + l_self_modules_head_modules_1_parameters_bias_ = ( + L_self_modules_head_modules_1_parameters_bias_ + ) + x = torch.nn.functional.interpolate( + l_x_, size=(240, 240), mode="bicubic", align_corners=False + ) + conv2d = torch.conv2d( + x, + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_, + (12, 12), + (0, 0), + (1, 1), + 1, + ) + x = ( + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ = None + flatten = conv2d.flatten(2) + conv2d = None + x_1 = flatten.transpose(1, 2) + flatten = None + cls_tokens = l_self_parameters_cls_token_0_.expand(1, -1, -1) + l_self_parameters_cls_token_0_ = None + x_ = torch.cat((cls_tokens, x_1), dim=1) + cls_tokens = x_1 = None + x__1 = x_ + l_self_parameters_pos_embed_0_ + x_ = l_self_parameters_pos_embed_0_ = None + x__2 = torch.nn.functional.dropout(x__1, 0.0, False, False) + x__1 = None + conv2d_1 = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ = None + flatten_1 = conv2d_1.flatten(2) + conv2d_1 = None + x_2 = flatten_1.transpose(1, 2) + flatten_1 = None + cls_tokens_1 = l_self_parameters_cls_token_1_.expand(1, -1, -1) + l_self_parameters_cls_token_1_ = None + x__3 = torch.cat((cls_tokens_1, x_2), dim=1) + cls_tokens_1 = x_2 = None + x__4 = x__3 + l_self_parameters_pos_embed_1_ + x__3 = l_self_parameters_pos_embed_1_ = None + x__5 = torch.nn.functional.dropout(x__4, 0.0, False, False) + x__4 = None + layer_norm = torch.nn.functional.layer_norm( + x__2, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear = torch._C._nn.linear( + layer_norm, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 401, 3, 12, 32) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_3 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_2 = x_3.transpose(1, 2) + x_3 = None + x_4 = transpose_2.reshape(1, 401, 384) + transpose_2 = None + x_5 = torch._C._nn.linear( + x_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_6 = torch.nn.functional.dropout(x_5, 0.0, False, False) + x_5 = None + x_7 = x__2 + x_6 + x__2 = x_6 = None + layer_norm_1 = torch.nn.functional.layer_norm( + x_7, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_8 = torch._C._nn.linear( + layer_norm_1, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_1 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_9 = torch._C._nn.gelu(x_8, approximate="none") + x_8 = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = torch._C._nn.linear( + x_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_7 + x_12 + x_7 = x_12 = None + layer_norm_2 = torch.nn.functional.layer_norm( + x__5, + (768,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_2, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_2 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_2 = linear_4.reshape(1, 197, 3, 12, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_14 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_3 = x_14.transpose(1, 2) + x_14 = None + x_15 = transpose_3.reshape(1, 197, 768) + transpose_3 = None + x_16 = torch._C._nn.linear( + x_15, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_15 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = x__5 + x_17 + x__5 = x_17 = None + layer_norm_3 = torch.nn.functional.layer_norm( + x_18, + (768,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_19 = torch._C._nn.linear( + layer_norm_3, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_3 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_20 = torch._C._nn.gelu(x_19, approximate="none") + x_19 = None + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = torch._C._nn.linear( + x_21, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_21 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_18 + x_23 + x_18 = x_23 = None + layer_norm_4 = torch.nn.functional.layer_norm( + x_24, + (768,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_8.reshape(1, 197, 3, 12, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_25 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_4 = x_25.transpose(1, 2) + x_25 = None + x_26 = transpose_4.reshape(1, 197, 768) + transpose_4 = None + x_27 = torch._C._nn.linear( + x_26, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_26 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_28 = torch.nn.functional.dropout(x_27, 0.0, False, False) + x_27 = None + x_29 = x_24 + x_28 + x_24 = x_28 = None + layer_norm_5 = torch.nn.functional.layer_norm( + x_29, + (768,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_30 = torch._C._nn.linear( + layer_norm_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_31 = torch._C._nn.gelu(x_30, approximate="none") + x_30 = None + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = torch._C._nn.linear( + x_32, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_32 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = x_29 + x_34 + x_29 = x_34 = None + layer_norm_6 = torch.nn.functional.layer_norm( + x_35, + (768,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_6, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_6 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_6 = linear_12.reshape(1, 197, 3, 12, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_36 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_5 = x_36.transpose(1, 2) + x_36 = None + x_37 = transpose_5.reshape(1, 197, 768) + transpose_5 = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_37 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_39 = torch.nn.functional.dropout(x_38, 0.0, False, False) + x_38 = None + x_40 = x_35 + x_39 + x_35 = x_39 = None + layer_norm_7 = torch.nn.functional.layer_norm( + x_40, + (768,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_41 = torch._C._nn.linear( + layer_norm_7, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_7 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_42 = torch._C._nn.gelu(x_41, approximate="none") + x_41 = None + x_43 = torch.nn.functional.dropout(x_42, 0.0, False, False) + x_42 = None + x_44 = torch._C._nn.linear( + x_43, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_43 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = x_40 + x_45 + x_40 = x_45 = None + layer_norm_8 = torch.nn.functional.layer_norm( + x_46, + (768,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_8, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_8 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_8 = linear_16.reshape(1, 197, 3, 12, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_47 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_6 = x_47.transpose(1, 2) + x_47 = None + x_48 = transpose_6.reshape(1, 197, 768) + transpose_6 = None + x_49 = torch._C._nn.linear( + x_48, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_48 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_50 = torch.nn.functional.dropout(x_49, 0.0, False, False) + x_49 = None + x_51 = x_46 + x_50 + x_46 = x_50 = None + layer_norm_9 = torch.nn.functional.layer_norm( + x_51, + (768,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_52 = torch._C._nn.linear( + layer_norm_9, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_9 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_53 = torch._C._nn.gelu(x_52, approximate="none") + x_52 = None + x_54 = torch.nn.functional.dropout(x_53, 0.0, False, False) + x_53 = None + x_55 = torch._C._nn.linear( + x_54, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_54 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_56 = torch.nn.functional.dropout(x_55, 0.0, False, False) + x_55 = None + x_57 = x_51 + x_56 + x_51 = x_56 = None + getitem_31 = x_13[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_1 = torch.nn.functional.layer_norm( + getitem_31, + (384,), + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_31 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_2 = torch._C._nn.gelu(input_1, approximate="none") + input_1 = None + input_3 = torch._C._nn.linear( + input_2, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_2 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_32 = x_57[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_4 = torch.nn.functional.layer_norm( + getitem_32, + (768,), + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_32 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_5 = torch._C._nn.gelu(input_4, approximate="none") + input_4 = None + input_6 = torch._C._nn.linear( + input_5, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_5 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_33 = x_57[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp = torch.cat((input_3, getitem_33), dim=1) + input_3 = getitem_33 = None + getitem_34 = tmp[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_12 = torch.nn.functional.layer_norm( + tmp, + (768,), + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_35 = layer_norm_12[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_22 = torch._C._nn.linear( + getitem_35, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_35 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_10 = linear_22.reshape(1, 1, 12, 64) + linear_22 = None + q_5 = reshape_10.permute(0, 2, 1, 3) + reshape_10 = None + linear_23 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_11 = linear_23.reshape(1, 197, 12, 64) + linear_23 = None + k_5 = reshape_11.permute(0, 2, 1, 3) + reshape_11 = None + linear_24 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_12 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_12 = linear_24.reshape(1, 197, 12, 64) + linear_24 = None + v_5 = reshape_12.permute(0, 2, 1, 3) + reshape_12 = None + transpose_7 = k_5.transpose(-2, -1) + k_5 = None + matmul = q_5 @ transpose_7 + q_5 = transpose_7 = None + attn = matmul * 0.125 + matmul = None + attn_1 = attn.softmax(dim=-1) + attn = None + attn_2 = torch.nn.functional.dropout(attn_1, 0.0, False, False) + attn_1 = None + matmul_1 = attn_2 @ v_5 + attn_2 = v_5 = None + transpose_8 = matmul_1.transpose(1, 2) + matmul_1 = None + x_58 = transpose_8.reshape(1, 1, 768) + transpose_8 = None + x_59 = torch._C._nn.linear( + x_58, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_58 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_60 = torch.nn.functional.dropout(x_59, 0.0, False, False) + x_59 = None + x_61 = getitem_34 + x_60 + getitem_34 = x_60 = None + getitem_36 = x_61[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_61 = None + input_7 = torch.nn.functional.layer_norm( + getitem_36, + (768,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_36 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_8 = torch._C._nn.gelu(input_7, approximate="none") + input_7 = None + input_9 = torch._C._nn.linear( + input_8, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_8 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_37 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_1 = torch.cat((input_9, getitem_37), dim=1) + input_9 = getitem_37 = None + getitem_38 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_13 = None + tmp_2 = torch.cat((input_6, getitem_38), dim=1) + input_6 = getitem_38 = None + getitem_39 = tmp_2[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_14 = torch.nn.functional.layer_norm( + tmp_2, + (384,), + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_2 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_40 = layer_norm_14[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_27 = torch._C._nn.linear( + getitem_40, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_40 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_14 = linear_27.reshape(1, 1, 12, 32) + linear_27 = None + q_6 = reshape_14.permute(0, 2, 1, 3) + reshape_14 = None + linear_28 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_15 = linear_28.reshape(1, 401, 12, 32) + linear_28 = None + k_6 = reshape_15.permute(0, 2, 1, 3) + reshape_15 = None + linear_29 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_14 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_16 = linear_29.reshape(1, 401, 12, 32) + linear_29 = None + v_6 = reshape_16.permute(0, 2, 1, 3) + reshape_16 = None + transpose_9 = k_6.transpose(-2, -1) + k_6 = None + matmul_2 = q_6 @ transpose_9 + q_6 = transpose_9 = None + attn_3 = matmul_2 * 0.1767766952966369 + matmul_2 = None + attn_4 = attn_3.softmax(dim=-1) + attn_3 = None + attn_5 = torch.nn.functional.dropout(attn_4, 0.0, False, False) + attn_4 = None + matmul_3 = attn_5 @ v_6 + attn_5 = v_6 = None + transpose_10 = matmul_3.transpose(1, 2) + matmul_3 = None + x_62 = transpose_10.reshape(1, 1, 384) + transpose_10 = None + x_63 = torch._C._nn.linear( + x_62, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_62 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_64 = torch.nn.functional.dropout(x_63, 0.0, False, False) + x_63 = None + x_65 = getitem_39 + x_64 + getitem_39 = x_64 = None + getitem_41 = x_65[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_65 = None + input_10 = torch.nn.functional.layer_norm( + getitem_41, + (384,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_41 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_11 = torch._C._nn.gelu(input_10, approximate="none") + input_10 = None + input_12 = torch._C._nn.linear( + input_11, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_11 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_42 = x_57[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_57 = None + tmp_3 = torch.cat((input_12, getitem_42), dim=1) + input_12 = getitem_42 = None + layer_norm_16 = torch.nn.functional.layer_norm( + tmp_1, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_32 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_16 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_18 = linear_32.reshape(1, 401, 3, 12, 32) + linear_32 = None + qkv_5 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_7 = unbind_5[0] + k_7 = unbind_5[1] + v_7 = unbind_5[2] + unbind_5 = None + x_66 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_11 = x_66.transpose(1, 2) + x_66 = None + x_67 = transpose_11.reshape(1, 401, 384) + transpose_11 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_67 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = tmp_1 + x_69 + tmp_1 = x_69 = None + layer_norm_17 = torch.nn.functional.layer_norm( + x_70, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_71 = torch._C._nn.linear( + layer_norm_17, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_17 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_72 = torch._C._nn.gelu(x_71, approximate="none") + x_71 = None + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = torch._C._nn.linear( + x_73, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_73 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_75 = torch.nn.functional.dropout(x_74, 0.0, False, False) + x_74 = None + x_76 = x_70 + x_75 + x_70 = x_75 = None + layer_norm_18 = torch.nn.functional.layer_norm( + tmp_3, + (768,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_18 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_20 = linear_36.reshape(1, 197, 3, 12, 64) + linear_36 = None + qkv_6 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_8 = unbind_6[0] + k_8 = unbind_6[1] + v_8 = unbind_6[2] + unbind_6 = None + x_77 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_12 = x_77.transpose(1, 2) + x_77 = None + x_78 = transpose_12.reshape(1, 197, 768) + transpose_12 = None + x_79 = torch._C._nn.linear( + x_78, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_78 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = tmp_3 + x_80 + tmp_3 = x_80 = None + layer_norm_19 = torch.nn.functional.layer_norm( + x_81, + (768,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_82 = torch._C._nn.linear( + layer_norm_19, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_19 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_83 = torch._C._nn.gelu(x_82, approximate="none") + x_82 = None + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = torch._C._nn.linear( + x_84, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_84 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_86 = torch.nn.functional.dropout(x_85, 0.0, False, False) + x_85 = None + x_87 = x_81 + x_86 + x_81 = x_86 = None + layer_norm_20 = torch.nn.functional.layer_norm( + x_87, + (768,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_20 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_22 = linear_40.reshape(1, 197, 3, 12, 64) + linear_40 = None + qkv_7 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_9 = unbind_7[0] + k_9 = unbind_7[1] + v_9 = unbind_7[2] + unbind_7 = None + x_88 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_13 = x_88.transpose(1, 2) + x_88 = None + x_89 = transpose_13.reshape(1, 197, 768) + transpose_13 = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_89 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = x_87 + x_91 + x_87 = x_91 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_92, + (768,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_93 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_21 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_94 = torch._C._nn.gelu(x_93, approximate="none") + x_93 = None + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = torch._C._nn.linear( + x_95, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_95 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_97 = torch.nn.functional.dropout(x_96, 0.0, False, False) + x_96 = None + x_98 = x_92 + x_97 + x_92 = x_97 = None + layer_norm_22 = torch.nn.functional.layer_norm( + x_98, + (768,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_22, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_22 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_24 = linear_44.reshape(1, 197, 3, 12, 64) + linear_44 = None + qkv_8 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_10 = unbind_8[0] + k_10 = unbind_8[1] + v_10 = unbind_8[2] + unbind_8 = None + x_99 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_14 = x_99.transpose(1, 2) + x_99 = None + x_100 = transpose_14.reshape(1, 197, 768) + transpose_14 = None + x_101 = torch._C._nn.linear( + x_100, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_100 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = x_98 + x_102 + x_98 = x_102 = None + layer_norm_23 = torch.nn.functional.layer_norm( + x_103, + (768,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_104 = torch._C._nn.linear( + layer_norm_23, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_23 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_105 = torch._C._nn.gelu(x_104, approximate="none") + x_104 = None + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch._C._nn.linear( + x_106, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_106 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_103 + x_108 + x_103 = x_108 = None + layer_norm_24 = torch.nn.functional.layer_norm( + x_109, + (768,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_24 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_26 = linear_48.reshape(1, 197, 3, 12, 64) + linear_48 = None + qkv_9 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_11 = unbind_9[0] + k_11 = unbind_9[1] + v_11 = unbind_9[2] + unbind_9 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_15 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_15.reshape(1, 197, 768) + transpose_15 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_109 + x_113 + x_109 = x_113 = None + layer_norm_25 = torch.nn.functional.layer_norm( + x_114, + (768,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_115 = torch._C._nn.linear( + layer_norm_25, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_25 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_116 = torch._C._nn.gelu(x_115, approximate="none") + x_115 = None + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = torch._C._nn.linear( + x_117, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_117 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_119 = torch.nn.functional.dropout(x_118, 0.0, False, False) + x_118 = None + x_120 = x_114 + x_119 + x_114 = x_119 = None + getitem_58 = x_76[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_13 = torch.nn.functional.layer_norm( + getitem_58, + (384,), + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_58 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_14 = torch._C._nn.gelu(input_13, approximate="none") + input_13 = None + input_15 = torch._C._nn.linear( + input_14, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_14 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_59 = x_120[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_16 = torch.nn.functional.layer_norm( + getitem_59, + (768,), + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_59 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_17 = torch._C._nn.gelu(input_16, approximate="none") + input_16 = None + input_18 = torch._C._nn.linear( + input_17, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_17 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_60 = x_120[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_4 = torch.cat((input_15, getitem_60), dim=1) + input_15 = getitem_60 = None + getitem_61 = tmp_4[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_28 = torch.nn.functional.layer_norm( + tmp_4, + (768,), + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_4 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_62 = layer_norm_28[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_54 = torch._C._nn.linear( + getitem_62, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_62 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_28 = linear_54.reshape(1, 1, 12, 64) + linear_54 = None + q_12 = reshape_28.permute(0, 2, 1, 3) + reshape_28 = None + linear_55 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_29 = linear_55.reshape(1, 197, 12, 64) + linear_55 = None + k_12 = reshape_29.permute(0, 2, 1, 3) + reshape_29 = None + linear_56 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_28 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_30 = linear_56.reshape(1, 197, 12, 64) + linear_56 = None + v_12 = reshape_30.permute(0, 2, 1, 3) + reshape_30 = None + transpose_16 = k_12.transpose(-2, -1) + k_12 = None + matmul_4 = q_12 @ transpose_16 + q_12 = transpose_16 = None + attn_6 = matmul_4 * 0.125 + matmul_4 = None + attn_7 = attn_6.softmax(dim=-1) + attn_6 = None + attn_8 = torch.nn.functional.dropout(attn_7, 0.0, False, False) + attn_7 = None + matmul_5 = attn_8 @ v_12 + attn_8 = v_12 = None + transpose_17 = matmul_5.transpose(1, 2) + matmul_5 = None + x_121 = transpose_17.reshape(1, 1, 768) + transpose_17 = None + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_121 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) + x_122 = None + x_124 = getitem_61 + x_123 + getitem_61 = x_123 = None + getitem_63 = x_124[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_124 = None + input_19 = torch.nn.functional.layer_norm( + getitem_63, + (768,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_63 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_20 = torch._C._nn.gelu(input_19, approximate="none") + input_19 = None + input_21 = torch._C._nn.linear( + input_20, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_20 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_64 = x_76[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_5 = torch.cat((input_21, getitem_64), dim=1) + input_21 = getitem_64 = None + getitem_65 = x_76[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_76 = None + tmp_6 = torch.cat((input_18, getitem_65), dim=1) + input_18 = getitem_65 = None + getitem_66 = tmp_6[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_30 = torch.nn.functional.layer_norm( + tmp_6, + (384,), + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_6 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_67 = layer_norm_30[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_59 = torch._C._nn.linear( + getitem_67, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_67 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_32 = linear_59.reshape(1, 1, 12, 32) + linear_59 = None + q_13 = reshape_32.permute(0, 2, 1, 3) + reshape_32 = None + linear_60 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_33 = linear_60.reshape(1, 401, 12, 32) + linear_60 = None + k_13 = reshape_33.permute(0, 2, 1, 3) + reshape_33 = None + linear_61 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_30 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_34 = linear_61.reshape(1, 401, 12, 32) + linear_61 = None + v_13 = reshape_34.permute(0, 2, 1, 3) + reshape_34 = None + transpose_18 = k_13.transpose(-2, -1) + k_13 = None + matmul_6 = q_13 @ transpose_18 + q_13 = transpose_18 = None + attn_9 = matmul_6 * 0.1767766952966369 + matmul_6 = None + attn_10 = attn_9.softmax(dim=-1) + attn_9 = None + attn_11 = torch.nn.functional.dropout(attn_10, 0.0, False, False) + attn_10 = None + matmul_7 = attn_11 @ v_13 + attn_11 = v_13 = None + transpose_19 = matmul_7.transpose(1, 2) + matmul_7 = None + x_125 = transpose_19.reshape(1, 1, 384) + transpose_19 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_125 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = getitem_66 + x_127 + getitem_66 = x_127 = None + getitem_68 = x_128[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_128 = None + input_22 = torch.nn.functional.layer_norm( + getitem_68, + (384,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_68 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_23 = torch._C._nn.gelu(input_22, approximate="none") + input_22 = None + input_24 = torch._C._nn.linear( + input_23, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_23 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_69 = x_120[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_120 = None + tmp_7 = torch.cat((input_24, getitem_69), dim=1) + input_24 = getitem_69 = None + layer_norm_32 = torch.nn.functional.layer_norm( + tmp_5, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_32 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_36 = linear_64.reshape(1, 401, 3, 12, 32) + linear_64 = None + qkv_10 = reshape_36.permute(2, 0, 3, 1, 4) + reshape_36 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_14 = unbind_10[0] + k_14 = unbind_10[1] + v_14 = unbind_10[2] + unbind_10 = None + x_129 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_20 = x_129.transpose(1, 2) + x_129 = None + x_130 = transpose_20.reshape(1, 401, 384) + transpose_20 = None + x_131 = torch._C._nn.linear( + x_130, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_130 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_132 = torch.nn.functional.dropout(x_131, 0.0, False, False) + x_131 = None + x_133 = tmp_5 + x_132 + tmp_5 = x_132 = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_133, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_134 = torch._C._nn.linear( + layer_norm_33, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_33 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_135 = torch._C._nn.gelu(x_134, approximate="none") + x_134 = None + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_133 + x_138 + x_133 = x_138 = None + layer_norm_34 = torch.nn.functional.layer_norm( + tmp_7, + (768,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_68 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_34 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_38 = linear_68.reshape(1, 197, 3, 12, 64) + linear_68 = None + qkv_11 = reshape_38.permute(2, 0, 3, 1, 4) + reshape_38 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_15 = unbind_11[0] + k_15 = unbind_11[1] + v_15 = unbind_11[2] + unbind_11 = None + x_140 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_21 = x_140.transpose(1, 2) + x_140 = None + x_141 = transpose_21.reshape(1, 197, 768) + transpose_21 = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_141 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_143 = torch.nn.functional.dropout(x_142, 0.0, False, False) + x_142 = None + x_144 = tmp_7 + x_143 + tmp_7 = x_143 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_144, + (768,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_145 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_35 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_146 = torch._C._nn.gelu(x_145, approximate="none") + x_145 = None + x_147 = torch.nn.functional.dropout(x_146, 0.0, False, False) + x_146 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_144 + x_149 + x_144 = x_149 = None + layer_norm_36 = torch.nn.functional.layer_norm( + x_150, + (768,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_72 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_36 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_40 = linear_72.reshape(1, 197, 3, 12, 64) + linear_72 = None + qkv_12 = reshape_40.permute(2, 0, 3, 1, 4) + reshape_40 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_16 = unbind_12[0] + k_16 = unbind_12[1] + v_16 = unbind_12[2] + unbind_12 = None + x_151 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, attn_mask=None, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_22 = x_151.transpose(1, 2) + x_151 = None + x_152 = transpose_22.reshape(1, 197, 768) + transpose_22 = None + x_153 = torch._C._nn.linear( + x_152, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_152 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_154 = torch.nn.functional.dropout(x_153, 0.0, False, False) + x_153 = None + x_155 = x_150 + x_154 + x_150 = x_154 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_155, + (768,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_156 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_37 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_157 = torch._C._nn.gelu(x_156, approximate="none") + x_156 = None + x_158 = torch.nn.functional.dropout(x_157, 0.0, False, False) + x_157 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_155 + x_160 + x_155 = x_160 = None + layer_norm_38 = torch.nn.functional.layer_norm( + x_161, + (768,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_76 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_38 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_42 = linear_76.reshape(1, 197, 3, 12, 64) + linear_76 = None + qkv_13 = reshape_42.permute(2, 0, 3, 1, 4) + reshape_42 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_17 = unbind_13[0] + k_17 = unbind_13[1] + v_17 = unbind_13[2] + unbind_13 = None + x_162 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, attn_mask=None, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_23 = x_162.transpose(1, 2) + x_162 = None + x_163 = transpose_23.reshape(1, 197, 768) + transpose_23 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_163 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_165 = torch.nn.functional.dropout(x_164, 0.0, False, False) + x_164 = None + x_166 = x_161 + x_165 + x_161 = x_165 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_166, + (768,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_167 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_39 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_168 = torch._C._nn.gelu(x_167, approximate="none") + x_167 = None + x_169 = torch.nn.functional.dropout(x_168, 0.0, False, False) + x_168 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_166 + x_171 + x_166 = x_171 = None + layer_norm_40 = torch.nn.functional.layer_norm( + x_172, + (768,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_40 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_44 = linear_80.reshape(1, 197, 3, 12, 64) + linear_80 = None + qkv_14 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_18 = unbind_14[0] + k_18 = unbind_14[1] + v_18 = unbind_14[2] + unbind_14 = None + x_173 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_24 = x_173.transpose(1, 2) + x_173 = None + x_174 = transpose_24.reshape(1, 197, 768) + transpose_24 = None + x_175 = torch._C._nn.linear( + x_174, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_174 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = x_172 + x_176 + x_172 = x_176 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_177, + (768,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_178 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_41 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_179 = torch._C._nn.gelu(x_178, approximate="none") + x_178 = None + x_180 = torch.nn.functional.dropout(x_179, 0.0, False, False) + x_179 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_177 + x_182 + x_177 = x_182 = None + getitem_85 = x_139[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_25 = torch.nn.functional.layer_norm( + getitem_85, + (384,), + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_85 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_26 = torch._C._nn.gelu(input_25, approximate="none") + input_25 = None + input_27 = torch._C._nn.linear( + input_26, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_26 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_86 = x_183[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_28 = torch.nn.functional.layer_norm( + getitem_86, + (768,), + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_86 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_29 = torch._C._nn.gelu(input_28, approximate="none") + input_28 = None + input_30 = torch._C._nn.linear( + input_29, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_29 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_87 = x_183[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_8 = torch.cat((input_27, getitem_87), dim=1) + input_27 = getitem_87 = None + getitem_88 = tmp_8[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_44 = torch.nn.functional.layer_norm( + tmp_8, + (768,), + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_8 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_89 = layer_norm_44[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_86 = torch._C._nn.linear( + getitem_89, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_89 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_46 = linear_86.reshape(1, 1, 12, 64) + linear_86 = None + q_19 = reshape_46.permute(0, 2, 1, 3) + reshape_46 = None + linear_87 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_47 = linear_87.reshape(1, 197, 12, 64) + linear_87 = None + k_19 = reshape_47.permute(0, 2, 1, 3) + reshape_47 = None + linear_88 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_44 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_48 = linear_88.reshape(1, 197, 12, 64) + linear_88 = None + v_19 = reshape_48.permute(0, 2, 1, 3) + reshape_48 = None + transpose_25 = k_19.transpose(-2, -1) + k_19 = None + matmul_8 = q_19 @ transpose_25 + q_19 = transpose_25 = None + attn_12 = matmul_8 * 0.125 + matmul_8 = None + attn_13 = attn_12.softmax(dim=-1) + attn_12 = None + attn_14 = torch.nn.functional.dropout(attn_13, 0.0, False, False) + attn_13 = None + matmul_9 = attn_14 @ v_19 + attn_14 = v_19 = None + transpose_26 = matmul_9.transpose(1, 2) + matmul_9 = None + x_184 = transpose_26.reshape(1, 1, 768) + transpose_26 = None + x_185 = torch._C._nn.linear( + x_184, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_184 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_186 = torch.nn.functional.dropout(x_185, 0.0, False, False) + x_185 = None + x_187 = getitem_88 + x_186 + getitem_88 = x_186 = None + getitem_90 = x_187[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_187 = None + input_31 = torch.nn.functional.layer_norm( + getitem_90, + (768,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_90 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_32 = torch._C._nn.gelu(input_31, approximate="none") + input_31 = None + input_33 = torch._C._nn.linear( + input_32, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_32 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_91 = x_139[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_9 = torch.cat((input_33, getitem_91), dim=1) + input_33 = getitem_91 = None + getitem_92 = x_139[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_139 = None + tmp_10 = torch.cat((input_30, getitem_92), dim=1) + input_30 = getitem_92 = None + getitem_93 = tmp_10[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_46 = torch.nn.functional.layer_norm( + tmp_10, + (384,), + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_10 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_94 = layer_norm_46[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_91 = torch._C._nn.linear( + getitem_94, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_94 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_50 = linear_91.reshape(1, 1, 12, 32) + linear_91 = None + q_20 = reshape_50.permute(0, 2, 1, 3) + reshape_50 = None + linear_92 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_51 = linear_92.reshape(1, 401, 12, 32) + linear_92 = None + k_20 = reshape_51.permute(0, 2, 1, 3) + reshape_51 = None + linear_93 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_46 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_52 = linear_93.reshape(1, 401, 12, 32) + linear_93 = None + v_20 = reshape_52.permute(0, 2, 1, 3) + reshape_52 = None + transpose_27 = k_20.transpose(-2, -1) + k_20 = None + matmul_10 = q_20 @ transpose_27 + q_20 = transpose_27 = None + attn_15 = matmul_10 * 0.1767766952966369 + matmul_10 = None + attn_16 = attn_15.softmax(dim=-1) + attn_15 = None + attn_17 = torch.nn.functional.dropout(attn_16, 0.0, False, False) + attn_16 = None + matmul_11 = attn_17 @ v_20 + attn_17 = v_20 = None + transpose_28 = matmul_11.transpose(1, 2) + matmul_11 = None + x_188 = transpose_28.reshape(1, 1, 384) + transpose_28 = None + x_189 = torch._C._nn.linear( + x_188, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_188 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = getitem_93 + x_190 + getitem_93 = x_190 = None + getitem_95 = x_191[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_191 = None + input_34 = torch.nn.functional.layer_norm( + getitem_95, + (384,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_95 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_35 = torch._C._nn.gelu(input_34, approximate="none") + input_34 = None + input_36 = torch._C._nn.linear( + input_35, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_35 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_96 = x_183[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_183 = None + tmp_11 = torch.cat((input_36, getitem_96), dim=1) + input_36 = getitem_96 = None + x_192 = torch.nn.functional.layer_norm( + tmp_9, + (384,), + l_self_modules_norm_modules_0_parameters_weight_, + l_self_modules_norm_modules_0_parameters_bias_, + 1e-06, + ) + tmp_9 = ( + l_self_modules_norm_modules_0_parameters_weight_ + ) = l_self_modules_norm_modules_0_parameters_bias_ = None + x_193 = torch.nn.functional.layer_norm( + tmp_11, + (768,), + l_self_modules_norm_modules_1_parameters_weight_, + l_self_modules_norm_modules_1_parameters_bias_, + 1e-06, + ) + tmp_11 = ( + l_self_modules_norm_modules_1_parameters_weight_ + ) = l_self_modules_norm_modules_1_parameters_bias_ = None + x_194 = x_192[(slice(None, None, None), 0)] + x_192 = None + x_195 = x_193[(slice(None, None, None), 0)] + x_193 = None + dropout_59 = torch.nn.functional.dropout(x_194, 0.0, False, False) + x_194 = None + dropout_60 = torch.nn.functional.dropout(x_195, 0.0, False, False) + x_195 = None + linear_96 = torch._C._nn.linear( + dropout_59, + l_self_modules_head_modules_0_parameters_weight_, + l_self_modules_head_modules_0_parameters_bias_, + ) + dropout_59 = ( + l_self_modules_head_modules_0_parameters_weight_ + ) = l_self_modules_head_modules_0_parameters_bias_ = None + linear_97 = torch._C._nn.linear( + dropout_60, + l_self_modules_head_modules_1_parameters_weight_, + l_self_modules_head_modules_1_parameters_bias_, + ) + dropout_60 = ( + l_self_modules_head_modules_1_parameters_weight_ + ) = l_self_modules_head_modules_1_parameters_bias_ = None + stack = torch.stack([linear_96, linear_97], dim=0) + linear_96 = linear_97 = None + x_196 = torch.mean(stack, dim=0) + stack = None + return (x_196,) diff --git a/samples/timm/crossvit_base_240.in1k/weight_meta.py b/samples/timm/crossvit_base_240.in1k/weight_meta.py new file mode 100644 index 0000000000..7f6f7c456e --- /dev/null +++ b/samples/timm/crossvit_base_240.in1k/weight_meta.py @@ -0,0 +1,3048 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.229 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_" + shape = [384, 3, 12, 12] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.028 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.027 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_0_: + name = "L_self_parameters_cls_token_0_" + shape = [1, 1, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_0_: + name = "L_self_parameters_pos_embed_0_" + shape = [1, 401, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_" + shape = [768, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_1_: + name = "L_self_parameters_cls_token_1_" + shape = [1, 1, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_1_: + name = "L_self_parameters_pos_embed_1_" + shape = [1, 197, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [768, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [768, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [768, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [768, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [768, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [768, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_weight_: + name = "L_self_modules_norm_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_bias_: + name = "L_self_modules_norm_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_weight_: + name = "L_self_modules_norm_modules_1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_bias_: + name = "L_self_modules_norm_modules_1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_weight_: + name = "L_self_modules_head_modules_0_parameters_weight_" + shape = [1000, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_bias_: + name = "L_self_modules_head_modules_0_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_weight_: + name = "L_self_modules_head_modules_1_parameters_weight_" + shape = [1000, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_bias_: + name = "L_self_modules_head_modules_1_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/crossvit_small_240.in1k/graph_hash.txt b/samples/timm/crossvit_small_240.in1k/graph_hash.txt new file mode 100644 index 0000000000..76341e912a --- /dev/null +++ b/samples/timm/crossvit_small_240.in1k/graph_hash.txt @@ -0,0 +1 @@ +4e02fa5c93460e927ac885c21327cd13c87d12faffd9a3573df684cde66ec654 \ No newline at end of file diff --git a/samples/timm/crossvit_small_240.in1k/graph_net.json b/samples/timm/crossvit_small_240.in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/crossvit_small_240.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/crossvit_small_240.in1k/input_meta.py b/samples/timm/crossvit_small_240.in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/crossvit_small_240.in1k/input_tensor_constraints.py b/samples/timm/crossvit_small_240.in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/crossvit_small_240.in1k/model.py b/samples/timm/crossvit_small_240.in1k/model.py new file mode 100644 index 0000000000..f8c01c9aa2 --- /dev/null +++ b/samples/timm/crossvit_small_240.in1k/model.py @@ -0,0 +1,2451 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_0_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_0_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_1_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ + ) + l_self_parameters_cls_token_0_ = L_self_parameters_cls_token_0_ + l_self_parameters_pos_embed_0_ = L_self_parameters_pos_embed_0_ + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ + ) + l_self_parameters_cls_token_1_ = L_self_parameters_cls_token_1_ + l_self_parameters_pos_embed_1_ = L_self_parameters_pos_embed_1_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_norm_modules_0_parameters_weight_ = ( + L_self_modules_norm_modules_0_parameters_weight_ + ) + l_self_modules_norm_modules_0_parameters_bias_ = ( + L_self_modules_norm_modules_0_parameters_bias_ + ) + l_self_modules_norm_modules_1_parameters_weight_ = ( + L_self_modules_norm_modules_1_parameters_weight_ + ) + l_self_modules_norm_modules_1_parameters_bias_ = ( + L_self_modules_norm_modules_1_parameters_bias_ + ) + l_self_modules_head_modules_0_parameters_weight_ = ( + L_self_modules_head_modules_0_parameters_weight_ + ) + l_self_modules_head_modules_0_parameters_bias_ = ( + L_self_modules_head_modules_0_parameters_bias_ + ) + l_self_modules_head_modules_1_parameters_weight_ = ( + L_self_modules_head_modules_1_parameters_weight_ + ) + l_self_modules_head_modules_1_parameters_bias_ = ( + L_self_modules_head_modules_1_parameters_bias_ + ) + x = torch.nn.functional.interpolate( + l_x_, size=(240, 240), mode="bicubic", align_corners=False + ) + conv2d = torch.conv2d( + x, + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_, + (12, 12), + (0, 0), + (1, 1), + 1, + ) + x = ( + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ = None + flatten = conv2d.flatten(2) + conv2d = None + x_1 = flatten.transpose(1, 2) + flatten = None + cls_tokens = l_self_parameters_cls_token_0_.expand(1, -1, -1) + l_self_parameters_cls_token_0_ = None + x_ = torch.cat((cls_tokens, x_1), dim=1) + cls_tokens = x_1 = None + x__1 = x_ + l_self_parameters_pos_embed_0_ + x_ = l_self_parameters_pos_embed_0_ = None + x__2 = torch.nn.functional.dropout(x__1, 0.0, False, False) + x__1 = None + conv2d_1 = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ = None + flatten_1 = conv2d_1.flatten(2) + conv2d_1 = None + x_2 = flatten_1.transpose(1, 2) + flatten_1 = None + cls_tokens_1 = l_self_parameters_cls_token_1_.expand(1, -1, -1) + l_self_parameters_cls_token_1_ = None + x__3 = torch.cat((cls_tokens_1, x_2), dim=1) + cls_tokens_1 = x_2 = None + x__4 = x__3 + l_self_parameters_pos_embed_1_ + x__3 = l_self_parameters_pos_embed_1_ = None + x__5 = torch.nn.functional.dropout(x__4, 0.0, False, False) + x__4 = None + layer_norm = torch.nn.functional.layer_norm( + x__2, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear = torch._C._nn.linear( + layer_norm, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 401, 3, 6, 32) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_3 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_2 = x_3.transpose(1, 2) + x_3 = None + x_4 = transpose_2.reshape(1, 401, 192) + transpose_2 = None + x_5 = torch._C._nn.linear( + x_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_6 = torch.nn.functional.dropout(x_5, 0.0, False, False) + x_5 = None + x_7 = x__2 + x_6 + x__2 = x_6 = None + layer_norm_1 = torch.nn.functional.layer_norm( + x_7, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_8 = torch._C._nn.linear( + layer_norm_1, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_1 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_9 = torch._C._nn.gelu(x_8, approximate="none") + x_8 = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = torch._C._nn.linear( + x_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_7 + x_12 + x_7 = x_12 = None + layer_norm_2 = torch.nn.functional.layer_norm( + x__5, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_2, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_2 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_2 = linear_4.reshape(1, 197, 3, 6, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_14 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_3 = x_14.transpose(1, 2) + x_14 = None + x_15 = transpose_3.reshape(1, 197, 384) + transpose_3 = None + x_16 = torch._C._nn.linear( + x_15, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_15 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = x__5 + x_17 + x__5 = x_17 = None + layer_norm_3 = torch.nn.functional.layer_norm( + x_18, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_19 = torch._C._nn.linear( + layer_norm_3, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_3 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_20 = torch._C._nn.gelu(x_19, approximate="none") + x_19 = None + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = torch._C._nn.linear( + x_21, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_21 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_18 + x_23 + x_18 = x_23 = None + layer_norm_4 = torch.nn.functional.layer_norm( + x_24, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_8.reshape(1, 197, 3, 6, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_25 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_4 = x_25.transpose(1, 2) + x_25 = None + x_26 = transpose_4.reshape(1, 197, 384) + transpose_4 = None + x_27 = torch._C._nn.linear( + x_26, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_26 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_28 = torch.nn.functional.dropout(x_27, 0.0, False, False) + x_27 = None + x_29 = x_24 + x_28 + x_24 = x_28 = None + layer_norm_5 = torch.nn.functional.layer_norm( + x_29, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_30 = torch._C._nn.linear( + layer_norm_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_31 = torch._C._nn.gelu(x_30, approximate="none") + x_30 = None + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = torch._C._nn.linear( + x_32, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_32 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = x_29 + x_34 + x_29 = x_34 = None + layer_norm_6 = torch.nn.functional.layer_norm( + x_35, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_6, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_6 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_6 = linear_12.reshape(1, 197, 3, 6, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_36 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_5 = x_36.transpose(1, 2) + x_36 = None + x_37 = transpose_5.reshape(1, 197, 384) + transpose_5 = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_37 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_39 = torch.nn.functional.dropout(x_38, 0.0, False, False) + x_38 = None + x_40 = x_35 + x_39 + x_35 = x_39 = None + layer_norm_7 = torch.nn.functional.layer_norm( + x_40, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_41 = torch._C._nn.linear( + layer_norm_7, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_7 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_42 = torch._C._nn.gelu(x_41, approximate="none") + x_41 = None + x_43 = torch.nn.functional.dropout(x_42, 0.0, False, False) + x_42 = None + x_44 = torch._C._nn.linear( + x_43, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_43 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = x_40 + x_45 + x_40 = x_45 = None + layer_norm_8 = torch.nn.functional.layer_norm( + x_46, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_8, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_8 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_8 = linear_16.reshape(1, 197, 3, 6, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_47 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_6 = x_47.transpose(1, 2) + x_47 = None + x_48 = transpose_6.reshape(1, 197, 384) + transpose_6 = None + x_49 = torch._C._nn.linear( + x_48, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_48 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_50 = torch.nn.functional.dropout(x_49, 0.0, False, False) + x_49 = None + x_51 = x_46 + x_50 + x_46 = x_50 = None + layer_norm_9 = torch.nn.functional.layer_norm( + x_51, + (384,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_52 = torch._C._nn.linear( + layer_norm_9, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_9 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_53 = torch._C._nn.gelu(x_52, approximate="none") + x_52 = None + x_54 = torch.nn.functional.dropout(x_53, 0.0, False, False) + x_53 = None + x_55 = torch._C._nn.linear( + x_54, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_54 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_56 = torch.nn.functional.dropout(x_55, 0.0, False, False) + x_55 = None + x_57 = x_51 + x_56 + x_51 = x_56 = None + getitem_31 = x_13[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_1 = torch.nn.functional.layer_norm( + getitem_31, + (192,), + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_31 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_2 = torch._C._nn.gelu(input_1, approximate="none") + input_1 = None + input_3 = torch._C._nn.linear( + input_2, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_2 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_32 = x_57[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_4 = torch.nn.functional.layer_norm( + getitem_32, + (384,), + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_32 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_5 = torch._C._nn.gelu(input_4, approximate="none") + input_4 = None + input_6 = torch._C._nn.linear( + input_5, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_5 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_33 = x_57[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp = torch.cat((input_3, getitem_33), dim=1) + input_3 = getitem_33 = None + getitem_34 = tmp[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_12 = torch.nn.functional.layer_norm( + tmp, + (384,), + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_35 = layer_norm_12[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_22 = torch._C._nn.linear( + getitem_35, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_35 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_10 = linear_22.reshape(1, 1, 6, 64) + linear_22 = None + q_5 = reshape_10.permute(0, 2, 1, 3) + reshape_10 = None + linear_23 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_11 = linear_23.reshape(1, 197, 6, 64) + linear_23 = None + k_5 = reshape_11.permute(0, 2, 1, 3) + reshape_11 = None + linear_24 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_12 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_12 = linear_24.reshape(1, 197, 6, 64) + linear_24 = None + v_5 = reshape_12.permute(0, 2, 1, 3) + reshape_12 = None + transpose_7 = k_5.transpose(-2, -1) + k_5 = None + matmul = q_5 @ transpose_7 + q_5 = transpose_7 = None + attn = matmul * 0.125 + matmul = None + attn_1 = attn.softmax(dim=-1) + attn = None + attn_2 = torch.nn.functional.dropout(attn_1, 0.0, False, False) + attn_1 = None + matmul_1 = attn_2 @ v_5 + attn_2 = v_5 = None + transpose_8 = matmul_1.transpose(1, 2) + matmul_1 = None + x_58 = transpose_8.reshape(1, 1, 384) + transpose_8 = None + x_59 = torch._C._nn.linear( + x_58, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_58 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_60 = torch.nn.functional.dropout(x_59, 0.0, False, False) + x_59 = None + x_61 = getitem_34 + x_60 + getitem_34 = x_60 = None + getitem_36 = x_61[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_61 = None + input_7 = torch.nn.functional.layer_norm( + getitem_36, + (384,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_36 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_8 = torch._C._nn.gelu(input_7, approximate="none") + input_7 = None + input_9 = torch._C._nn.linear( + input_8, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_8 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_37 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_1 = torch.cat((input_9, getitem_37), dim=1) + input_9 = getitem_37 = None + getitem_38 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_13 = None + tmp_2 = torch.cat((input_6, getitem_38), dim=1) + input_6 = getitem_38 = None + getitem_39 = tmp_2[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_14 = torch.nn.functional.layer_norm( + tmp_2, + (192,), + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_2 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_40 = layer_norm_14[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_27 = torch._C._nn.linear( + getitem_40, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_40 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_14 = linear_27.reshape(1, 1, 6, 32) + linear_27 = None + q_6 = reshape_14.permute(0, 2, 1, 3) + reshape_14 = None + linear_28 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_15 = linear_28.reshape(1, 401, 6, 32) + linear_28 = None + k_6 = reshape_15.permute(0, 2, 1, 3) + reshape_15 = None + linear_29 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_14 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_16 = linear_29.reshape(1, 401, 6, 32) + linear_29 = None + v_6 = reshape_16.permute(0, 2, 1, 3) + reshape_16 = None + transpose_9 = k_6.transpose(-2, -1) + k_6 = None + matmul_2 = q_6 @ transpose_9 + q_6 = transpose_9 = None + attn_3 = matmul_2 * 0.1767766952966369 + matmul_2 = None + attn_4 = attn_3.softmax(dim=-1) + attn_3 = None + attn_5 = torch.nn.functional.dropout(attn_4, 0.0, False, False) + attn_4 = None + matmul_3 = attn_5 @ v_6 + attn_5 = v_6 = None + transpose_10 = matmul_3.transpose(1, 2) + matmul_3 = None + x_62 = transpose_10.reshape(1, 1, 192) + transpose_10 = None + x_63 = torch._C._nn.linear( + x_62, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_62 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_64 = torch.nn.functional.dropout(x_63, 0.0, False, False) + x_63 = None + x_65 = getitem_39 + x_64 + getitem_39 = x_64 = None + getitem_41 = x_65[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_65 = None + input_10 = torch.nn.functional.layer_norm( + getitem_41, + (192,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_41 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_11 = torch._C._nn.gelu(input_10, approximate="none") + input_10 = None + input_12 = torch._C._nn.linear( + input_11, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_11 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_42 = x_57[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_57 = None + tmp_3 = torch.cat((input_12, getitem_42), dim=1) + input_12 = getitem_42 = None + layer_norm_16 = torch.nn.functional.layer_norm( + tmp_1, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_32 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_16 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_18 = linear_32.reshape(1, 401, 3, 6, 32) + linear_32 = None + qkv_5 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_7 = unbind_5[0] + k_7 = unbind_5[1] + v_7 = unbind_5[2] + unbind_5 = None + x_66 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_11 = x_66.transpose(1, 2) + x_66 = None + x_67 = transpose_11.reshape(1, 401, 192) + transpose_11 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_67 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = tmp_1 + x_69 + tmp_1 = x_69 = None + layer_norm_17 = torch.nn.functional.layer_norm( + x_70, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_71 = torch._C._nn.linear( + layer_norm_17, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_17 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_72 = torch._C._nn.gelu(x_71, approximate="none") + x_71 = None + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = torch._C._nn.linear( + x_73, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_73 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_75 = torch.nn.functional.dropout(x_74, 0.0, False, False) + x_74 = None + x_76 = x_70 + x_75 + x_70 = x_75 = None + layer_norm_18 = torch.nn.functional.layer_norm( + tmp_3, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_18 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_20 = linear_36.reshape(1, 197, 3, 6, 64) + linear_36 = None + qkv_6 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_8 = unbind_6[0] + k_8 = unbind_6[1] + v_8 = unbind_6[2] + unbind_6 = None + x_77 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_12 = x_77.transpose(1, 2) + x_77 = None + x_78 = transpose_12.reshape(1, 197, 384) + transpose_12 = None + x_79 = torch._C._nn.linear( + x_78, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_78 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = tmp_3 + x_80 + tmp_3 = x_80 = None + layer_norm_19 = torch.nn.functional.layer_norm( + x_81, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_82 = torch._C._nn.linear( + layer_norm_19, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_19 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_83 = torch._C._nn.gelu(x_82, approximate="none") + x_82 = None + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = torch._C._nn.linear( + x_84, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_84 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_86 = torch.nn.functional.dropout(x_85, 0.0, False, False) + x_85 = None + x_87 = x_81 + x_86 + x_81 = x_86 = None + layer_norm_20 = torch.nn.functional.layer_norm( + x_87, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_20 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_22 = linear_40.reshape(1, 197, 3, 6, 64) + linear_40 = None + qkv_7 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_9 = unbind_7[0] + k_9 = unbind_7[1] + v_9 = unbind_7[2] + unbind_7 = None + x_88 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_13 = x_88.transpose(1, 2) + x_88 = None + x_89 = transpose_13.reshape(1, 197, 384) + transpose_13 = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_89 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = x_87 + x_91 + x_87 = x_91 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_92, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_93 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_21 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_94 = torch._C._nn.gelu(x_93, approximate="none") + x_93 = None + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = torch._C._nn.linear( + x_95, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_95 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_97 = torch.nn.functional.dropout(x_96, 0.0, False, False) + x_96 = None + x_98 = x_92 + x_97 + x_92 = x_97 = None + layer_norm_22 = torch.nn.functional.layer_norm( + x_98, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_22, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_22 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_24 = linear_44.reshape(1, 197, 3, 6, 64) + linear_44 = None + qkv_8 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_10 = unbind_8[0] + k_10 = unbind_8[1] + v_10 = unbind_8[2] + unbind_8 = None + x_99 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_14 = x_99.transpose(1, 2) + x_99 = None + x_100 = transpose_14.reshape(1, 197, 384) + transpose_14 = None + x_101 = torch._C._nn.linear( + x_100, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_100 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = x_98 + x_102 + x_98 = x_102 = None + layer_norm_23 = torch.nn.functional.layer_norm( + x_103, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_104 = torch._C._nn.linear( + layer_norm_23, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_23 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_105 = torch._C._nn.gelu(x_104, approximate="none") + x_104 = None + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch._C._nn.linear( + x_106, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_106 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_103 + x_108 + x_103 = x_108 = None + layer_norm_24 = torch.nn.functional.layer_norm( + x_109, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_24 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_26 = linear_48.reshape(1, 197, 3, 6, 64) + linear_48 = None + qkv_9 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_11 = unbind_9[0] + k_11 = unbind_9[1] + v_11 = unbind_9[2] + unbind_9 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_15 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_15.reshape(1, 197, 384) + transpose_15 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_109 + x_113 + x_109 = x_113 = None + layer_norm_25 = torch.nn.functional.layer_norm( + x_114, + (384,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_115 = torch._C._nn.linear( + layer_norm_25, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_25 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_116 = torch._C._nn.gelu(x_115, approximate="none") + x_115 = None + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = torch._C._nn.linear( + x_117, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_117 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_119 = torch.nn.functional.dropout(x_118, 0.0, False, False) + x_118 = None + x_120 = x_114 + x_119 + x_114 = x_119 = None + getitem_58 = x_76[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_13 = torch.nn.functional.layer_norm( + getitem_58, + (192,), + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_58 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_14 = torch._C._nn.gelu(input_13, approximate="none") + input_13 = None + input_15 = torch._C._nn.linear( + input_14, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_14 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_59 = x_120[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_16 = torch.nn.functional.layer_norm( + getitem_59, + (384,), + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_59 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_17 = torch._C._nn.gelu(input_16, approximate="none") + input_16 = None + input_18 = torch._C._nn.linear( + input_17, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_17 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_60 = x_120[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_4 = torch.cat((input_15, getitem_60), dim=1) + input_15 = getitem_60 = None + getitem_61 = tmp_4[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_28 = torch.nn.functional.layer_norm( + tmp_4, + (384,), + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_4 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_62 = layer_norm_28[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_54 = torch._C._nn.linear( + getitem_62, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_62 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_28 = linear_54.reshape(1, 1, 6, 64) + linear_54 = None + q_12 = reshape_28.permute(0, 2, 1, 3) + reshape_28 = None + linear_55 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_29 = linear_55.reshape(1, 197, 6, 64) + linear_55 = None + k_12 = reshape_29.permute(0, 2, 1, 3) + reshape_29 = None + linear_56 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_28 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_30 = linear_56.reshape(1, 197, 6, 64) + linear_56 = None + v_12 = reshape_30.permute(0, 2, 1, 3) + reshape_30 = None + transpose_16 = k_12.transpose(-2, -1) + k_12 = None + matmul_4 = q_12 @ transpose_16 + q_12 = transpose_16 = None + attn_6 = matmul_4 * 0.125 + matmul_4 = None + attn_7 = attn_6.softmax(dim=-1) + attn_6 = None + attn_8 = torch.nn.functional.dropout(attn_7, 0.0, False, False) + attn_7 = None + matmul_5 = attn_8 @ v_12 + attn_8 = v_12 = None + transpose_17 = matmul_5.transpose(1, 2) + matmul_5 = None + x_121 = transpose_17.reshape(1, 1, 384) + transpose_17 = None + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_121 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) + x_122 = None + x_124 = getitem_61 + x_123 + getitem_61 = x_123 = None + getitem_63 = x_124[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_124 = None + input_19 = torch.nn.functional.layer_norm( + getitem_63, + (384,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_63 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_20 = torch._C._nn.gelu(input_19, approximate="none") + input_19 = None + input_21 = torch._C._nn.linear( + input_20, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_20 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_64 = x_76[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_5 = torch.cat((input_21, getitem_64), dim=1) + input_21 = getitem_64 = None + getitem_65 = x_76[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_76 = None + tmp_6 = torch.cat((input_18, getitem_65), dim=1) + input_18 = getitem_65 = None + getitem_66 = tmp_6[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_30 = torch.nn.functional.layer_norm( + tmp_6, + (192,), + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_6 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_67 = layer_norm_30[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_59 = torch._C._nn.linear( + getitem_67, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_67 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_32 = linear_59.reshape(1, 1, 6, 32) + linear_59 = None + q_13 = reshape_32.permute(0, 2, 1, 3) + reshape_32 = None + linear_60 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_33 = linear_60.reshape(1, 401, 6, 32) + linear_60 = None + k_13 = reshape_33.permute(0, 2, 1, 3) + reshape_33 = None + linear_61 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_30 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_34 = linear_61.reshape(1, 401, 6, 32) + linear_61 = None + v_13 = reshape_34.permute(0, 2, 1, 3) + reshape_34 = None + transpose_18 = k_13.transpose(-2, -1) + k_13 = None + matmul_6 = q_13 @ transpose_18 + q_13 = transpose_18 = None + attn_9 = matmul_6 * 0.1767766952966369 + matmul_6 = None + attn_10 = attn_9.softmax(dim=-1) + attn_9 = None + attn_11 = torch.nn.functional.dropout(attn_10, 0.0, False, False) + attn_10 = None + matmul_7 = attn_11 @ v_13 + attn_11 = v_13 = None + transpose_19 = matmul_7.transpose(1, 2) + matmul_7 = None + x_125 = transpose_19.reshape(1, 1, 192) + transpose_19 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_125 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = getitem_66 + x_127 + getitem_66 = x_127 = None + getitem_68 = x_128[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_128 = None + input_22 = torch.nn.functional.layer_norm( + getitem_68, + (192,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_68 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_23 = torch._C._nn.gelu(input_22, approximate="none") + input_22 = None + input_24 = torch._C._nn.linear( + input_23, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_23 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_69 = x_120[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_120 = None + tmp_7 = torch.cat((input_24, getitem_69), dim=1) + input_24 = getitem_69 = None + layer_norm_32 = torch.nn.functional.layer_norm( + tmp_5, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_32 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_36 = linear_64.reshape(1, 401, 3, 6, 32) + linear_64 = None + qkv_10 = reshape_36.permute(2, 0, 3, 1, 4) + reshape_36 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_14 = unbind_10[0] + k_14 = unbind_10[1] + v_14 = unbind_10[2] + unbind_10 = None + x_129 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_20 = x_129.transpose(1, 2) + x_129 = None + x_130 = transpose_20.reshape(1, 401, 192) + transpose_20 = None + x_131 = torch._C._nn.linear( + x_130, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_130 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_132 = torch.nn.functional.dropout(x_131, 0.0, False, False) + x_131 = None + x_133 = tmp_5 + x_132 + tmp_5 = x_132 = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_133, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_134 = torch._C._nn.linear( + layer_norm_33, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_33 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_135 = torch._C._nn.gelu(x_134, approximate="none") + x_134 = None + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_133 + x_138 + x_133 = x_138 = None + layer_norm_34 = torch.nn.functional.layer_norm( + tmp_7, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_68 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_34 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_38 = linear_68.reshape(1, 197, 3, 6, 64) + linear_68 = None + qkv_11 = reshape_38.permute(2, 0, 3, 1, 4) + reshape_38 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_15 = unbind_11[0] + k_15 = unbind_11[1] + v_15 = unbind_11[2] + unbind_11 = None + x_140 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_21 = x_140.transpose(1, 2) + x_140 = None + x_141 = transpose_21.reshape(1, 197, 384) + transpose_21 = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_141 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_143 = torch.nn.functional.dropout(x_142, 0.0, False, False) + x_142 = None + x_144 = tmp_7 + x_143 + tmp_7 = x_143 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_144, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_145 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_35 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_146 = torch._C._nn.gelu(x_145, approximate="none") + x_145 = None + x_147 = torch.nn.functional.dropout(x_146, 0.0, False, False) + x_146 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_144 + x_149 + x_144 = x_149 = None + layer_norm_36 = torch.nn.functional.layer_norm( + x_150, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_72 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_36 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_40 = linear_72.reshape(1, 197, 3, 6, 64) + linear_72 = None + qkv_12 = reshape_40.permute(2, 0, 3, 1, 4) + reshape_40 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_16 = unbind_12[0] + k_16 = unbind_12[1] + v_16 = unbind_12[2] + unbind_12 = None + x_151 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, attn_mask=None, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_22 = x_151.transpose(1, 2) + x_151 = None + x_152 = transpose_22.reshape(1, 197, 384) + transpose_22 = None + x_153 = torch._C._nn.linear( + x_152, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_152 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_154 = torch.nn.functional.dropout(x_153, 0.0, False, False) + x_153 = None + x_155 = x_150 + x_154 + x_150 = x_154 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_155, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_156 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_37 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_157 = torch._C._nn.gelu(x_156, approximate="none") + x_156 = None + x_158 = torch.nn.functional.dropout(x_157, 0.0, False, False) + x_157 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_155 + x_160 + x_155 = x_160 = None + layer_norm_38 = torch.nn.functional.layer_norm( + x_161, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_76 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_38 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_42 = linear_76.reshape(1, 197, 3, 6, 64) + linear_76 = None + qkv_13 = reshape_42.permute(2, 0, 3, 1, 4) + reshape_42 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_17 = unbind_13[0] + k_17 = unbind_13[1] + v_17 = unbind_13[2] + unbind_13 = None + x_162 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, attn_mask=None, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_23 = x_162.transpose(1, 2) + x_162 = None + x_163 = transpose_23.reshape(1, 197, 384) + transpose_23 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_163 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_165 = torch.nn.functional.dropout(x_164, 0.0, False, False) + x_164 = None + x_166 = x_161 + x_165 + x_161 = x_165 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_166, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_167 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_39 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_168 = torch._C._nn.gelu(x_167, approximate="none") + x_167 = None + x_169 = torch.nn.functional.dropout(x_168, 0.0, False, False) + x_168 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_166 + x_171 + x_166 = x_171 = None + layer_norm_40 = torch.nn.functional.layer_norm( + x_172, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_40 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_44 = linear_80.reshape(1, 197, 3, 6, 64) + linear_80 = None + qkv_14 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_18 = unbind_14[0] + k_18 = unbind_14[1] + v_18 = unbind_14[2] + unbind_14 = None + x_173 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_24 = x_173.transpose(1, 2) + x_173 = None + x_174 = transpose_24.reshape(1, 197, 384) + transpose_24 = None + x_175 = torch._C._nn.linear( + x_174, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_174 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = x_172 + x_176 + x_172 = x_176 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_177, + (384,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_178 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_41 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_179 = torch._C._nn.gelu(x_178, approximate="none") + x_178 = None + x_180 = torch.nn.functional.dropout(x_179, 0.0, False, False) + x_179 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_177 + x_182 + x_177 = x_182 = None + getitem_85 = x_139[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_25 = torch.nn.functional.layer_norm( + getitem_85, + (192,), + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_85 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_26 = torch._C._nn.gelu(input_25, approximate="none") + input_25 = None + input_27 = torch._C._nn.linear( + input_26, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_26 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_86 = x_183[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_28 = torch.nn.functional.layer_norm( + getitem_86, + (384,), + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_86 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_29 = torch._C._nn.gelu(input_28, approximate="none") + input_28 = None + input_30 = torch._C._nn.linear( + input_29, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_29 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_87 = x_183[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_8 = torch.cat((input_27, getitem_87), dim=1) + input_27 = getitem_87 = None + getitem_88 = tmp_8[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_44 = torch.nn.functional.layer_norm( + tmp_8, + (384,), + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_8 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_89 = layer_norm_44[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_86 = torch._C._nn.linear( + getitem_89, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_89 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_46 = linear_86.reshape(1, 1, 6, 64) + linear_86 = None + q_19 = reshape_46.permute(0, 2, 1, 3) + reshape_46 = None + linear_87 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_47 = linear_87.reshape(1, 197, 6, 64) + linear_87 = None + k_19 = reshape_47.permute(0, 2, 1, 3) + reshape_47 = None + linear_88 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_44 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_48 = linear_88.reshape(1, 197, 6, 64) + linear_88 = None + v_19 = reshape_48.permute(0, 2, 1, 3) + reshape_48 = None + transpose_25 = k_19.transpose(-2, -1) + k_19 = None + matmul_8 = q_19 @ transpose_25 + q_19 = transpose_25 = None + attn_12 = matmul_8 * 0.125 + matmul_8 = None + attn_13 = attn_12.softmax(dim=-1) + attn_12 = None + attn_14 = torch.nn.functional.dropout(attn_13, 0.0, False, False) + attn_13 = None + matmul_9 = attn_14 @ v_19 + attn_14 = v_19 = None + transpose_26 = matmul_9.transpose(1, 2) + matmul_9 = None + x_184 = transpose_26.reshape(1, 1, 384) + transpose_26 = None + x_185 = torch._C._nn.linear( + x_184, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_184 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_186 = torch.nn.functional.dropout(x_185, 0.0, False, False) + x_185 = None + x_187 = getitem_88 + x_186 + getitem_88 = x_186 = None + getitem_90 = x_187[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_187 = None + input_31 = torch.nn.functional.layer_norm( + getitem_90, + (384,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_90 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_32 = torch._C._nn.gelu(input_31, approximate="none") + input_31 = None + input_33 = torch._C._nn.linear( + input_32, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_32 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_91 = x_139[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_9 = torch.cat((input_33, getitem_91), dim=1) + input_33 = getitem_91 = None + getitem_92 = x_139[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_139 = None + tmp_10 = torch.cat((input_30, getitem_92), dim=1) + input_30 = getitem_92 = None + getitem_93 = tmp_10[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_46 = torch.nn.functional.layer_norm( + tmp_10, + (192,), + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_10 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_94 = layer_norm_46[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_91 = torch._C._nn.linear( + getitem_94, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_94 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_50 = linear_91.reshape(1, 1, 6, 32) + linear_91 = None + q_20 = reshape_50.permute(0, 2, 1, 3) + reshape_50 = None + linear_92 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_51 = linear_92.reshape(1, 401, 6, 32) + linear_92 = None + k_20 = reshape_51.permute(0, 2, 1, 3) + reshape_51 = None + linear_93 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_46 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_52 = linear_93.reshape(1, 401, 6, 32) + linear_93 = None + v_20 = reshape_52.permute(0, 2, 1, 3) + reshape_52 = None + transpose_27 = k_20.transpose(-2, -1) + k_20 = None + matmul_10 = q_20 @ transpose_27 + q_20 = transpose_27 = None + attn_15 = matmul_10 * 0.1767766952966369 + matmul_10 = None + attn_16 = attn_15.softmax(dim=-1) + attn_15 = None + attn_17 = torch.nn.functional.dropout(attn_16, 0.0, False, False) + attn_16 = None + matmul_11 = attn_17 @ v_20 + attn_17 = v_20 = None + transpose_28 = matmul_11.transpose(1, 2) + matmul_11 = None + x_188 = transpose_28.reshape(1, 1, 192) + transpose_28 = None + x_189 = torch._C._nn.linear( + x_188, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_188 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = getitem_93 + x_190 + getitem_93 = x_190 = None + getitem_95 = x_191[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_191 = None + input_34 = torch.nn.functional.layer_norm( + getitem_95, + (192,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_95 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_35 = torch._C._nn.gelu(input_34, approximate="none") + input_34 = None + input_36 = torch._C._nn.linear( + input_35, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_35 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_96 = x_183[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_183 = None + tmp_11 = torch.cat((input_36, getitem_96), dim=1) + input_36 = getitem_96 = None + x_192 = torch.nn.functional.layer_norm( + tmp_9, + (192,), + l_self_modules_norm_modules_0_parameters_weight_, + l_self_modules_norm_modules_0_parameters_bias_, + 1e-06, + ) + tmp_9 = ( + l_self_modules_norm_modules_0_parameters_weight_ + ) = l_self_modules_norm_modules_0_parameters_bias_ = None + x_193 = torch.nn.functional.layer_norm( + tmp_11, + (384,), + l_self_modules_norm_modules_1_parameters_weight_, + l_self_modules_norm_modules_1_parameters_bias_, + 1e-06, + ) + tmp_11 = ( + l_self_modules_norm_modules_1_parameters_weight_ + ) = l_self_modules_norm_modules_1_parameters_bias_ = None + x_194 = x_192[(slice(None, None, None), 0)] + x_192 = None + x_195 = x_193[(slice(None, None, None), 0)] + x_193 = None + dropout_59 = torch.nn.functional.dropout(x_194, 0.0, False, False) + x_194 = None + dropout_60 = torch.nn.functional.dropout(x_195, 0.0, False, False) + x_195 = None + linear_96 = torch._C._nn.linear( + dropout_59, + l_self_modules_head_modules_0_parameters_weight_, + l_self_modules_head_modules_0_parameters_bias_, + ) + dropout_59 = ( + l_self_modules_head_modules_0_parameters_weight_ + ) = l_self_modules_head_modules_0_parameters_bias_ = None + linear_97 = torch._C._nn.linear( + dropout_60, + l_self_modules_head_modules_1_parameters_weight_, + l_self_modules_head_modules_1_parameters_bias_, + ) + dropout_60 = ( + l_self_modules_head_modules_1_parameters_weight_ + ) = l_self_modules_head_modules_1_parameters_bias_ = None + stack = torch.stack([linear_96, linear_97], dim=0) + linear_96 = linear_97 = None + x_196 = torch.mean(stack, dim=0) + stack = None + return (x_196,) diff --git a/samples/timm/crossvit_small_240.in1k/weight_meta.py b/samples/timm/crossvit_small_240.in1k/weight_meta.py new file mode 100644 index 0000000000..5fc073d134 --- /dev/null +++ b/samples/timm/crossvit_small_240.in1k/weight_meta.py @@ -0,0 +1,3048 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.228 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_" + shape = [192, 3, 12, 12] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.028 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.027 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_0_: + name = "L_self_parameters_cls_token_0_" + shape = [1, 1, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_0_: + name = "L_self_parameters_pos_embed_0_" + shape = [1, 401, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_" + shape = [384, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_1_: + name = "L_self_parameters_cls_token_1_" + shape = [1, 1, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_1_: + name = "L_self_parameters_pos_embed_1_" + shape = [1, 197, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [384, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_weight_: + name = "L_self_modules_norm_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_bias_: + name = "L_self_modules_norm_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_weight_: + name = "L_self_modules_norm_modules_1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_bias_: + name = "L_self_modules_norm_modules_1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_weight_: + name = "L_self_modules_head_modules_0_parameters_weight_" + shape = [1000, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_bias_: + name = "L_self_modules_head_modules_0_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_weight_: + name = "L_self_modules_head_modules_1_parameters_weight_" + shape = [1000, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_bias_: + name = "L_self_modules_head_modules_1_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/crossvit_tiny_240.in1k/graph_hash.txt b/samples/timm/crossvit_tiny_240.in1k/graph_hash.txt new file mode 100644 index 0000000000..b1530ba6b4 --- /dev/null +++ b/samples/timm/crossvit_tiny_240.in1k/graph_hash.txt @@ -0,0 +1 @@ +63cd9da64f85a0be1d1bcf7b2ed6260cfecbf0e6393ac01b3b155379bce98355 \ No newline at end of file diff --git a/samples/timm/crossvit_tiny_240.in1k/graph_net.json b/samples/timm/crossvit_tiny_240.in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/crossvit_tiny_240.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/crossvit_tiny_240.in1k/input_meta.py b/samples/timm/crossvit_tiny_240.in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/crossvit_tiny_240.in1k/input_tensor_constraints.py b/samples/timm/crossvit_tiny_240.in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/crossvit_tiny_240.in1k/model.py b/samples/timm/crossvit_tiny_240.in1k/model.py new file mode 100644 index 0000000000..131b835971 --- /dev/null +++ b/samples/timm/crossvit_tiny_240.in1k/model.py @@ -0,0 +1,2451 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_0_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_0_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_1_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_1_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ + ) + l_self_parameters_cls_token_0_ = L_self_parameters_cls_token_0_ + l_self_parameters_pos_embed_0_ = L_self_parameters_pos_embed_0_ + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ + ) + l_self_parameters_cls_token_1_ = L_self_parameters_cls_token_1_ + l_self_parameters_pos_embed_1_ = L_self_parameters_pos_embed_1_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ + l_self_modules_norm_modules_0_parameters_weight_ = ( + L_self_modules_norm_modules_0_parameters_weight_ + ) + l_self_modules_norm_modules_0_parameters_bias_ = ( + L_self_modules_norm_modules_0_parameters_bias_ + ) + l_self_modules_norm_modules_1_parameters_weight_ = ( + L_self_modules_norm_modules_1_parameters_weight_ + ) + l_self_modules_norm_modules_1_parameters_bias_ = ( + L_self_modules_norm_modules_1_parameters_bias_ + ) + l_self_modules_head_modules_0_parameters_weight_ = ( + L_self_modules_head_modules_0_parameters_weight_ + ) + l_self_modules_head_modules_0_parameters_bias_ = ( + L_self_modules_head_modules_0_parameters_bias_ + ) + l_self_modules_head_modules_1_parameters_weight_ = ( + L_self_modules_head_modules_1_parameters_weight_ + ) + l_self_modules_head_modules_1_parameters_bias_ = ( + L_self_modules_head_modules_1_parameters_bias_ + ) + x = torch.nn.functional.interpolate( + l_x_, size=(240, 240), mode="bicubic", align_corners=False + ) + conv2d = torch.conv2d( + x, + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_, + (12, 12), + (0, 0), + (1, 1), + 1, + ) + x = ( + l_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_ = None + flatten = conv2d.flatten(2) + conv2d = None + x_1 = flatten.transpose(1, 2) + flatten = None + cls_tokens = l_self_parameters_cls_token_0_.expand(1, -1, -1) + l_self_parameters_cls_token_0_ = None + x_ = torch.cat((cls_tokens, x_1), dim=1) + cls_tokens = x_1 = None + x__1 = x_ + l_self_parameters_pos_embed_0_ + x_ = l_self_parameters_pos_embed_0_ = None + x__2 = torch.nn.functional.dropout(x__1, 0.0, False, False) + x__1 = None + conv2d_1 = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_ = None + flatten_1 = conv2d_1.flatten(2) + conv2d_1 = None + x_2 = flatten_1.transpose(1, 2) + flatten_1 = None + cls_tokens_1 = l_self_parameters_cls_token_1_.expand(1, -1, -1) + l_self_parameters_cls_token_1_ = None + x__3 = torch.cat((cls_tokens_1, x_2), dim=1) + cls_tokens_1 = x_2 = None + x__4 = x__3 + l_self_parameters_pos_embed_1_ + x__3 = l_self_parameters_pos_embed_1_ = None + x__5 = torch.nn.functional.dropout(x__4, 0.0, False, False) + x__4 = None + layer_norm = torch.nn.functional.layer_norm( + x__2, + (96,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear = torch._C._nn.linear( + layer_norm, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape = linear.reshape(1, 401, 3, 3, 32) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_3 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_2 = x_3.transpose(1, 2) + x_3 = None + x_4 = transpose_2.reshape(1, 401, 96) + transpose_2 = None + x_5 = torch._C._nn.linear( + x_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_6 = torch.nn.functional.dropout(x_5, 0.0, False, False) + x_5 = None + x_7 = x__2 + x_6 + x__2 = x_6 = None + layer_norm_1 = torch.nn.functional.layer_norm( + x_7, + (96,), + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_8 = torch._C._nn.linear( + layer_norm_1, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_1 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_9 = torch._C._nn.gelu(x_8, approximate="none") + x_8 = None + x_10 = torch.nn.functional.dropout(x_9, 0.0, False, False) + x_9 = None + x_11 = torch._C._nn.linear( + x_10, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_10 = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_12 = torch.nn.functional.dropout(x_11, 0.0, False, False) + x_11 = None + x_13 = x_7 + x_12 + x_7 = x_12 = None + layer_norm_2 = torch.nn.functional.layer_norm( + x__5, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_4 = torch._C._nn.linear( + layer_norm_2, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_2 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_2 = linear_4.reshape(1, 197, 3, 3, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_14 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_3 = x_14.transpose(1, 2) + x_14 = None + x_15 = transpose_3.reshape(1, 197, 192) + transpose_3 = None + x_16 = torch._C._nn.linear( + x_15, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_15 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_17 = torch.nn.functional.dropout(x_16, 0.0, False, False) + x_16 = None + x_18 = x__5 + x_17 + x__5 = x_17 = None + layer_norm_3 = torch.nn.functional.layer_norm( + x_18, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_19 = torch._C._nn.linear( + layer_norm_3, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_3 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_20 = torch._C._nn.gelu(x_19, approximate="none") + x_19 = None + x_21 = torch.nn.functional.dropout(x_20, 0.0, False, False) + x_20 = None + x_22 = torch._C._nn.linear( + x_21, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_21 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_23 = torch.nn.functional.dropout(x_22, 0.0, False, False) + x_22 = None + x_24 = x_18 + x_23 + x_18 = x_23 = None + layer_norm_4 = torch.nn.functional.layer_norm( + x_24, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_8 = torch._C._nn.linear( + layer_norm_4, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_4 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_4 = linear_8.reshape(1, 197, 3, 3, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_25 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_4 = x_25.transpose(1, 2) + x_25 = None + x_26 = transpose_4.reshape(1, 197, 192) + transpose_4 = None + x_27 = torch._C._nn.linear( + x_26, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_26 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_28 = torch.nn.functional.dropout(x_27, 0.0, False, False) + x_27 = None + x_29 = x_24 + x_28 + x_24 = x_28 = None + layer_norm_5 = torch.nn.functional.layer_norm( + x_29, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_30 = torch._C._nn.linear( + layer_norm_5, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_5 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_31 = torch._C._nn.gelu(x_30, approximate="none") + x_30 = None + x_32 = torch.nn.functional.dropout(x_31, 0.0, False, False) + x_31 = None + x_33 = torch._C._nn.linear( + x_32, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_32 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_34 = torch.nn.functional.dropout(x_33, 0.0, False, False) + x_33 = None + x_35 = x_29 + x_34 + x_29 = x_34 = None + layer_norm_6 = torch.nn.functional.layer_norm( + x_35, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_12 = torch._C._nn.linear( + layer_norm_6, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_6 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_6 = linear_12.reshape(1, 197, 3, 3, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_36 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_5 = x_36.transpose(1, 2) + x_36 = None + x_37 = transpose_5.reshape(1, 197, 192) + transpose_5 = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_37 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_39 = torch.nn.functional.dropout(x_38, 0.0, False, False) + x_38 = None + x_40 = x_35 + x_39 + x_35 = x_39 = None + layer_norm_7 = torch.nn.functional.layer_norm( + x_40, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_41 = torch._C._nn.linear( + layer_norm_7, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_7 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_42 = torch._C._nn.gelu(x_41, approximate="none") + x_41 = None + x_43 = torch.nn.functional.dropout(x_42, 0.0, False, False) + x_42 = None + x_44 = torch._C._nn.linear( + x_43, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_43 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_45 = torch.nn.functional.dropout(x_44, 0.0, False, False) + x_44 = None + x_46 = x_40 + x_45 + x_40 = x_45 = None + layer_norm_8 = torch.nn.functional.layer_norm( + x_46, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_16 = torch._C._nn.linear( + layer_norm_8, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_8 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_8 = linear_16.reshape(1, 197, 3, 3, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_47 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_6 = x_47.transpose(1, 2) + x_47 = None + x_48 = transpose_6.reshape(1, 197, 192) + transpose_6 = None + x_49 = torch._C._nn.linear( + x_48, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_48 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_50 = torch.nn.functional.dropout(x_49, 0.0, False, False) + x_49 = None + x_51 = x_46 + x_50 + x_46 = x_50 = None + layer_norm_9 = torch.nn.functional.layer_norm( + x_51, + (192,), + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_52 = torch._C._nn.linear( + layer_norm_9, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_9 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_53 = torch._C._nn.gelu(x_52, approximate="none") + x_52 = None + x_54 = torch.nn.functional.dropout(x_53, 0.0, False, False) + x_53 = None + x_55 = torch._C._nn.linear( + x_54, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_54 = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_56 = torch.nn.functional.dropout(x_55, 0.0, False, False) + x_55 = None + x_57 = x_51 + x_56 + x_51 = x_56 = None + getitem_31 = x_13[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_1 = torch.nn.functional.layer_norm( + getitem_31, + (96,), + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_31 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_2 = torch._C._nn.gelu(input_1, approximate="none") + input_1 = None + input_3 = torch._C._nn.linear( + input_2, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_2 = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_32 = x_57[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_4 = torch.nn.functional.layer_norm( + getitem_32, + (192,), + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_32 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_5 = torch._C._nn.gelu(input_4, approximate="none") + input_4 = None + input_6 = torch._C._nn.linear( + input_5, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_5 = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_33 = x_57[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp = torch.cat((input_3, getitem_33), dim=1) + input_3 = getitem_33 = None + getitem_34 = tmp[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_12 = torch.nn.functional.layer_norm( + tmp, + (192,), + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_35 = layer_norm_12[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_22 = torch._C._nn.linear( + getitem_35, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_35 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_10 = linear_22.reshape(1, 1, 3, 64) + linear_22 = None + q_5 = reshape_10.permute(0, 2, 1, 3) + reshape_10 = None + linear_23 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_11 = linear_23.reshape(1, 197, 3, 64) + linear_23 = None + k_5 = reshape_11.permute(0, 2, 1, 3) + reshape_11 = None + linear_24 = torch._C._nn.linear( + layer_norm_12, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_12 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_12 = linear_24.reshape(1, 197, 3, 64) + linear_24 = None + v_5 = reshape_12.permute(0, 2, 1, 3) + reshape_12 = None + transpose_7 = k_5.transpose(-2, -1) + k_5 = None + matmul = q_5 @ transpose_7 + q_5 = transpose_7 = None + attn = matmul * 0.125 + matmul = None + attn_1 = attn.softmax(dim=-1) + attn = None + attn_2 = torch.nn.functional.dropout(attn_1, 0.0, False, False) + attn_1 = None + matmul_1 = attn_2 @ v_5 + attn_2 = v_5 = None + transpose_8 = matmul_1.transpose(1, 2) + matmul_1 = None + x_58 = transpose_8.reshape(1, 1, 192) + transpose_8 = None + x_59 = torch._C._nn.linear( + x_58, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_58 = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_60 = torch.nn.functional.dropout(x_59, 0.0, False, False) + x_59 = None + x_61 = getitem_34 + x_60 + getitem_34 = x_60 = None + getitem_36 = x_61[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_61 = None + input_7 = torch.nn.functional.layer_norm( + getitem_36, + (192,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_36 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_8 = torch._C._nn.gelu(input_7, approximate="none") + input_7 = None + input_9 = torch._C._nn.linear( + input_8, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_8 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_37 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_1 = torch.cat((input_9, getitem_37), dim=1) + input_9 = getitem_37 = None + getitem_38 = x_13[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_13 = None + tmp_2 = torch.cat((input_6, getitem_38), dim=1) + input_6 = getitem_38 = None + getitem_39 = tmp_2[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_14 = torch.nn.functional.layer_norm( + tmp_2, + (96,), + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_2 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_40 = layer_norm_14[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_27 = torch._C._nn.linear( + getitem_40, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_40 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_14 = linear_27.reshape(1, 1, 3, 32) + linear_27 = None + q_6 = reshape_14.permute(0, 2, 1, 3) + reshape_14 = None + linear_28 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_15 = linear_28.reshape(1, 401, 3, 32) + linear_28 = None + k_6 = reshape_15.permute(0, 2, 1, 3) + reshape_15 = None + linear_29 = torch._C._nn.linear( + layer_norm_14, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_14 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_16 = linear_29.reshape(1, 401, 3, 32) + linear_29 = None + v_6 = reshape_16.permute(0, 2, 1, 3) + reshape_16 = None + transpose_9 = k_6.transpose(-2, -1) + k_6 = None + matmul_2 = q_6 @ transpose_9 + q_6 = transpose_9 = None + attn_3 = matmul_2 * 0.1767766952966369 + matmul_2 = None + attn_4 = attn_3.softmax(dim=-1) + attn_3 = None + attn_5 = torch.nn.functional.dropout(attn_4, 0.0, False, False) + attn_4 = None + matmul_3 = attn_5 @ v_6 + attn_5 = v_6 = None + transpose_10 = matmul_3.transpose(1, 2) + matmul_3 = None + x_62 = transpose_10.reshape(1, 1, 96) + transpose_10 = None + x_63 = torch._C._nn.linear( + x_62, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_62 = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_64 = torch.nn.functional.dropout(x_63, 0.0, False, False) + x_63 = None + x_65 = getitem_39 + x_64 + getitem_39 = x_64 = None + getitem_41 = x_65[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_65 = None + input_10 = torch.nn.functional.layer_norm( + getitem_41, + (96,), + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_41 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_11 = torch._C._nn.gelu(input_10, approximate="none") + input_10 = None + input_12 = torch._C._nn.linear( + input_11, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_11 = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_42 = x_57[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_57 = None + tmp_3 = torch.cat((input_12, getitem_42), dim=1) + input_12 = getitem_42 = None + layer_norm_16 = torch.nn.functional.layer_norm( + tmp_1, + (96,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_32 = torch._C._nn.linear( + layer_norm_16, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_16 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_18 = linear_32.reshape(1, 401, 3, 3, 32) + linear_32 = None + qkv_5 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_7 = unbind_5[0] + k_7 = unbind_5[1] + v_7 = unbind_5[2] + unbind_5 = None + x_66 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_11 = x_66.transpose(1, 2) + x_66 = None + x_67 = transpose_11.reshape(1, 401, 96) + transpose_11 = None + x_68 = torch._C._nn.linear( + x_67, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_67 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_69 = torch.nn.functional.dropout(x_68, 0.0, False, False) + x_68 = None + x_70 = tmp_1 + x_69 + tmp_1 = x_69 = None + layer_norm_17 = torch.nn.functional.layer_norm( + x_70, + (96,), + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_71 = torch._C._nn.linear( + layer_norm_17, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_17 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_72 = torch._C._nn.gelu(x_71, approximate="none") + x_71 = None + x_73 = torch.nn.functional.dropout(x_72, 0.0, False, False) + x_72 = None + x_74 = torch._C._nn.linear( + x_73, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_73 = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_75 = torch.nn.functional.dropout(x_74, 0.0, False, False) + x_74 = None + x_76 = x_70 + x_75 + x_70 = x_75 = None + layer_norm_18 = torch.nn.functional.layer_norm( + tmp_3, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_36 = torch._C._nn.linear( + layer_norm_18, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_18 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_20 = linear_36.reshape(1, 197, 3, 3, 64) + linear_36 = None + qkv_6 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_8 = unbind_6[0] + k_8 = unbind_6[1] + v_8 = unbind_6[2] + unbind_6 = None + x_77 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_12 = x_77.transpose(1, 2) + x_77 = None + x_78 = transpose_12.reshape(1, 197, 192) + transpose_12 = None + x_79 = torch._C._nn.linear( + x_78, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_78 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_80 = torch.nn.functional.dropout(x_79, 0.0, False, False) + x_79 = None + x_81 = tmp_3 + x_80 + tmp_3 = x_80 = None + layer_norm_19 = torch.nn.functional.layer_norm( + x_81, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_82 = torch._C._nn.linear( + layer_norm_19, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_19 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_83 = torch._C._nn.gelu(x_82, approximate="none") + x_82 = None + x_84 = torch.nn.functional.dropout(x_83, 0.0, False, False) + x_83 = None + x_85 = torch._C._nn.linear( + x_84, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_84 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_86 = torch.nn.functional.dropout(x_85, 0.0, False, False) + x_85 = None + x_87 = x_81 + x_86 + x_81 = x_86 = None + layer_norm_20 = torch.nn.functional.layer_norm( + x_87, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_40 = torch._C._nn.linear( + layer_norm_20, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_20 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_22 = linear_40.reshape(1, 197, 3, 3, 64) + linear_40 = None + qkv_7 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_9 = unbind_7[0] + k_9 = unbind_7[1] + v_9 = unbind_7[2] + unbind_7 = None + x_88 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_13 = x_88.transpose(1, 2) + x_88 = None + x_89 = transpose_13.reshape(1, 197, 192) + transpose_13 = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_89 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_91 = torch.nn.functional.dropout(x_90, 0.0, False, False) + x_90 = None + x_92 = x_87 + x_91 + x_87 = x_91 = None + layer_norm_21 = torch.nn.functional.layer_norm( + x_92, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_93 = torch._C._nn.linear( + layer_norm_21, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_21 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_94 = torch._C._nn.gelu(x_93, approximate="none") + x_93 = None + x_95 = torch.nn.functional.dropout(x_94, 0.0, False, False) + x_94 = None + x_96 = torch._C._nn.linear( + x_95, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_95 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_97 = torch.nn.functional.dropout(x_96, 0.0, False, False) + x_96 = None + x_98 = x_92 + x_97 + x_92 = x_97 = None + layer_norm_22 = torch.nn.functional.layer_norm( + x_98, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_44 = torch._C._nn.linear( + layer_norm_22, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_22 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_24 = linear_44.reshape(1, 197, 3, 3, 64) + linear_44 = None + qkv_8 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_10 = unbind_8[0] + k_10 = unbind_8[1] + v_10 = unbind_8[2] + unbind_8 = None + x_99 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_14 = x_99.transpose(1, 2) + x_99 = None + x_100 = transpose_14.reshape(1, 197, 192) + transpose_14 = None + x_101 = torch._C._nn.linear( + x_100, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_100 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_102 = torch.nn.functional.dropout(x_101, 0.0, False, False) + x_101 = None + x_103 = x_98 + x_102 + x_98 = x_102 = None + layer_norm_23 = torch.nn.functional.layer_norm( + x_103, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_104 = torch._C._nn.linear( + layer_norm_23, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_23 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_105 = torch._C._nn.gelu(x_104, approximate="none") + x_104 = None + x_106 = torch.nn.functional.dropout(x_105, 0.0, False, False) + x_105 = None + x_107 = torch._C._nn.linear( + x_106, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_106 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_108 = torch.nn.functional.dropout(x_107, 0.0, False, False) + x_107 = None + x_109 = x_103 + x_108 + x_103 = x_108 = None + layer_norm_24 = torch.nn.functional.layer_norm( + x_109, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_48 = torch._C._nn.linear( + layer_norm_24, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_24 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_26 = linear_48.reshape(1, 197, 3, 3, 64) + linear_48 = None + qkv_9 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_11 = unbind_9[0] + k_11 = unbind_9[1] + v_11 = unbind_9[2] + unbind_9 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_15 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_15.reshape(1, 197, 192) + transpose_15 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_109 + x_113 + x_109 = x_113 = None + layer_norm_25 = torch.nn.functional.layer_norm( + x_114, + (192,), + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_115 = torch._C._nn.linear( + layer_norm_25, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_25 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_116 = torch._C._nn.gelu(x_115, approximate="none") + x_115 = None + x_117 = torch.nn.functional.dropout(x_116, 0.0, False, False) + x_116 = None + x_118 = torch._C._nn.linear( + x_117, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_117 = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_119 = torch.nn.functional.dropout(x_118, 0.0, False, False) + x_118 = None + x_120 = x_114 + x_119 + x_114 = x_119 = None + getitem_58 = x_76[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_13 = torch.nn.functional.layer_norm( + getitem_58, + (96,), + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_58 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_14 = torch._C._nn.gelu(input_13, approximate="none") + input_13 = None + input_15 = torch._C._nn.linear( + input_14, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_14 = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_59 = x_120[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_16 = torch.nn.functional.layer_norm( + getitem_59, + (192,), + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_59 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_17 = torch._C._nn.gelu(input_16, approximate="none") + input_16 = None + input_18 = torch._C._nn.linear( + input_17, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_17 = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_60 = x_120[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_4 = torch.cat((input_15, getitem_60), dim=1) + input_15 = getitem_60 = None + getitem_61 = tmp_4[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_28 = torch.nn.functional.layer_norm( + tmp_4, + (192,), + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_4 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_62 = layer_norm_28[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_54 = torch._C._nn.linear( + getitem_62, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_62 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_28 = linear_54.reshape(1, 1, 3, 64) + linear_54 = None + q_12 = reshape_28.permute(0, 2, 1, 3) + reshape_28 = None + linear_55 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_29 = linear_55.reshape(1, 197, 3, 64) + linear_55 = None + k_12 = reshape_29.permute(0, 2, 1, 3) + reshape_29 = None + linear_56 = torch._C._nn.linear( + layer_norm_28, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_28 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_30 = linear_56.reshape(1, 197, 3, 64) + linear_56 = None + v_12 = reshape_30.permute(0, 2, 1, 3) + reshape_30 = None + transpose_16 = k_12.transpose(-2, -1) + k_12 = None + matmul_4 = q_12 @ transpose_16 + q_12 = transpose_16 = None + attn_6 = matmul_4 * 0.125 + matmul_4 = None + attn_7 = attn_6.softmax(dim=-1) + attn_6 = None + attn_8 = torch.nn.functional.dropout(attn_7, 0.0, False, False) + attn_7 = None + matmul_5 = attn_8 @ v_12 + attn_8 = v_12 = None + transpose_17 = matmul_5.transpose(1, 2) + matmul_5 = None + x_121 = transpose_17.reshape(1, 1, 192) + transpose_17 = None + x_122 = torch._C._nn.linear( + x_121, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_121 = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_123 = torch.nn.functional.dropout(x_122, 0.0, False, False) + x_122 = None + x_124 = getitem_61 + x_123 + getitem_61 = x_123 = None + getitem_63 = x_124[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_124 = None + input_19 = torch.nn.functional.layer_norm( + getitem_63, + (192,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_63 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_20 = torch._C._nn.gelu(input_19, approximate="none") + input_19 = None + input_21 = torch._C._nn.linear( + input_20, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_20 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_64 = x_76[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_5 = torch.cat((input_21, getitem_64), dim=1) + input_21 = getitem_64 = None + getitem_65 = x_76[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_76 = None + tmp_6 = torch.cat((input_18, getitem_65), dim=1) + input_18 = getitem_65 = None + getitem_66 = tmp_6[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_30 = torch.nn.functional.layer_norm( + tmp_6, + (96,), + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_6 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_67 = layer_norm_30[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_59 = torch._C._nn.linear( + getitem_67, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_67 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_32 = linear_59.reshape(1, 1, 3, 32) + linear_59 = None + q_13 = reshape_32.permute(0, 2, 1, 3) + reshape_32 = None + linear_60 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_33 = linear_60.reshape(1, 401, 3, 32) + linear_60 = None + k_13 = reshape_33.permute(0, 2, 1, 3) + reshape_33 = None + linear_61 = torch._C._nn.linear( + layer_norm_30, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_30 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_34 = linear_61.reshape(1, 401, 3, 32) + linear_61 = None + v_13 = reshape_34.permute(0, 2, 1, 3) + reshape_34 = None + transpose_18 = k_13.transpose(-2, -1) + k_13 = None + matmul_6 = q_13 @ transpose_18 + q_13 = transpose_18 = None + attn_9 = matmul_6 * 0.1767766952966369 + matmul_6 = None + attn_10 = attn_9.softmax(dim=-1) + attn_9 = None + attn_11 = torch.nn.functional.dropout(attn_10, 0.0, False, False) + attn_10 = None + matmul_7 = attn_11 @ v_13 + attn_11 = v_13 = None + transpose_19 = matmul_7.transpose(1, 2) + matmul_7 = None + x_125 = transpose_19.reshape(1, 1, 96) + transpose_19 = None + x_126 = torch._C._nn.linear( + x_125, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_125 = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_127 = torch.nn.functional.dropout(x_126, 0.0, False, False) + x_126 = None + x_128 = getitem_66 + x_127 + getitem_66 = x_127 = None + getitem_68 = x_128[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_128 = None + input_22 = torch.nn.functional.layer_norm( + getitem_68, + (96,), + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_68 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_23 = torch._C._nn.gelu(input_22, approximate="none") + input_22 = None + input_24 = torch._C._nn.linear( + input_23, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_23 = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_69 = x_120[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_120 = None + tmp_7 = torch.cat((input_24, getitem_69), dim=1) + input_24 = getitem_69 = None + layer_norm_32 = torch.nn.functional.layer_norm( + tmp_5, + (96,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_ = (None) + linear_64 = torch._C._nn.linear( + layer_norm_32, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_32 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_36 = linear_64.reshape(1, 401, 3, 3, 32) + linear_64 = None + qkv_10 = reshape_36.permute(2, 0, 3, 1, 4) + reshape_36 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_14 = unbind_10[0] + k_14 = unbind_10[1] + v_14 = unbind_10[2] + unbind_10 = None + x_129 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_20 = x_129.transpose(1, 2) + x_129 = None + x_130 = transpose_20.reshape(1, 401, 96) + transpose_20 = None + x_131 = torch._C._nn.linear( + x_130, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_130 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_132 = torch.nn.functional.dropout(x_131, 0.0, False, False) + x_131 = None + x_133 = tmp_5 + x_132 + tmp_5 = x_132 = None + layer_norm_33 = torch.nn.functional.layer_norm( + x_133, + (96,), + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_ = (None) + x_134 = torch._C._nn.linear( + layer_norm_33, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_33 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_135 = torch._C._nn.gelu(x_134, approximate="none") + x_134 = None + x_136 = torch.nn.functional.dropout(x_135, 0.0, False, False) + x_135 = None + x_137 = torch._C._nn.linear( + x_136, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_136 = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_138 = torch.nn.functional.dropout(x_137, 0.0, False, False) + x_137 = None + x_139 = x_133 + x_138 + x_133 = x_138 = None + layer_norm_34 = torch.nn.functional.layer_norm( + tmp_7, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_ = (None) + linear_68 = torch._C._nn.linear( + layer_norm_34, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_34 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_38 = linear_68.reshape(1, 197, 3, 3, 64) + linear_68 = None + qkv_11 = reshape_38.permute(2, 0, 3, 1, 4) + reshape_38 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_15 = unbind_11[0] + k_15 = unbind_11[1] + v_15 = unbind_11[2] + unbind_11 = None + x_140 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_21 = x_140.transpose(1, 2) + x_140 = None + x_141 = transpose_21.reshape(1, 197, 192) + transpose_21 = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_141 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_143 = torch.nn.functional.dropout(x_142, 0.0, False, False) + x_142 = None + x_144 = tmp_7 + x_143 + tmp_7 = x_143 = None + layer_norm_35 = torch.nn.functional.layer_norm( + x_144, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_ = (None) + x_145 = torch._C._nn.linear( + layer_norm_35, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_35 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_146 = torch._C._nn.gelu(x_145, approximate="none") + x_145 = None + x_147 = torch.nn.functional.dropout(x_146, 0.0, False, False) + x_146 = None + x_148 = torch._C._nn.linear( + x_147, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_147 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_149 = torch.nn.functional.dropout(x_148, 0.0, False, False) + x_148 = None + x_150 = x_144 + x_149 + x_144 = x_149 = None + layer_norm_36 = torch.nn.functional.layer_norm( + x_150, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_ = (None) + linear_72 = torch._C._nn.linear( + layer_norm_36, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_36 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_40 = linear_72.reshape(1, 197, 3, 3, 64) + linear_72 = None + qkv_12 = reshape_40.permute(2, 0, 3, 1, 4) + reshape_40 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_16 = unbind_12[0] + k_16 = unbind_12[1] + v_16 = unbind_12[2] + unbind_12 = None + x_151 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, attn_mask=None, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_22 = x_151.transpose(1, 2) + x_151 = None + x_152 = transpose_22.reshape(1, 197, 192) + transpose_22 = None + x_153 = torch._C._nn.linear( + x_152, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_152 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_154 = torch.nn.functional.dropout(x_153, 0.0, False, False) + x_153 = None + x_155 = x_150 + x_154 + x_150 = x_154 = None + layer_norm_37 = torch.nn.functional.layer_norm( + x_155, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_ = (None) + x_156 = torch._C._nn.linear( + layer_norm_37, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_37 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_157 = torch._C._nn.gelu(x_156, approximate="none") + x_156 = None + x_158 = torch.nn.functional.dropout(x_157, 0.0, False, False) + x_157 = None + x_159 = torch._C._nn.linear( + x_158, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_158 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_160 = torch.nn.functional.dropout(x_159, 0.0, False, False) + x_159 = None + x_161 = x_155 + x_160 + x_155 = x_160 = None + layer_norm_38 = torch.nn.functional.layer_norm( + x_161, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_ = (None) + linear_76 = torch._C._nn.linear( + layer_norm_38, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_38 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_42 = linear_76.reshape(1, 197, 3, 3, 64) + linear_76 = None + qkv_13 = reshape_42.permute(2, 0, 3, 1, 4) + reshape_42 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_17 = unbind_13[0] + k_17 = unbind_13[1] + v_17 = unbind_13[2] + unbind_13 = None + x_162 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, attn_mask=None, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_23 = x_162.transpose(1, 2) + x_162 = None + x_163 = transpose_23.reshape(1, 197, 192) + transpose_23 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_163 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_ = (None) + x_165 = torch.nn.functional.dropout(x_164, 0.0, False, False) + x_164 = None + x_166 = x_161 + x_165 + x_161 = x_165 = None + layer_norm_39 = torch.nn.functional.layer_norm( + x_166, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_ = (None) + x_167 = torch._C._nn.linear( + layer_norm_39, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_39 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_168 = torch._C._nn.gelu(x_167, approximate="none") + x_167 = None + x_169 = torch.nn.functional.dropout(x_168, 0.0, False, False) + x_168 = None + x_170 = torch._C._nn.linear( + x_169, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_169 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_171 = torch.nn.functional.dropout(x_170, 0.0, False, False) + x_170 = None + x_172 = x_166 + x_171 + x_166 = x_171 = None + layer_norm_40 = torch.nn.functional.layer_norm( + x_172, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_ = (None) + linear_80 = torch._C._nn.linear( + layer_norm_40, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + layer_norm_40 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_ = (None) + reshape_44 = linear_80.reshape(1, 197, 3, 3, 64) + linear_80 = None + qkv_14 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_18 = unbind_14[0] + k_18 = unbind_14[1] + v_18 = unbind_14[2] + unbind_14 = None + x_173 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_24 = x_173.transpose(1, 2) + x_173 = None + x_174 = transpose_24.reshape(1, 197, 192) + transpose_24 = None + x_175 = torch._C._nn.linear( + x_174, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_174 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_ = (None) + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = x_172 + x_176 + x_172 = x_176 = None + layer_norm_41 = torch.nn.functional.layer_norm( + x_177, + (192,), + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_ = (None) + x_178 = torch._C._nn.linear( + layer_norm_41, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + layer_norm_41 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_ = (None) + x_179 = torch._C._nn.gelu(x_178, approximate="none") + x_178 = None + x_180 = torch.nn.functional.dropout(x_179, 0.0, False, False) + x_179 = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_180 = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_ = (None) + x_182 = torch.nn.functional.dropout(x_181, 0.0, False, False) + x_181 = None + x_183 = x_177 + x_182 + x_177 = x_182 = None + getitem_85 = x_139[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_25 = torch.nn.functional.layer_norm( + getitem_85, + (96,), + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_85 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_ = (None) + input_26 = torch._C._nn.gelu(input_25, approximate="none") + input_25 = None + input_27 = torch._C._nn.linear( + input_26, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_, + ) + input_26 = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_86 = x_183[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + input_28 = torch.nn.functional.layer_norm( + getitem_86, + (192,), + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_86 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_ = (None) + input_29 = torch._C._nn.gelu(input_28, approximate="none") + input_28 = None + input_30 = torch._C._nn.linear( + input_29, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_, + ) + input_29 = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_87 = x_183[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_8 = torch.cat((input_27, getitem_87), dim=1) + input_27 = getitem_87 = None + getitem_88 = tmp_8[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_44 = torch.nn.functional.layer_norm( + tmp_8, + (192,), + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_8 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_ = (None) + getitem_89 = layer_norm_44[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_86 = torch._C._nn.linear( + getitem_89, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_, + ) + getitem_89 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_46 = linear_86.reshape(1, 1, 3, 64) + linear_86 = None + q_19 = reshape_46.permute(0, 2, 1, 3) + reshape_46 = None + linear_87 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_47 = linear_87.reshape(1, 197, 3, 64) + linear_87 = None + k_19 = reshape_47.permute(0, 2, 1, 3) + reshape_47 = None + linear_88 = torch._C._nn.linear( + layer_norm_44, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_44 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_48 = linear_88.reshape(1, 197, 3, 64) + linear_88 = None + v_19 = reshape_48.permute(0, 2, 1, 3) + reshape_48 = None + transpose_25 = k_19.transpose(-2, -1) + k_19 = None + matmul_8 = q_19 @ transpose_25 + q_19 = transpose_25 = None + attn_12 = matmul_8 * 0.125 + matmul_8 = None + attn_13 = attn_12.softmax(dim=-1) + attn_12 = None + attn_14 = torch.nn.functional.dropout(attn_13, 0.0, False, False) + attn_13 = None + matmul_9 = attn_14 @ v_19 + attn_14 = v_19 = None + transpose_26 = matmul_9.transpose(1, 2) + matmul_9 = None + x_184 = transpose_26.reshape(1, 1, 192) + transpose_26 = None + x_185 = torch._C._nn.linear( + x_184, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_184 = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_ = (None) + x_186 = torch.nn.functional.dropout(x_185, 0.0, False, False) + x_185 = None + x_187 = getitem_88 + x_186 + getitem_88 = x_186 = None + getitem_90 = x_187[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_187 = None + input_31 = torch.nn.functional.layer_norm( + getitem_90, + (192,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_, + 1e-06, + ) + getitem_90 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_ = (None) + input_32 = torch._C._nn.gelu(input_31, approximate="none") + input_31 = None + input_33 = torch._C._nn.linear( + input_32, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_, + ) + input_32 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_ = (None) + getitem_91 = x_139[(slice(None, None, None), slice(1, None, None), Ellipsis)] + tmp_9 = torch.cat((input_33, getitem_91), dim=1) + input_33 = getitem_91 = None + getitem_92 = x_139[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_139 = None + tmp_10 = torch.cat((input_30, getitem_92), dim=1) + input_30 = getitem_92 = None + getitem_93 = tmp_10[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + layer_norm_46 = torch.nn.functional.layer_norm( + tmp_10, + (96,), + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + tmp_10 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_ = (None) + getitem_94 = layer_norm_46[ + (slice(None, None, None), slice(0, 1, None), Ellipsis) + ] + linear_91 = torch._C._nn.linear( + getitem_94, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_, + ) + getitem_94 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_ = (None) + reshape_50 = linear_91.reshape(1, 1, 3, 32) + linear_91 = None + q_20 = reshape_50.permute(0, 2, 1, 3) + reshape_50 = None + linear_92 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_, + ) + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_ = (None) + reshape_51 = linear_92.reshape(1, 401, 3, 32) + linear_92 = None + k_20 = reshape_51.permute(0, 2, 1, 3) + reshape_51 = None + linear_93 = torch._C._nn.linear( + layer_norm_46, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_, + ) + layer_norm_46 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_ = (None) + reshape_52 = linear_93.reshape(1, 401, 3, 32) + linear_93 = None + v_20 = reshape_52.permute(0, 2, 1, 3) + reshape_52 = None + transpose_27 = k_20.transpose(-2, -1) + k_20 = None + matmul_10 = q_20 @ transpose_27 + q_20 = transpose_27 = None + attn_15 = matmul_10 * 0.1767766952966369 + matmul_10 = None + attn_16 = attn_15.softmax(dim=-1) + attn_15 = None + attn_17 = torch.nn.functional.dropout(attn_16, 0.0, False, False) + attn_16 = None + matmul_11 = attn_17 @ v_20 + attn_17 = v_20 = None + transpose_28 = matmul_11.transpose(1, 2) + matmul_11 = None + x_188 = transpose_28.reshape(1, 1, 96) + transpose_28 = None + x_189 = torch._C._nn.linear( + x_188, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_188 = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_ = l_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_ = (None) + x_190 = torch.nn.functional.dropout(x_189, 0.0, False, False) + x_189 = None + x_191 = getitem_93 + x_190 + getitem_93 = x_190 = None + getitem_95 = x_191[(slice(None, None, None), slice(0, 1, None), Ellipsis)] + x_191 = None + input_34 = torch.nn.functional.layer_norm( + getitem_95, + (96,), + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_, + 1e-06, + ) + getitem_95 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_ = (None) + input_35 = torch._C._nn.gelu(input_34, approximate="none") + input_34 = None + input_36 = torch._C._nn.linear( + input_35, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_, + ) + input_35 = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_ = l_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_ = (None) + getitem_96 = x_183[(slice(None, None, None), slice(1, None, None), Ellipsis)] + x_183 = None + tmp_11 = torch.cat((input_36, getitem_96), dim=1) + input_36 = getitem_96 = None + x_192 = torch.nn.functional.layer_norm( + tmp_9, + (96,), + l_self_modules_norm_modules_0_parameters_weight_, + l_self_modules_norm_modules_0_parameters_bias_, + 1e-06, + ) + tmp_9 = ( + l_self_modules_norm_modules_0_parameters_weight_ + ) = l_self_modules_norm_modules_0_parameters_bias_ = None + x_193 = torch.nn.functional.layer_norm( + tmp_11, + (192,), + l_self_modules_norm_modules_1_parameters_weight_, + l_self_modules_norm_modules_1_parameters_bias_, + 1e-06, + ) + tmp_11 = ( + l_self_modules_norm_modules_1_parameters_weight_ + ) = l_self_modules_norm_modules_1_parameters_bias_ = None + x_194 = x_192[(slice(None, None, None), 0)] + x_192 = None + x_195 = x_193[(slice(None, None, None), 0)] + x_193 = None + dropout_59 = torch.nn.functional.dropout(x_194, 0.0, False, False) + x_194 = None + dropout_60 = torch.nn.functional.dropout(x_195, 0.0, False, False) + x_195 = None + linear_96 = torch._C._nn.linear( + dropout_59, + l_self_modules_head_modules_0_parameters_weight_, + l_self_modules_head_modules_0_parameters_bias_, + ) + dropout_59 = ( + l_self_modules_head_modules_0_parameters_weight_ + ) = l_self_modules_head_modules_0_parameters_bias_ = None + linear_97 = torch._C._nn.linear( + dropout_60, + l_self_modules_head_modules_1_parameters_weight_, + l_self_modules_head_modules_1_parameters_bias_, + ) + dropout_60 = ( + l_self_modules_head_modules_1_parameters_weight_ + ) = l_self_modules_head_modules_1_parameters_bias_ = None + stack = torch.stack([linear_96, linear_97], dim=0) + linear_96 = linear_97 = None + x_196 = torch.mean(stack, dim=0) + stack = None + return (x_196,) diff --git a/samples/timm/crossvit_tiny_240.in1k/weight_meta.py b/samples/timm/crossvit_tiny_240.in1k/weight_meta.py new file mode 100644 index 0000000000..1013e8ace6 --- /dev/null +++ b/samples/timm/crossvit_tiny_240.in1k/weight_meta.py @@ -0,0 +1,3048 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.228 + std = 1.288 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_0_modules_proj_parameters_weight_" + shape = [96, 3, 12, 12] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.028 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_0_modules_proj_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.005 + std = 0.028 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_0_: + name = "L_self_parameters_cls_token_0_" + shape = [1, 1, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.004 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_0_: + name = "L_self_parameters_pos_embed_0_" + shape = [1, 401, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_1_modules_proj_parameters_weight_" + shape = [192, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_1_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_1_: + name = "L_self_parameters_cls_token_1_" + shape = [1, 1, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_1_: + name = "L_self_parameters_pos_embed_1_" + shape = [1, 197, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [288, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [96, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [96, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [288, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [96, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [96, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [288, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_norm2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [384, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [96, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_0_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_blocks_modules_1_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_weight_" + shape = [192, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_0_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_weight_" + shape = [96, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_projs_modules_1_modules_2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wq_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wk_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_wv_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_weight_" + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_0_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_weight_" + shape = [96, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_0_modules_2_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_norm1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wq_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wk_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_wv_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_weight_" + shape = [96, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_fusion_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_weight_" + shape = [192, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_revert_projs_modules_1_modules_2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_weight_: + name = "L_self_modules_norm_modules_0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_0_parameters_bias_: + name = "L_self_modules_norm_modules_0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_weight_: + name = "L_self_modules_norm_modules_1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_modules_1_parameters_bias_: + name = "L_self_modules_norm_modules_1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_weight_: + name = "L_self_modules_head_modules_0_parameters_weight_" + shape = [1000, 96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_0_parameters_bias_: + name = "L_self_modules_head_modules_0_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_weight_: + name = "L_self_modules_head_modules_1_parameters_weight_" + shape = [1000, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_1_parameters_bias_: + name = "L_self_modules_head_modules_1_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cs3darknet_focus_l.c2ns_in1k/graph_hash.txt b/samples/timm/cs3darknet_focus_l.c2ns_in1k/graph_hash.txt new file mode 100644 index 0000000000..41a4d772ce --- /dev/null +++ b/samples/timm/cs3darknet_focus_l.c2ns_in1k/graph_hash.txt @@ -0,0 +1 @@ +b1b5d407a9ce4756c0f5f058742f0f1c7d9451a8b87c22a969d7db3f00854a06 \ No newline at end of file diff --git a/samples/timm/cs3darknet_focus_l.c2ns_in1k/graph_net.json b/samples/timm/cs3darknet_focus_l.c2ns_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/cs3darknet_focus_l.c2ns_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cs3darknet_focus_l.c2ns_in1k/input_meta.py b/samples/timm/cs3darknet_focus_l.c2ns_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cs3darknet_focus_l.c2ns_in1k/input_tensor_constraints.py b/samples/timm/cs3darknet_focus_l.c2ns_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cs3darknet_focus_l.c2ns_in1k/model.py b/samples/timm/cs3darknet_focus_l.c2ns_in1k/model.py new file mode 100644 index 0000000000..67fa432fe3 --- /dev/null +++ b/samples/timm/cs3darknet_focus_l.c2ns_in1k/model.py @@ -0,0 +1,1992 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (2, 2), + (2, 2), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.silu(x_1, inplace=True) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_5 = torch.nn.functional.silu(x_4, inplace=True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_5 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_8 = torch.nn.functional.silu(x_7, inplace=True) + x_7 = None + split = x_8.split(64, dim=1) + x_8 = None + x1 = split[0] + x2 = split[1] + split = None + x_9 = torch.conv2d( + x1, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_11 = torch.nn.functional.silu(x_10, inplace=True) + x_10 = None + x_12 = torch.conv2d( + x_11, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_11 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_13 = torch.nn.functional.batch_norm( + x_12, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_14 = torch.nn.functional.silu(x_13, inplace=True) + x_13 = None + x_15 = x_14 + x1 + x_14 = x1 = None + x_16 = torch.conv2d( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_17 = torch.nn.functional.batch_norm( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_18 = torch.nn.functional.silu(x_17, inplace=True) + x_17 = None + x_19 = torch.conv2d( + x_18, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_18 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_20 = torch.nn.functional.batch_norm( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_19 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_21 = torch.nn.functional.silu(x_20, inplace=True) + x_20 = None + x_22 = x_21 + x_15 + x_21 = x_15 = None + x_23 = torch.conv2d( + x_22, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_24 = torch.nn.functional.batch_norm( + x_23, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_25 = torch.nn.functional.silu(x_24, inplace=True) + x_24 = None + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_27 = torch.nn.functional.batch_norm( + x_26, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_28 = torch.nn.functional.silu(x_27, inplace=True) + x_27 = None + x_29 = x_28 + x_22 + x_28 = x_22 = None + cat = torch.cat([x_29, x2], dim=1) + x_29 = x2 = None + x_30 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_31 = torch.nn.functional.batch_norm( + x_30, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_30 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_32 = torch.nn.functional.silu(x_31, inplace=True) + x_31 = None + x_33 = torch.conv2d( + x_32, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_32 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_34 = torch.nn.functional.batch_norm( + x_33, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_33 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_35 = torch.nn.functional.silu(x_34, inplace=True) + x_34 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_37 = torch.nn.functional.batch_norm( + x_36, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_36 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_38 = torch.nn.functional.silu(x_37, inplace=True) + x_37 = None + split_1 = x_38.split(128, dim=1) + x_38 = None + x1_1 = split_1[0] + x2_1 = split_1[1] + split_1 = None + x_39 = torch.conv2d( + x1_1, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_40 = torch.nn.functional.batch_norm( + x_39, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_39 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_41 = torch.nn.functional.silu(x_40, inplace=True) + x_40 = None + x_42 = torch.conv2d( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_43 = torch.nn.functional.batch_norm( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_44 = torch.nn.functional.silu(x_43, inplace=True) + x_43 = None + x_45 = x_44 + x1_1 + x_44 = x1_1 = None + x_46 = torch.conv2d( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_47 = torch.nn.functional.batch_norm( + x_46, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_48 = torch.nn.functional.silu(x_47, inplace=True) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_50 = torch.nn.functional.batch_norm( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_49 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_51 = torch.nn.functional.silu(x_50, inplace=True) + x_50 = None + x_52 = x_51 + x_45 + x_51 = x_45 = None + x_53 = torch.conv2d( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_54 = torch.nn.functional.batch_norm( + x_53, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_53 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_55 = torch.nn.functional.silu(x_54, inplace=True) + x_54 = None + x_56 = torch.conv2d( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_55 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_57 = torch.nn.functional.batch_norm( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_58 = torch.nn.functional.silu(x_57, inplace=True) + x_57 = None + x_59 = x_58 + x_52 + x_58 = x_52 = None + x_60 = torch.conv2d( + x_59, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_61 = torch.nn.functional.batch_norm( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_60 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_62 = torch.nn.functional.silu(x_61, inplace=True) + x_61 = None + x_63 = torch.conv2d( + x_62, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_64 = torch.nn.functional.batch_norm( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_65 = torch.nn.functional.silu(x_64, inplace=True) + x_64 = None + x_66 = x_65 + x_59 + x_65 = x_59 = None + x_67 = torch.conv2d( + x_66, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_68 = torch.nn.functional.batch_norm( + x_67, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_67 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_69 = torch.nn.functional.silu(x_68, inplace=True) + x_68 = None + x_70 = torch.conv2d( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_71 = torch.nn.functional.batch_norm( + x_70, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_70 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_72 = torch.nn.functional.silu(x_71, inplace=True) + x_71 = None + x_73 = x_72 + x_66 + x_72 = x_66 = None + x_74 = torch.conv2d( + x_73, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_75 = torch.nn.functional.batch_norm( + x_74, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_74 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_76 = torch.nn.functional.silu(x_75, inplace=True) + x_75 = None + x_77 = torch.conv2d( + x_76, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_76 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_78 = torch.nn.functional.batch_norm( + x_77, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_77 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_79 = torch.nn.functional.silu(x_78, inplace=True) + x_78 = None + x_80 = x_79 + x_73 + x_79 = x_73 = None + cat_1 = torch.cat([x_80, x2_1], dim=1) + x_80 = x2_1 = None + x_81 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_82 = torch.nn.functional.batch_norm( + x_81, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_81 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_83 = torch.nn.functional.silu(x_82, inplace=True) + x_82 = None + x_84 = torch.conv2d( + x_83, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_83 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_85 = torch.nn.functional.batch_norm( + x_84, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_84 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_86 = torch.nn.functional.silu(x_85, inplace=True) + x_85 = None + x_87 = torch.conv2d( + x_86, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_86 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_88 = torch.nn.functional.batch_norm( + x_87, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_87 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_89 = torch.nn.functional.silu(x_88, inplace=True) + x_88 = None + split_2 = x_89.split(256, dim=1) + x_89 = None + x1_2 = split_2[0] + x2_2 = split_2[1] + split_2 = None + x_90 = torch.conv2d( + x1_2, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_91 = torch.nn.functional.batch_norm( + x_90, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_90 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_92 = torch.nn.functional.silu(x_91, inplace=True) + x_91 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_94 = torch.nn.functional.batch_norm( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_95 = torch.nn.functional.silu(x_94, inplace=True) + x_94 = None + x_96 = x_95 + x1_2 + x_95 = x1_2 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_98 = torch.nn.functional.batch_norm( + x_97, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_97 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_99 = torch.nn.functional.silu(x_98, inplace=True) + x_98 = None + x_100 = torch.conv2d( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_101 = torch.nn.functional.batch_norm( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_102 = torch.nn.functional.silu(x_101, inplace=True) + x_101 = None + x_103 = x_102 + x_96 + x_102 = x_96 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_105 = torch.nn.functional.batch_norm( + x_104, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_104 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_106 = torch.nn.functional.silu(x_105, inplace=True) + x_105 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_106 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_108 = torch.nn.functional.batch_norm( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_107 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_109 = torch.nn.functional.silu(x_108, inplace=True) + x_108 = None + x_110 = x_109 + x_103 + x_109 = x_103 = None + x_111 = torch.conv2d( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_112 = torch.nn.functional.batch_norm( + x_111, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_111 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_113 = torch.nn.functional.silu(x_112, inplace=True) + x_112 = None + x_114 = torch.conv2d( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_113 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_115 = torch.nn.functional.batch_norm( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_116 = torch.nn.functional.silu(x_115, inplace=True) + x_115 = None + x_117 = x_116 + x_110 + x_116 = x_110 = None + x_118 = torch.conv2d( + x_117, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_119 = torch.nn.functional.batch_norm( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_118 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_120 = torch.nn.functional.silu(x_119, inplace=True) + x_119 = None + x_121 = torch.conv2d( + x_120, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_122 = torch.nn.functional.batch_norm( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_123 = torch.nn.functional.silu(x_122, inplace=True) + x_122 = None + x_124 = x_123 + x_117 + x_123 = x_117 = None + x_125 = torch.conv2d( + x_124, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_126 = torch.nn.functional.batch_norm( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_125 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_127 = torch.nn.functional.silu(x_126, inplace=True) + x_126 = None + x_128 = torch.conv2d( + x_127, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_127 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_129 = torch.nn.functional.batch_norm( + x_128, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_128 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_130 = torch.nn.functional.silu(x_129, inplace=True) + x_129 = None + x_131 = x_130 + x_124 + x_130 = x_124 = None + x_132 = torch.conv2d( + x_131, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_133 = torch.nn.functional.batch_norm( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_134 = torch.nn.functional.silu(x_133, inplace=True) + x_133 = None + x_135 = torch.conv2d( + x_134, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_134 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_136 = torch.nn.functional.batch_norm( + x_135, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_135 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_137 = torch.nn.functional.silu(x_136, inplace=True) + x_136 = None + x_138 = x_137 + x_131 + x_137 = x_131 = None + x_139 = torch.conv2d( + x_138, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_140 = torch.nn.functional.batch_norm( + x_139, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_139 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_141 = torch.nn.functional.silu(x_140, inplace=True) + x_140 = None + x_142 = torch.conv2d( + x_141, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_141 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_143 = torch.nn.functional.batch_norm( + x_142, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_144 = torch.nn.functional.silu(x_143, inplace=True) + x_143 = None + x_145 = x_144 + x_138 + x_144 = x_138 = None + x_146 = torch.conv2d( + x_145, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_147 = torch.nn.functional.batch_norm( + x_146, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_146 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = (None) + x_148 = torch.nn.functional.silu(x_147, inplace=True) + x_147 = None + x_149 = torch.conv2d( + x_148, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_148 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = (None) + x_150 = torch.nn.functional.batch_norm( + x_149, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_149 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = (None) + x_151 = torch.nn.functional.silu(x_150, inplace=True) + x_150 = None + x_152 = x_151 + x_145 + x_151 = x_145 = None + cat_2 = torch.cat([x_152, x2_2], dim=1) + x_152 = x2_2 = None + x_153 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_154 = torch.nn.functional.batch_norm( + x_153, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_153 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_155 = torch.nn.functional.silu(x_154, inplace=True) + x_154 = None + x_156 = torch.conv2d( + x_155, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_155 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_157 = torch.nn.functional.batch_norm( + x_156, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_156 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_158 = torch.nn.functional.silu(x_157, inplace=True) + x_157 = None + x_159 = torch.conv2d( + x_158, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_158 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_160 = torch.nn.functional.batch_norm( + x_159, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_159 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_161 = torch.nn.functional.silu(x_160, inplace=True) + x_160 = None + split_3 = x_161.split(512, dim=1) + x_161 = None + x1_3 = split_3[0] + x2_3 = split_3[1] + split_3 = None + x_162 = torch.conv2d( + x1_3, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_163 = torch.nn.functional.batch_norm( + x_162, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_162 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_164 = torch.nn.functional.silu(x_163, inplace=True) + x_163 = None + x_165 = torch.conv2d( + x_164, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_164 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_166 = torch.nn.functional.batch_norm( + x_165, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_165 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_167 = torch.nn.functional.silu(x_166, inplace=True) + x_166 = None + x_168 = x_167 + x1_3 + x_167 = x1_3 = None + x_169 = torch.conv2d( + x_168, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_170 = torch.nn.functional.batch_norm( + x_169, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_169 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_171 = torch.nn.functional.silu(x_170, inplace=True) + x_170 = None + x_172 = torch.conv2d( + x_171, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_171 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_173 = torch.nn.functional.batch_norm( + x_172, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_172 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_174 = torch.nn.functional.silu(x_173, inplace=True) + x_173 = None + x_175 = x_174 + x_168 + x_174 = x_168 = None + x_176 = torch.conv2d( + x_175, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_177 = torch.nn.functional.batch_norm( + x_176, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_176 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_178 = torch.nn.functional.silu(x_177, inplace=True) + x_177 = None + x_179 = torch.conv2d( + x_178, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_178 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_180 = torch.nn.functional.batch_norm( + x_179, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_179 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_181 = torch.nn.functional.silu(x_180, inplace=True) + x_180 = None + x_182 = x_181 + x_175 + x_181 = x_175 = None + cat_3 = torch.cat([x_182, x2_3], dim=1) + x_182 = x2_3 = None + x_183 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_184 = torch.nn.functional.batch_norm( + x_183, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_183 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_185 = torch.nn.functional.silu(x_184, inplace=True) + x_184 = None + x_186 = torch.nn.functional.adaptive_avg_pool2d(x_185, 1) + x_185 = None + x_187 = x_186.flatten(1, -1) + x_186 = None + x_188 = torch.nn.functional.dropout(x_187, 0.0, False, False) + x_187 = None + x_189 = torch._C._nn.linear( + x_188, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_188 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_189,) diff --git a/samples/timm/cs3darknet_focus_l.c2ns_in1k/weight_meta.py b/samples/timm/cs3darknet_focus_l.c2ns_in1k/weight_meta.py new file mode 100644 index 0000000000..f035c3b0cf --- /dev/null +++ b/samples/timm/cs3darknet_focus_l.c2ns_in1k/weight_meta.py @@ -0,0 +1,2812 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 3, 6, 6] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.228 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.005 + std = 0.180 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.124 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.124 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [1024, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cs3darknet_focus_m.c2ns_in1k/graph_hash.txt b/samples/timm/cs3darknet_focus_m.c2ns_in1k/graph_hash.txt new file mode 100644 index 0000000000..a339e5e87f --- /dev/null +++ b/samples/timm/cs3darknet_focus_m.c2ns_in1k/graph_hash.txt @@ -0,0 +1 @@ +f3fa74260ceba665ad095e7ef6c381c42711ac3f3ca201cd4825c0f522f7dbde \ No newline at end of file diff --git a/samples/timm/cs3darknet_focus_m.c2ns_in1k/graph_net.json b/samples/timm/cs3darknet_focus_m.c2ns_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/cs3darknet_focus_m.c2ns_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cs3darknet_focus_m.c2ns_in1k/input_meta.py b/samples/timm/cs3darknet_focus_m.c2ns_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cs3darknet_focus_m.c2ns_in1k/input_tensor_constraints.py b/samples/timm/cs3darknet_focus_m.c2ns_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cs3darknet_focus_m.c2ns_in1k/model.py b/samples/timm/cs3darknet_focus_m.c2ns_in1k/model.py new file mode 100644 index 0000000000..75b43c5d73 --- /dev/null +++ b/samples/timm/cs3darknet_focus_m.c2ns_in1k/model.py @@ -0,0 +1,1502 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (2, 2), + (2, 2), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.silu(x_1, inplace=True) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_5 = torch.nn.functional.silu(x_4, inplace=True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_5 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_8 = torch.nn.functional.silu(x_7, inplace=True) + x_7 = None + split = x_8.split(48, dim=1) + x_8 = None + x1 = split[0] + x2 = split[1] + split = None + x_9 = torch.conv2d( + x1, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_11 = torch.nn.functional.silu(x_10, inplace=True) + x_10 = None + x_12 = torch.conv2d( + x_11, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_11 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_13 = torch.nn.functional.batch_norm( + x_12, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_14 = torch.nn.functional.silu(x_13, inplace=True) + x_13 = None + x_15 = x_14 + x1 + x_14 = x1 = None + x_16 = torch.conv2d( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_17 = torch.nn.functional.batch_norm( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_18 = torch.nn.functional.silu(x_17, inplace=True) + x_17 = None + x_19 = torch.conv2d( + x_18, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_18 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_20 = torch.nn.functional.batch_norm( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_19 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_21 = torch.nn.functional.silu(x_20, inplace=True) + x_20 = None + x_22 = x_21 + x_15 + x_21 = x_15 = None + cat = torch.cat([x_22, x2], dim=1) + x_22 = x2 = None + x_23 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_24 = torch.nn.functional.batch_norm( + x_23, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_23 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_25 = torch.nn.functional.silu(x_24, inplace=True) + x_24 = None + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_25 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_27 = torch.nn.functional.batch_norm( + x_26, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_26 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_28 = torch.nn.functional.silu(x_27, inplace=True) + x_27 = None + x_29 = torch.conv2d( + x_28, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_28 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_30 = torch.nn.functional.batch_norm( + x_29, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_29 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_31 = torch.nn.functional.silu(x_30, inplace=True) + x_30 = None + split_1 = x_31.split(96, dim=1) + x_31 = None + x1_1 = split_1[0] + x2_1 = split_1[1] + split_1 = None + x_32 = torch.conv2d( + x1_1, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_33 = torch.nn.functional.batch_norm( + x_32, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_32 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_34 = torch.nn.functional.silu(x_33, inplace=True) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_36 = torch.nn.functional.batch_norm( + x_35, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_35 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_37 = torch.nn.functional.silu(x_36, inplace=True) + x_36 = None + x_38 = x_37 + x1_1 + x_37 = x1_1 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_40 = torch.nn.functional.batch_norm( + x_39, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_39 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_41 = torch.nn.functional.silu(x_40, inplace=True) + x_40 = None + x_42 = torch.conv2d( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_43 = torch.nn.functional.batch_norm( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_44 = torch.nn.functional.silu(x_43, inplace=True) + x_43 = None + x_45 = x_44 + x_38 + x_44 = x_38 = None + x_46 = torch.conv2d( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_47 = torch.nn.functional.batch_norm( + x_46, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_48 = torch.nn.functional.silu(x_47, inplace=True) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_50 = torch.nn.functional.batch_norm( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_49 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_51 = torch.nn.functional.silu(x_50, inplace=True) + x_50 = None + x_52 = x_51 + x_45 + x_51 = x_45 = None + x_53 = torch.conv2d( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_54 = torch.nn.functional.batch_norm( + x_53, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_53 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_55 = torch.nn.functional.silu(x_54, inplace=True) + x_54 = None + x_56 = torch.conv2d( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_55 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_57 = torch.nn.functional.batch_norm( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_58 = torch.nn.functional.silu(x_57, inplace=True) + x_57 = None + x_59 = x_58 + x_52 + x_58 = x_52 = None + cat_1 = torch.cat([x_59, x2_1], dim=1) + x_59 = x2_1 = None + x_60 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_61 = torch.nn.functional.batch_norm( + x_60, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_60 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_62 = torch.nn.functional.silu(x_61, inplace=True) + x_61 = None + x_63 = torch.conv2d( + x_62, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_62 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_64 = torch.nn.functional.batch_norm( + x_63, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_63 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_65 = torch.nn.functional.silu(x_64, inplace=True) + x_64 = None + x_66 = torch.conv2d( + x_65, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_65 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_67 = torch.nn.functional.batch_norm( + x_66, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_66 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_68 = torch.nn.functional.silu(x_67, inplace=True) + x_67 = None + split_2 = x_68.split(192, dim=1) + x_68 = None + x1_2 = split_2[0] + x2_2 = split_2[1] + split_2 = None + x_69 = torch.conv2d( + x1_2, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_70 = torch.nn.functional.batch_norm( + x_69, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_69 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_71 = torch.nn.functional.silu(x_70, inplace=True) + x_70 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_73 = torch.nn.functional.batch_norm( + x_72, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_72 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_74 = torch.nn.functional.silu(x_73, inplace=True) + x_73 = None + x_75 = x_74 + x1_2 + x_74 = x1_2 = None + x_76 = torch.conv2d( + x_75, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_77 = torch.nn.functional.batch_norm( + x_76, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_78 = torch.nn.functional.silu(x_77, inplace=True) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_80 = torch.nn.functional.batch_norm( + x_79, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_79 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_81 = torch.nn.functional.silu(x_80, inplace=True) + x_80 = None + x_82 = x_81 + x_75 + x_81 = x_75 = None + x_83 = torch.conv2d( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_84 = torch.nn.functional.batch_norm( + x_83, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_83 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_85 = torch.nn.functional.silu(x_84, inplace=True) + x_84 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_85 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_87 = torch.nn.functional.batch_norm( + x_86, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_86 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_88 = torch.nn.functional.silu(x_87, inplace=True) + x_87 = None + x_89 = x_88 + x_82 + x_88 = x_82 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_91 = torch.nn.functional.batch_norm( + x_90, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_90 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_92 = torch.nn.functional.silu(x_91, inplace=True) + x_91 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_94 = torch.nn.functional.batch_norm( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_95 = torch.nn.functional.silu(x_94, inplace=True) + x_94 = None + x_96 = x_95 + x_89 + x_95 = x_89 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_98 = torch.nn.functional.batch_norm( + x_97, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_97 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_99 = torch.nn.functional.silu(x_98, inplace=True) + x_98 = None + x_100 = torch.conv2d( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_101 = torch.nn.functional.batch_norm( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_102 = torch.nn.functional.silu(x_101, inplace=True) + x_101 = None + x_103 = x_102 + x_96 + x_102 = x_96 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_105 = torch.nn.functional.batch_norm( + x_104, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_104 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_106 = torch.nn.functional.silu(x_105, inplace=True) + x_105 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_106 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_108 = torch.nn.functional.batch_norm( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_107 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_109 = torch.nn.functional.silu(x_108, inplace=True) + x_108 = None + x_110 = x_109 + x_103 + x_109 = x_103 = None + cat_2 = torch.cat([x_110, x2_2], dim=1) + x_110 = x2_2 = None + x_111 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_112 = torch.nn.functional.batch_norm( + x_111, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_111 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_113 = torch.nn.functional.silu(x_112, inplace=True) + x_112 = None + x_114 = torch.conv2d( + x_113, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_113 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_115 = torch.nn.functional.batch_norm( + x_114, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_114 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_116 = torch.nn.functional.silu(x_115, inplace=True) + x_115 = None + x_117 = torch.conv2d( + x_116, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_116 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_118 = torch.nn.functional.batch_norm( + x_117, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_117 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_119 = torch.nn.functional.silu(x_118, inplace=True) + x_118 = None + split_3 = x_119.split(384, dim=1) + x_119 = None + x1_3 = split_3[0] + x2_3 = split_3[1] + split_3 = None + x_120 = torch.conv2d( + x1_3, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_121 = torch.nn.functional.batch_norm( + x_120, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_120 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_122 = torch.nn.functional.silu(x_121, inplace=True) + x_121 = None + x_123 = torch.conv2d( + x_122, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_122 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_124 = torch.nn.functional.batch_norm( + x_123, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_125 = torch.nn.functional.silu(x_124, inplace=True) + x_124 = None + x_126 = x_125 + x1_3 + x_125 = x1_3 = None + x_127 = torch.conv2d( + x_126, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_128 = torch.nn.functional.batch_norm( + x_127, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_127 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_129 = torch.nn.functional.silu(x_128, inplace=True) + x_128 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_131 = torch.nn.functional.batch_norm( + x_130, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_130 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_132 = torch.nn.functional.silu(x_131, inplace=True) + x_131 = None + x_133 = x_132 + x_126 + x_132 = x_126 = None + cat_3 = torch.cat([x_133, x2_3], dim=1) + x_133 = x2_3 = None + x_134 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_135 = torch.nn.functional.batch_norm( + x_134, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_134 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_136 = torch.nn.functional.silu(x_135, inplace=True) + x_135 = None + x_137 = torch.nn.functional.adaptive_avg_pool2d(x_136, 1) + x_136 = None + x_138 = x_137.flatten(1, -1) + x_137 = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = torch._C._nn.linear( + x_139, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_139 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_140,) diff --git a/samples/timm/cs3darknet_focus_m.c2ns_in1k/weight_meta.py b/samples/timm/cs3darknet_focus_m.c2ns_in1k/weight_meta.py new file mode 100644 index 0000000000..3ce5e8ad9e --- /dev/null +++ b/samples/timm/cs3darknet_focus_m.c2ns_in1k/weight_meta.py @@ -0,0 +1,2112 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [48, 3, 6, 6] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.223 + std = 1.288 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_" + shape = [48, 48, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.068 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [96, 48, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.144 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [48, 48, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.207 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [48, 48, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.069 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [48, 48, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.204 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [48, 48, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.068 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [96, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.145 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [96, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [192, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [96, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.145 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [96, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [96, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.144 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [96, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [96, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.144 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [96, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [96, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.144 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [96, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.101 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.101 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.101 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [384, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [384, 384, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [768, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [384, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [384, 384, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [384, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [384, 384, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/graph_hash.txt b/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/graph_hash.txt new file mode 100644 index 0000000000..7cbd7875c3 --- /dev/null +++ b/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/graph_hash.txt @@ -0,0 +1 @@ +338be4333e6e92fd045ec42c48b512f53c2984df5171d281b13dfe16fe00d35a \ No newline at end of file diff --git a/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/graph_net.json b/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/input_meta.py b/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/input_tensor_constraints.py b/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/model.py b/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/model.py new file mode 100644 index 0000000000..30be2eebc4 --- /dev/null +++ b/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/model.py @@ -0,0 +1,1152 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (2, 2), + (2, 2), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.silu(x_1, inplace=True) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_5 = torch.nn.functional.silu(x_4, inplace=True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_5 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_8 = torch.nn.functional.silu(x_7, inplace=True) + x_7 = None + split = x_8.split(32, dim=1) + x_8 = None + x1 = split[0] + x2 = split[1] + split = None + x_9 = torch.conv2d( + x1, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_11 = torch.nn.functional.silu(x_10, inplace=True) + x_10 = None + x_12 = torch.conv2d( + x_11, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_11 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_13 = torch.nn.functional.batch_norm( + x_12, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_14 = torch.nn.functional.silu(x_13, inplace=True) + x_13 = None + x_15 = x_14 + x1 + x_14 = x1 = None + cat = torch.cat([x_15, x2], dim=1) + x_15 = x2 = None + x_16 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_17 = torch.nn.functional.batch_norm( + x_16, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_16 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_18 = torch.nn.functional.silu(x_17, inplace=True) + x_17 = None + x_19 = torch.conv2d( + x_18, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_18 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_20 = torch.nn.functional.batch_norm( + x_19, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_19 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_21 = torch.nn.functional.silu(x_20, inplace=True) + x_20 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_23 = torch.nn.functional.batch_norm( + x_22, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_22 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_24 = torch.nn.functional.silu(x_23, inplace=True) + x_23 = None + split_1 = x_24.split(64, dim=1) + x_24 = None + x1_1 = split_1[0] + x2_1 = split_1[1] + split_1 = None + x_25 = torch.conv2d( + x1_1, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_26 = torch.nn.functional.batch_norm( + x_25, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_25 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_27 = torch.nn.functional.silu(x_26, inplace=True) + x_26 = None + x_28 = torch.conv2d( + x_27, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_29 = torch.nn.functional.batch_norm( + x_28, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_28 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_30 = torch.nn.functional.silu(x_29, inplace=True) + x_29 = None + x_31 = x_30 + x1_1 + x_30 = x1_1 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_33 = torch.nn.functional.batch_norm( + x_32, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_32 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_34 = torch.nn.functional.silu(x_33, inplace=True) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_36 = torch.nn.functional.batch_norm( + x_35, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_35 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_37 = torch.nn.functional.silu(x_36, inplace=True) + x_36 = None + x_38 = x_37 + x_31 + x_37 = x_31 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_40 = torch.nn.functional.batch_norm( + x_39, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_39 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_41 = torch.nn.functional.silu(x_40, inplace=True) + x_40 = None + x_42 = torch.conv2d( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_41 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_43 = torch.nn.functional.batch_norm( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_44 = torch.nn.functional.silu(x_43, inplace=True) + x_43 = None + x_45 = x_44 + x_38 + x_44 = x_38 = None + cat_1 = torch.cat([x_45, x2_1], dim=1) + x_45 = x2_1 = None + x_46 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_47 = torch.nn.functional.batch_norm( + x_46, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_46 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_48 = torch.nn.functional.silu(x_47, inplace=True) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_50 = torch.nn.functional.batch_norm( + x_49, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_49 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_51 = torch.nn.functional.silu(x_50, inplace=True) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_53 = torch.nn.functional.batch_norm( + x_52, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_52 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_54 = torch.nn.functional.silu(x_53, inplace=True) + x_53 = None + split_2 = x_54.split(128, dim=1) + x_54 = None + x1_2 = split_2[0] + x2_2 = split_2[1] + split_2 = None + x_55 = torch.conv2d( + x1_2, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_56 = torch.nn.functional.batch_norm( + x_55, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_55 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_57 = torch.nn.functional.silu(x_56, inplace=True) + x_56 = None + x_58 = torch.conv2d( + x_57, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_57 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_59 = torch.nn.functional.batch_norm( + x_58, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_58 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_60 = torch.nn.functional.silu(x_59, inplace=True) + x_59 = None + x_61 = x_60 + x1_2 + x_60 = x1_2 = None + x_62 = torch.conv2d( + x_61, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_63 = torch.nn.functional.batch_norm( + x_62, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_62 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_64 = torch.nn.functional.silu(x_63, inplace=True) + x_63 = None + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_66 = torch.nn.functional.batch_norm( + x_65, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_65 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_67 = torch.nn.functional.silu(x_66, inplace=True) + x_66 = None + x_68 = x_67 + x_61 + x_67 = x_61 = None + x_69 = torch.conv2d( + x_68, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_70 = torch.nn.functional.batch_norm( + x_69, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_69 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_71 = torch.nn.functional.silu(x_70, inplace=True) + x_70 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_73 = torch.nn.functional.batch_norm( + x_72, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_72 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_74 = torch.nn.functional.silu(x_73, inplace=True) + x_73 = None + x_75 = x_74 + x_68 + x_74 = x_68 = None + x_76 = torch.conv2d( + x_75, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_77 = torch.nn.functional.batch_norm( + x_76, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_76 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_78 = torch.nn.functional.silu(x_77, inplace=True) + x_77 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_80 = torch.nn.functional.batch_norm( + x_79, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_79 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_81 = torch.nn.functional.silu(x_80, inplace=True) + x_80 = None + x_82 = x_81 + x_75 + x_81 = x_75 = None + cat_2 = torch.cat([x_82, x2_2], dim=1) + x_82 = x2_2 = None + x_83 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_84 = torch.nn.functional.batch_norm( + x_83, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_83 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_85 = torch.nn.functional.silu(x_84, inplace=True) + x_84 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_85 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_87 = torch.nn.functional.batch_norm( + x_86, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_86 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_88 = torch.nn.functional.silu(x_87, inplace=True) + x_87 = None + x_89 = torch.conv2d( + x_88, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_88 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_90 = torch.nn.functional.batch_norm( + x_89, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_89 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_91 = torch.nn.functional.silu(x_90, inplace=True) + x_90 = None + split_3 = x_91.split(256, dim=1) + x_91 = None + x1_3 = split_3[0] + x2_3 = split_3[1] + split_3 = None + x_92 = torch.conv2d( + x1_3, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_93 = torch.nn.functional.batch_norm( + x_92, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_92 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_94 = torch.nn.functional.silu(x_93, inplace=True) + x_93 = None + x_95 = torch.conv2d( + x_94, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_94 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_96 = torch.nn.functional.batch_norm( + x_95, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_95 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_97 = torch.nn.functional.silu(x_96, inplace=True) + x_96 = None + x_98 = x_97 + x1_3 + x_97 = x1_3 = None + cat_3 = torch.cat([x_98, x2_3], dim=1) + x_98 = x2_3 = None + x_99 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_100 = torch.nn.functional.batch_norm( + x_99, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_99 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_101 = torch.nn.functional.silu(x_100, inplace=True) + x_100 = None + x_102 = torch.nn.functional.adaptive_avg_pool2d(x_101, 1) + x_101 = None + x_103 = x_102.flatten(1, -1) + x_102 = None + x_104 = torch.nn.functional.dropout(x_103, 0.0, False, False) + x_103 = None + x_105 = torch._C._nn.linear( + x_104, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_104 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_105,) diff --git a/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/weight_meta.py b/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/weight_meta.py new file mode 100644 index 0000000000..f525f7fe60 --- /dev/null +++ b/samples/timm/cs3darknet_focus_s.ra4_e3600_r256_in1k/weight_meta.py @@ -0,0 +1,1612 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [32, 3, 6, 6] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.228 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_" + shape = [32, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [64, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.009 + std = 0.178 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [32, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.243 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [32, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.127 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cs3darknet_l.c2ns_in1k/graph_hash.txt b/samples/timm/cs3darknet_l.c2ns_in1k/graph_hash.txt new file mode 100644 index 0000000000..bf730a1093 --- /dev/null +++ b/samples/timm/cs3darknet_l.c2ns_in1k/graph_hash.txt @@ -0,0 +1 @@ +c3068f037595d3dc7ca29f97628506445b72363bc464669645887ead284dff75 \ No newline at end of file diff --git a/samples/timm/cs3darknet_l.c2ns_in1k/graph_net.json b/samples/timm/cs3darknet_l.c2ns_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/cs3darknet_l.c2ns_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cs3darknet_l.c2ns_in1k/input_meta.py b/samples/timm/cs3darknet_l.c2ns_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cs3darknet_l.c2ns_in1k/input_tensor_constraints.py b/samples/timm/cs3darknet_l.c2ns_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cs3darknet_l.c2ns_in1k/model.py b/samples/timm/cs3darknet_l.c2ns_in1k/model.py new file mode 100644 index 0000000000..5dd3a1f134 --- /dev/null +++ b/samples/timm/cs3darknet_l.c2ns_in1k/model.py @@ -0,0 +1,2041 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.silu(x_1, inplace=True) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = None + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = None + x_5 = torch.nn.functional.silu(x_4, inplace=True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_5 = l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_8 = torch.nn.functional.silu(x_7, inplace=True) + x_7 = None + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_8 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_11 = torch.nn.functional.silu(x_10, inplace=True) + x_10 = None + split = x_11.split(64, dim=1) + x_11 = None + x1 = split[0] + x2 = split[1] + split = None + x_12 = torch.conv2d( + x1, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_13 = torch.nn.functional.batch_norm( + x_12, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_14 = torch.nn.functional.silu(x_13, inplace=True) + x_13 = None + x_15 = torch.conv2d( + x_14, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_16 = torch.nn.functional.batch_norm( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_17 = torch.nn.functional.silu(x_16, inplace=True) + x_16 = None + x_18 = x_17 + x1 + x_17 = x1 = None + x_19 = torch.conv2d( + x_18, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_20 = torch.nn.functional.batch_norm( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_19 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_21 = torch.nn.functional.silu(x_20, inplace=True) + x_20 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_23 = torch.nn.functional.batch_norm( + x_22, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_22 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_24 = torch.nn.functional.silu(x_23, inplace=True) + x_23 = None + x_25 = x_24 + x_18 + x_24 = x_18 = None + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_27 = torch.nn.functional.batch_norm( + x_26, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_28 = torch.nn.functional.silu(x_27, inplace=True) + x_27 = None + x_29 = torch.conv2d( + x_28, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_28 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_30 = torch.nn.functional.batch_norm( + x_29, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_31 = torch.nn.functional.silu(x_30, inplace=True) + x_30 = None + x_32 = x_31 + x_25 + x_31 = x_25 = None + cat = torch.cat([x_32, x2], dim=1) + x_32 = x2 = None + x_33 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_34 = torch.nn.functional.batch_norm( + x_33, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_33 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_35 = torch.nn.functional.silu(x_34, inplace=True) + x_34 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_37 = torch.nn.functional.batch_norm( + x_36, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_36 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_38 = torch.nn.functional.silu(x_37, inplace=True) + x_37 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_40 = torch.nn.functional.batch_norm( + x_39, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_39 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_41 = torch.nn.functional.silu(x_40, inplace=True) + x_40 = None + split_1 = x_41.split(128, dim=1) + x_41 = None + x1_1 = split_1[0] + x2_1 = split_1[1] + split_1 = None + x_42 = torch.conv2d( + x1_1, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_43 = torch.nn.functional.batch_norm( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_44 = torch.nn.functional.silu(x_43, inplace=True) + x_43 = None + x_45 = torch.conv2d( + x_44, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_44 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_46 = torch.nn.functional.batch_norm( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_47 = torch.nn.functional.silu(x_46, inplace=True) + x_46 = None + x_48 = x_47 + x1_1 + x_47 = x1_1 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_50 = torch.nn.functional.batch_norm( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_49 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_51 = torch.nn.functional.silu(x_50, inplace=True) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_53 = torch.nn.functional.batch_norm( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_54 = torch.nn.functional.silu(x_53, inplace=True) + x_53 = None + x_55 = x_54 + x_48 + x_54 = x_48 = None + x_56 = torch.conv2d( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_57 = torch.nn.functional.batch_norm( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_58 = torch.nn.functional.silu(x_57, inplace=True) + x_57 = None + x_59 = torch.conv2d( + x_58, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_58 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_60 = torch.nn.functional.batch_norm( + x_59, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_61 = torch.nn.functional.silu(x_60, inplace=True) + x_60 = None + x_62 = x_61 + x_55 + x_61 = x_55 = None + x_63 = torch.conv2d( + x_62, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_64 = torch.nn.functional.batch_norm( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_65 = torch.nn.functional.silu(x_64, inplace=True) + x_64 = None + x_66 = torch.conv2d( + x_65, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_65 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_67 = torch.nn.functional.batch_norm( + x_66, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_66 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_68 = torch.nn.functional.silu(x_67, inplace=True) + x_67 = None + x_69 = x_68 + x_62 + x_68 = x_62 = None + x_70 = torch.conv2d( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_71 = torch.nn.functional.batch_norm( + x_70, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_70 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_72 = torch.nn.functional.silu(x_71, inplace=True) + x_71 = None + x_73 = torch.conv2d( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_74 = torch.nn.functional.batch_norm( + x_73, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_73 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_75 = torch.nn.functional.silu(x_74, inplace=True) + x_74 = None + x_76 = x_75 + x_69 + x_75 = x_69 = None + x_77 = torch.conv2d( + x_76, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_78 = torch.nn.functional.batch_norm( + x_77, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_77 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_79 = torch.nn.functional.silu(x_78, inplace=True) + x_78 = None + x_80 = torch.conv2d( + x_79, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_79 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_81 = torch.nn.functional.batch_norm( + x_80, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_82 = torch.nn.functional.silu(x_81, inplace=True) + x_81 = None + x_83 = x_82 + x_76 + x_82 = x_76 = None + cat_1 = torch.cat([x_83, x2_1], dim=1) + x_83 = x2_1 = None + x_84 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_85 = torch.nn.functional.batch_norm( + x_84, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_84 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_86 = torch.nn.functional.silu(x_85, inplace=True) + x_85 = None + x_87 = torch.conv2d( + x_86, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_86 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_88 = torch.nn.functional.batch_norm( + x_87, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_87 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_89 = torch.nn.functional.silu(x_88, inplace=True) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_91 = torch.nn.functional.batch_norm( + x_90, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_90 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_92 = torch.nn.functional.silu(x_91, inplace=True) + x_91 = None + split_2 = x_92.split(256, dim=1) + x_92 = None + x1_2 = split_2[0] + x2_2 = split_2[1] + split_2 = None + x_93 = torch.conv2d( + x1_2, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_94 = torch.nn.functional.batch_norm( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_95 = torch.nn.functional.silu(x_94, inplace=True) + x_94 = None + x_96 = torch.conv2d( + x_95, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_95 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_97 = torch.nn.functional.batch_norm( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_96 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_98 = torch.nn.functional.silu(x_97, inplace=True) + x_97 = None + x_99 = x_98 + x1_2 + x_98 = x1_2 = None + x_100 = torch.conv2d( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_101 = torch.nn.functional.batch_norm( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_102 = torch.nn.functional.silu(x_101, inplace=True) + x_101 = None + x_103 = torch.conv2d( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_104 = torch.nn.functional.batch_norm( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_105 = torch.nn.functional.silu(x_104, inplace=True) + x_104 = None + x_106 = x_105 + x_99 + x_105 = x_99 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_108 = torch.nn.functional.batch_norm( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_107 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_109 = torch.nn.functional.silu(x_108, inplace=True) + x_108 = None + x_110 = torch.conv2d( + x_109, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_111 = torch.nn.functional.batch_norm( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_112 = torch.nn.functional.silu(x_111, inplace=True) + x_111 = None + x_113 = x_112 + x_106 + x_112 = x_106 = None + x_114 = torch.conv2d( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_115 = torch.nn.functional.batch_norm( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_116 = torch.nn.functional.silu(x_115, inplace=True) + x_115 = None + x_117 = torch.conv2d( + x_116, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_116 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_118 = torch.nn.functional.batch_norm( + x_117, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_117 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_119 = torch.nn.functional.silu(x_118, inplace=True) + x_118 = None + x_120 = x_119 + x_113 + x_119 = x_113 = None + x_121 = torch.conv2d( + x_120, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_122 = torch.nn.functional.batch_norm( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_123 = torch.nn.functional.silu(x_122, inplace=True) + x_122 = None + x_124 = torch.conv2d( + x_123, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_123 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_125 = torch.nn.functional.batch_norm( + x_124, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_124 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_126 = torch.nn.functional.silu(x_125, inplace=True) + x_125 = None + x_127 = x_126 + x_120 + x_126 = x_120 = None + x_128 = torch.conv2d( + x_127, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_129 = torch.nn.functional.batch_norm( + x_128, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_128 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_130 = torch.nn.functional.silu(x_129, inplace=True) + x_129 = None + x_131 = torch.conv2d( + x_130, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_130 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_132 = torch.nn.functional.batch_norm( + x_131, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_133 = torch.nn.functional.silu(x_132, inplace=True) + x_132 = None + x_134 = x_133 + x_127 + x_133 = x_127 = None + x_135 = torch.conv2d( + x_134, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_136 = torch.nn.functional.batch_norm( + x_135, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_135 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_137 = torch.nn.functional.silu(x_136, inplace=True) + x_136 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_139 = torch.nn.functional.batch_norm( + x_138, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_138 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_140 = torch.nn.functional.silu(x_139, inplace=True) + x_139 = None + x_141 = x_140 + x_134 + x_140 = x_134 = None + x_142 = torch.conv2d( + x_141, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_143 = torch.nn.functional.batch_norm( + x_142, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_144 = torch.nn.functional.silu(x_143, inplace=True) + x_143 = None + x_145 = torch.conv2d( + x_144, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_144 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_146 = torch.nn.functional.batch_norm( + x_145, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_145 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_147 = torch.nn.functional.silu(x_146, inplace=True) + x_146 = None + x_148 = x_147 + x_141 + x_147 = x_141 = None + x_149 = torch.conv2d( + x_148, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_150 = torch.nn.functional.batch_norm( + x_149, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_149 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = (None) + x_151 = torch.nn.functional.silu(x_150, inplace=True) + x_150 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_151 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = (None) + x_153 = torch.nn.functional.batch_norm( + x_152, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_152 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = (None) + x_154 = torch.nn.functional.silu(x_153, inplace=True) + x_153 = None + x_155 = x_154 + x_148 + x_154 = x_148 = None + cat_2 = torch.cat([x_155, x2_2], dim=1) + x_155 = x2_2 = None + x_156 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_157 = torch.nn.functional.batch_norm( + x_156, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_156 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_158 = torch.nn.functional.silu(x_157, inplace=True) + x_157 = None + x_159 = torch.conv2d( + x_158, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_158 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_160 = torch.nn.functional.batch_norm( + x_159, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_159 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_161 = torch.nn.functional.silu(x_160, inplace=True) + x_160 = None + x_162 = torch.conv2d( + x_161, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_161 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_163 = torch.nn.functional.batch_norm( + x_162, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_162 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_164 = torch.nn.functional.silu(x_163, inplace=True) + x_163 = None + split_3 = x_164.split(512, dim=1) + x_164 = None + x1_3 = split_3[0] + x2_3 = split_3[1] + split_3 = None + x_165 = torch.conv2d( + x1_3, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_166 = torch.nn.functional.batch_norm( + x_165, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_165 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_167 = torch.nn.functional.silu(x_166, inplace=True) + x_166 = None + x_168 = torch.conv2d( + x_167, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_167 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_169 = torch.nn.functional.batch_norm( + x_168, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_168 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_170 = torch.nn.functional.silu(x_169, inplace=True) + x_169 = None + x_171 = x_170 + x1_3 + x_170 = x1_3 = None + x_172 = torch.conv2d( + x_171, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_173 = torch.nn.functional.batch_norm( + x_172, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_172 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_174 = torch.nn.functional.silu(x_173, inplace=True) + x_173 = None + x_175 = torch.conv2d( + x_174, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_174 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_176 = torch.nn.functional.batch_norm( + x_175, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_175 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_177 = torch.nn.functional.silu(x_176, inplace=True) + x_176 = None + x_178 = x_177 + x_171 + x_177 = x_171 = None + x_179 = torch.conv2d( + x_178, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_180 = torch.nn.functional.batch_norm( + x_179, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_179 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_181 = torch.nn.functional.silu(x_180, inplace=True) + x_180 = None + x_182 = torch.conv2d( + x_181, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_181 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_183 = torch.nn.functional.batch_norm( + x_182, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_182 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_184 = torch.nn.functional.silu(x_183, inplace=True) + x_183 = None + x_185 = x_184 + x_178 + x_184 = x_178 = None + cat_3 = torch.cat([x_185, x2_3], dim=1) + x_185 = x2_3 = None + x_186 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_187 = torch.nn.functional.batch_norm( + x_186, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_186 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_188 = torch.nn.functional.silu(x_187, inplace=True) + x_187 = None + x_189 = torch.nn.functional.adaptive_avg_pool2d(x_188, 1) + x_188 = None + x_190 = x_189.flatten(1, -1) + x_189 = None + x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) + x_190 = None + x_192 = torch._C._nn.linear( + x_191, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_191 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_192,) diff --git a/samples/timm/cs3darknet_l.c2ns_in1k/weight_meta.py b/samples/timm/cs3darknet_l.c2ns_in1k/weight_meta.py new file mode 100644 index 0000000000..a4a7a3daa6 --- /dev/null +++ b/samples/timm/cs3darknet_l.c2ns_in1k/weight_meta.py @@ -0,0 +1,2862 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [32, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.085 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.226 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.123 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.060 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.004 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.005 + std = 0.179 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.124 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.124 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [1024, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cs3darknet_m.c2ns_in1k/graph_hash.txt b/samples/timm/cs3darknet_m.c2ns_in1k/graph_hash.txt new file mode 100644 index 0000000000..8df2869197 --- /dev/null +++ b/samples/timm/cs3darknet_m.c2ns_in1k/graph_hash.txt @@ -0,0 +1 @@ +cebfb5f58a4ad3894476513d82b2d34bd5c612a4662c6224d95dfb054105f459 \ No newline at end of file diff --git a/samples/timm/cs3darknet_m.c2ns_in1k/graph_net.json b/samples/timm/cs3darknet_m.c2ns_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/cs3darknet_m.c2ns_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cs3darknet_m.c2ns_in1k/input_meta.py b/samples/timm/cs3darknet_m.c2ns_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cs3darknet_m.c2ns_in1k/input_tensor_constraints.py b/samples/timm/cs3darknet_m.c2ns_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cs3darknet_m.c2ns_in1k/model.py b/samples/timm/cs3darknet_m.c2ns_in1k/model.py new file mode 100644 index 0000000000..0a058a1a8f --- /dev/null +++ b/samples/timm/cs3darknet_m.c2ns_in1k/model.py @@ -0,0 +1,1551 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.silu(x_1, inplace=True) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = None + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = None + x_5 = torch.nn.functional.silu(x_4, inplace=True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_5 = l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_8 = torch.nn.functional.silu(x_7, inplace=True) + x_7 = None + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_8 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_11 = torch.nn.functional.silu(x_10, inplace=True) + x_10 = None + split = x_11.split(48, dim=1) + x_11 = None + x1 = split[0] + x2 = split[1] + split = None + x_12 = torch.conv2d( + x1, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_13 = torch.nn.functional.batch_norm( + x_12, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_14 = torch.nn.functional.silu(x_13, inplace=True) + x_13 = None + x_15 = torch.conv2d( + x_14, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_16 = torch.nn.functional.batch_norm( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_17 = torch.nn.functional.silu(x_16, inplace=True) + x_16 = None + x_18 = x_17 + x1 + x_17 = x1 = None + x_19 = torch.conv2d( + x_18, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_20 = torch.nn.functional.batch_norm( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_19 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_21 = torch.nn.functional.silu(x_20, inplace=True) + x_20 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_23 = torch.nn.functional.batch_norm( + x_22, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_22 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_24 = torch.nn.functional.silu(x_23, inplace=True) + x_23 = None + x_25 = x_24 + x_18 + x_24 = x_18 = None + cat = torch.cat([x_25, x2], dim=1) + x_25 = x2 = None + x_26 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_27 = torch.nn.functional.batch_norm( + x_26, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_26 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_28 = torch.nn.functional.silu(x_27, inplace=True) + x_27 = None + x_29 = torch.conv2d( + x_28, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_28 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_30 = torch.nn.functional.batch_norm( + x_29, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_29 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_31 = torch.nn.functional.silu(x_30, inplace=True) + x_30 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_33 = torch.nn.functional.batch_norm( + x_32, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_32 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_34 = torch.nn.functional.silu(x_33, inplace=True) + x_33 = None + split_1 = x_34.split(96, dim=1) + x_34 = None + x1_1 = split_1[0] + x2_1 = split_1[1] + split_1 = None + x_35 = torch.conv2d( + x1_1, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_36 = torch.nn.functional.batch_norm( + x_35, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_35 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_37 = torch.nn.functional.silu(x_36, inplace=True) + x_36 = None + x_38 = torch.conv2d( + x_37, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_37 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_39 = torch.nn.functional.batch_norm( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_38 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_40 = torch.nn.functional.silu(x_39, inplace=True) + x_39 = None + x_41 = x_40 + x1_1 + x_40 = x1_1 = None + x_42 = torch.conv2d( + x_41, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_43 = torch.nn.functional.batch_norm( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_44 = torch.nn.functional.silu(x_43, inplace=True) + x_43 = None + x_45 = torch.conv2d( + x_44, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_44 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_46 = torch.nn.functional.batch_norm( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_47 = torch.nn.functional.silu(x_46, inplace=True) + x_46 = None + x_48 = x_47 + x_41 + x_47 = x_41 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_50 = torch.nn.functional.batch_norm( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_49 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_51 = torch.nn.functional.silu(x_50, inplace=True) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_53 = torch.nn.functional.batch_norm( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_54 = torch.nn.functional.silu(x_53, inplace=True) + x_53 = None + x_55 = x_54 + x_48 + x_54 = x_48 = None + x_56 = torch.conv2d( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_57 = torch.nn.functional.batch_norm( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_58 = torch.nn.functional.silu(x_57, inplace=True) + x_57 = None + x_59 = torch.conv2d( + x_58, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_58 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_60 = torch.nn.functional.batch_norm( + x_59, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_61 = torch.nn.functional.silu(x_60, inplace=True) + x_60 = None + x_62 = x_61 + x_55 + x_61 = x_55 = None + cat_1 = torch.cat([x_62, x2_1], dim=1) + x_62 = x2_1 = None + x_63 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_64 = torch.nn.functional.batch_norm( + x_63, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_63 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_65 = torch.nn.functional.silu(x_64, inplace=True) + x_64 = None + x_66 = torch.conv2d( + x_65, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_65 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_67 = torch.nn.functional.batch_norm( + x_66, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_66 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_68 = torch.nn.functional.silu(x_67, inplace=True) + x_67 = None + x_69 = torch.conv2d( + x_68, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_68 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_70 = torch.nn.functional.batch_norm( + x_69, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_69 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_71 = torch.nn.functional.silu(x_70, inplace=True) + x_70 = None + split_2 = x_71.split(192, dim=1) + x_71 = None + x1_2 = split_2[0] + x2_2 = split_2[1] + split_2 = None + x_72 = torch.conv2d( + x1_2, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_73 = torch.nn.functional.batch_norm( + x_72, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_72 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_74 = torch.nn.functional.silu(x_73, inplace=True) + x_73 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_74 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_76 = torch.nn.functional.batch_norm( + x_75, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_75 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_77 = torch.nn.functional.silu(x_76, inplace=True) + x_76 = None + x_78 = x_77 + x1_2 + x_77 = x1_2 = None + x_79 = torch.conv2d( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_80 = torch.nn.functional.batch_norm( + x_79, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_79 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_81 = torch.nn.functional.silu(x_80, inplace=True) + x_80 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_83 = torch.nn.functional.batch_norm( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_82 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_84 = torch.nn.functional.silu(x_83, inplace=True) + x_83 = None + x_85 = x_84 + x_78 + x_84 = x_78 = None + x_86 = torch.conv2d( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_87 = torch.nn.functional.batch_norm( + x_86, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_86 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_88 = torch.nn.functional.silu(x_87, inplace=True) + x_87 = None + x_89 = torch.conv2d( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_88 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_90 = torch.nn.functional.batch_norm( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_91 = torch.nn.functional.silu(x_90, inplace=True) + x_90 = None + x_92 = x_91 + x_85 + x_91 = x_85 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_94 = torch.nn.functional.batch_norm( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_95 = torch.nn.functional.silu(x_94, inplace=True) + x_94 = None + x_96 = torch.conv2d( + x_95, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_95 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_97 = torch.nn.functional.batch_norm( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_96 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_98 = torch.nn.functional.silu(x_97, inplace=True) + x_97 = None + x_99 = x_98 + x_92 + x_98 = x_92 = None + x_100 = torch.conv2d( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_101 = torch.nn.functional.batch_norm( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_102 = torch.nn.functional.silu(x_101, inplace=True) + x_101 = None + x_103 = torch.conv2d( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_104 = torch.nn.functional.batch_norm( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_105 = torch.nn.functional.silu(x_104, inplace=True) + x_104 = None + x_106 = x_105 + x_99 + x_105 = x_99 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_108 = torch.nn.functional.batch_norm( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_107 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_109 = torch.nn.functional.silu(x_108, inplace=True) + x_108 = None + x_110 = torch.conv2d( + x_109, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_111 = torch.nn.functional.batch_norm( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_112 = torch.nn.functional.silu(x_111, inplace=True) + x_111 = None + x_113 = x_112 + x_106 + x_112 = x_106 = None + cat_2 = torch.cat([x_113, x2_2], dim=1) + x_113 = x2_2 = None + x_114 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_115 = torch.nn.functional.batch_norm( + x_114, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_114 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_116 = torch.nn.functional.silu(x_115, inplace=True) + x_115 = None + x_117 = torch.conv2d( + x_116, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_116 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_118 = torch.nn.functional.batch_norm( + x_117, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_117 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_119 = torch.nn.functional.silu(x_118, inplace=True) + x_118 = None + x_120 = torch.conv2d( + x_119, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_119 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_121 = torch.nn.functional.batch_norm( + x_120, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_120 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_122 = torch.nn.functional.silu(x_121, inplace=True) + x_121 = None + split_3 = x_122.split(384, dim=1) + x_122 = None + x1_3 = split_3[0] + x2_3 = split_3[1] + split_3 = None + x_123 = torch.conv2d( + x1_3, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_124 = torch.nn.functional.batch_norm( + x_123, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_125 = torch.nn.functional.silu(x_124, inplace=True) + x_124 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_125 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_127 = torch.nn.functional.batch_norm( + x_126, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_126 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_128 = torch.nn.functional.silu(x_127, inplace=True) + x_127 = None + x_129 = x_128 + x1_3 + x_128 = x1_3 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_131 = torch.nn.functional.batch_norm( + x_130, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_130 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_132 = torch.nn.functional.silu(x_131, inplace=True) + x_131 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_132 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_134 = torch.nn.functional.batch_norm( + x_133, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_133 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_135 = torch.nn.functional.silu(x_134, inplace=True) + x_134 = None + x_136 = x_135 + x_129 + x_135 = x_129 = None + cat_3 = torch.cat([x_136, x2_3], dim=1) + x_136 = x2_3 = None + x_137 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_138 = torch.nn.functional.batch_norm( + x_137, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_137 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_139 = torch.nn.functional.silu(x_138, inplace=True) + x_138 = None + x_140 = torch.nn.functional.adaptive_avg_pool2d(x_139, 1) + x_139 = None + x_141 = x_140.flatten(1, -1) + x_140 = None + x_142 = torch.nn.functional.dropout(x_141, 0.0, False, False) + x_141 = None + x_143 = torch._C._nn.linear( + x_142, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_142 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_143,) diff --git a/samples/timm/cs3darknet_m.c2ns_in1k/weight_meta.py b/samples/timm/cs3darknet_m.c2ns_in1k/weight_meta.py new file mode 100644 index 0000000000..b5cf320a17 --- /dev/null +++ b/samples/timm/cs3darknet_m.c2ns_in1k/weight_meta.py @@ -0,0 +1,2162 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [24, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.092 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.227 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_" + shape = [48, 24, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.069 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_" + shape = [48, 48, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.068 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [96, 48, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.141 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [48, 48, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.005 + std = 0.203 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [48, 48, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.068 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [48, 48, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.207 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [48, 48, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.068 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [48] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [96, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.143 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [96, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [192, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [96, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.144 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [96, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [96, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.145 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [96, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [96, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.144 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [96, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [96, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.144 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [96, 96, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [384, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [192, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [384, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [384, 384, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [768, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [384, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [384, 384, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [384, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [384, 384, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [768, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cs3darknet_x.c2ns_in1k/graph_hash.txt b/samples/timm/cs3darknet_x.c2ns_in1k/graph_hash.txt new file mode 100644 index 0000000000..8a4e633649 --- /dev/null +++ b/samples/timm/cs3darknet_x.c2ns_in1k/graph_hash.txt @@ -0,0 +1 @@ +2d3ba3b73cded31585e47c5b7453f289cd329b85d1f959c316adaa2d4b0cbc92 \ No newline at end of file diff --git a/samples/timm/cs3darknet_x.c2ns_in1k/graph_net.json b/samples/timm/cs3darknet_x.c2ns_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/cs3darknet_x.c2ns_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cs3darknet_x.c2ns_in1k/input_meta.py b/samples/timm/cs3darknet_x.c2ns_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cs3darknet_x.c2ns_in1k/input_tensor_constraints.py b/samples/timm/cs3darknet_x.c2ns_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cs3darknet_x.c2ns_in1k/model.py b/samples/timm/cs3darknet_x.c2ns_in1k/model.py new file mode 100644 index 0000000000..4b502d1754 --- /dev/null +++ b/samples/timm/cs3darknet_x.c2ns_in1k/model.py @@ -0,0 +1,2251 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.silu(x_1, inplace=True) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = None + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = None + x_5 = torch.nn.functional.silu(x_4, inplace=True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_5 = l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_8 = torch.nn.functional.silu(x_7, inplace=True) + x_7 = None + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_8 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_11 = torch.nn.functional.silu(x_10, inplace=True) + x_10 = None + split = x_11.split(80, dim=1) + x_11 = None + x1 = split[0] + x2 = split[1] + split = None + x_12 = torch.conv2d( + x1, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_13 = torch.nn.functional.batch_norm( + x_12, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_14 = torch.nn.functional.silu(x_13, inplace=True) + x_13 = None + x_15 = torch.conv2d( + x_14, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_16 = torch.nn.functional.batch_norm( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_17 = torch.nn.functional.silu(x_16, inplace=True) + x_16 = None + x_18 = x_17 + x1 + x_17 = x1 = None + x_19 = torch.conv2d( + x_18, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_20 = torch.nn.functional.batch_norm( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_19 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_21 = torch.nn.functional.silu(x_20, inplace=True) + x_20 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_23 = torch.nn.functional.batch_norm( + x_22, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_22 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_24 = torch.nn.functional.silu(x_23, inplace=True) + x_23 = None + x_25 = x_24 + x_18 + x_24 = x_18 = None + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_27 = torch.nn.functional.batch_norm( + x_26, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_28 = torch.nn.functional.silu(x_27, inplace=True) + x_27 = None + x_29 = torch.conv2d( + x_28, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_28 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_30 = torch.nn.functional.batch_norm( + x_29, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_31 = torch.nn.functional.silu(x_30, inplace=True) + x_30 = None + x_32 = x_31 + x_25 + x_31 = x_25 = None + cat = torch.cat([x_32, x2], dim=1) + x_32 = x2 = None + x_33 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_34 = torch.nn.functional.batch_norm( + x_33, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_33 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_35 = torch.nn.functional.silu(x_34, inplace=True) + x_34 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_37 = torch.nn.functional.batch_norm( + x_36, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_36 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_38 = torch.nn.functional.silu(x_37, inplace=True) + x_37 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_40 = torch.nn.functional.batch_norm( + x_39, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_39 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_41 = torch.nn.functional.silu(x_40, inplace=True) + x_40 = None + split_1 = x_41.split(160, dim=1) + x_41 = None + x1_1 = split_1[0] + x2_1 = split_1[1] + split_1 = None + x_42 = torch.conv2d( + x1_1, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_43 = torch.nn.functional.batch_norm( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_44 = torch.nn.functional.silu(x_43, inplace=True) + x_43 = None + x_45 = torch.conv2d( + x_44, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_44 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_46 = torch.nn.functional.batch_norm( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_47 = torch.nn.functional.silu(x_46, inplace=True) + x_46 = None + x_48 = x_47 + x1_1 + x_47 = x1_1 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_50 = torch.nn.functional.batch_norm( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_49 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_51 = torch.nn.functional.silu(x_50, inplace=True) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_53 = torch.nn.functional.batch_norm( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_54 = torch.nn.functional.silu(x_53, inplace=True) + x_53 = None + x_55 = x_54 + x_48 + x_54 = x_48 = None + x_56 = torch.conv2d( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_57 = torch.nn.functional.batch_norm( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_58 = torch.nn.functional.silu(x_57, inplace=True) + x_57 = None + x_59 = torch.conv2d( + x_58, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_58 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_60 = torch.nn.functional.batch_norm( + x_59, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_61 = torch.nn.functional.silu(x_60, inplace=True) + x_60 = None + x_62 = x_61 + x_55 + x_61 = x_55 = None + x_63 = torch.conv2d( + x_62, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_64 = torch.nn.functional.batch_norm( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_65 = torch.nn.functional.silu(x_64, inplace=True) + x_64 = None + x_66 = torch.conv2d( + x_65, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_65 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_67 = torch.nn.functional.batch_norm( + x_66, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_66 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_68 = torch.nn.functional.silu(x_67, inplace=True) + x_67 = None + x_69 = x_68 + x_62 + x_68 = x_62 = None + x_70 = torch.conv2d( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_71 = torch.nn.functional.batch_norm( + x_70, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_70 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_72 = torch.nn.functional.silu(x_71, inplace=True) + x_71 = None + x_73 = torch.conv2d( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_74 = torch.nn.functional.batch_norm( + x_73, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_73 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_75 = torch.nn.functional.silu(x_74, inplace=True) + x_74 = None + x_76 = x_75 + x_69 + x_75 = x_69 = None + x_77 = torch.conv2d( + x_76, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_78 = torch.nn.functional.batch_norm( + x_77, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_77 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_79 = torch.nn.functional.silu(x_78, inplace=True) + x_78 = None + x_80 = torch.conv2d( + x_79, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_79 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_81 = torch.nn.functional.batch_norm( + x_80, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_82 = torch.nn.functional.silu(x_81, inplace=True) + x_81 = None + x_83 = x_82 + x_76 + x_82 = x_76 = None + x_84 = torch.conv2d( + x_83, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_85 = torch.nn.functional.batch_norm( + x_84, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_86 = torch.nn.functional.silu(x_85, inplace=True) + x_85 = None + x_87 = torch.conv2d( + x_86, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_88 = torch.nn.functional.batch_norm( + x_87, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_87 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_89 = torch.nn.functional.silu(x_88, inplace=True) + x_88 = None + x_90 = x_89 + x_83 + x_89 = x_83 = None + cat_1 = torch.cat([x_90, x2_1], dim=1) + x_90 = x2_1 = None + x_91 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_92 = torch.nn.functional.batch_norm( + x_91, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_91 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_93 = torch.nn.functional.silu(x_92, inplace=True) + x_92 = None + x_94 = torch.conv2d( + x_93, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_93 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_95 = torch.nn.functional.batch_norm( + x_94, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_94 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_96 = torch.nn.functional.silu(x_95, inplace=True) + x_95 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_96 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_98 = torch.nn.functional.batch_norm( + x_97, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_97 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_99 = torch.nn.functional.silu(x_98, inplace=True) + x_98 = None + split_2 = x_99.split(320, dim=1) + x_99 = None + x1_2 = split_2[0] + x2_2 = split_2[1] + split_2 = None + x_100 = torch.conv2d( + x1_2, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_101 = torch.nn.functional.batch_norm( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_102 = torch.nn.functional.silu(x_101, inplace=True) + x_101 = None + x_103 = torch.conv2d( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_104 = torch.nn.functional.batch_norm( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_105 = torch.nn.functional.silu(x_104, inplace=True) + x_104 = None + x_106 = x_105 + x1_2 + x_105 = x1_2 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_108 = torch.nn.functional.batch_norm( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_107 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_109 = torch.nn.functional.silu(x_108, inplace=True) + x_108 = None + x_110 = torch.conv2d( + x_109, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_111 = torch.nn.functional.batch_norm( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_112 = torch.nn.functional.silu(x_111, inplace=True) + x_111 = None + x_113 = x_112 + x_106 + x_112 = x_106 = None + x_114 = torch.conv2d( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_115 = torch.nn.functional.batch_norm( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_116 = torch.nn.functional.silu(x_115, inplace=True) + x_115 = None + x_117 = torch.conv2d( + x_116, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_116 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_118 = torch.nn.functional.batch_norm( + x_117, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_117 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_119 = torch.nn.functional.silu(x_118, inplace=True) + x_118 = None + x_120 = x_119 + x_113 + x_119 = x_113 = None + x_121 = torch.conv2d( + x_120, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_122 = torch.nn.functional.batch_norm( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_123 = torch.nn.functional.silu(x_122, inplace=True) + x_122 = None + x_124 = torch.conv2d( + x_123, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_123 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_125 = torch.nn.functional.batch_norm( + x_124, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_124 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_126 = torch.nn.functional.silu(x_125, inplace=True) + x_125 = None + x_127 = x_126 + x_120 + x_126 = x_120 = None + x_128 = torch.conv2d( + x_127, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_129 = torch.nn.functional.batch_norm( + x_128, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_128 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_130 = torch.nn.functional.silu(x_129, inplace=True) + x_129 = None + x_131 = torch.conv2d( + x_130, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_130 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_132 = torch.nn.functional.batch_norm( + x_131, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_133 = torch.nn.functional.silu(x_132, inplace=True) + x_132 = None + x_134 = x_133 + x_127 + x_133 = x_127 = None + x_135 = torch.conv2d( + x_134, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_136 = torch.nn.functional.batch_norm( + x_135, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_135 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_137 = torch.nn.functional.silu(x_136, inplace=True) + x_136 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_139 = torch.nn.functional.batch_norm( + x_138, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_138 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_140 = torch.nn.functional.silu(x_139, inplace=True) + x_139 = None + x_141 = x_140 + x_134 + x_140 = x_134 = None + x_142 = torch.conv2d( + x_141, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_143 = torch.nn.functional.batch_norm( + x_142, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_144 = torch.nn.functional.silu(x_143, inplace=True) + x_143 = None + x_145 = torch.conv2d( + x_144, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_144 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_146 = torch.nn.functional.batch_norm( + x_145, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_145 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_147 = torch.nn.functional.silu(x_146, inplace=True) + x_146 = None + x_148 = x_147 + x_141 + x_147 = x_141 = None + x_149 = torch.conv2d( + x_148, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_150 = torch.nn.functional.batch_norm( + x_149, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_149 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_151 = torch.nn.functional.silu(x_150, inplace=True) + x_150 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_151 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_153 = torch.nn.functional.batch_norm( + x_152, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_152 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_154 = torch.nn.functional.silu(x_153, inplace=True) + x_153 = None + x_155 = x_154 + x_148 + x_154 = x_148 = None + x_156 = torch.conv2d( + x_155, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_157 = torch.nn.functional.batch_norm( + x_156, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_156 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = (None) + x_158 = torch.nn.functional.silu(x_157, inplace=True) + x_157 = None + x_159 = torch.conv2d( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = (None) + x_160 = torch.nn.functional.batch_norm( + x_159, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_159 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = (None) + x_161 = torch.nn.functional.silu(x_160, inplace=True) + x_160 = None + x_162 = x_161 + x_155 + x_161 = x_155 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_164 = torch.nn.functional.batch_norm( + x_163, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_163 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_ = (None) + x_165 = torch.nn.functional.silu(x_164, inplace=True) + x_164 = None + x_166 = torch.conv2d( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_ = (None) + x_167 = torch.nn.functional.batch_norm( + x_166, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_166 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_ = (None) + x_168 = torch.nn.functional.silu(x_167, inplace=True) + x_167 = None + x_169 = x_168 + x_162 + x_168 = x_162 = None + x_170 = torch.conv2d( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_171 = torch.nn.functional.batch_norm( + x_170, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_170 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_ = (None) + x_172 = torch.nn.functional.silu(x_171, inplace=True) + x_171 = None + x_173 = torch.conv2d( + x_172, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_172 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_ = (None) + x_174 = torch.nn.functional.batch_norm( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_173 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_ = (None) + x_175 = torch.nn.functional.silu(x_174, inplace=True) + x_174 = None + x_176 = x_175 + x_169 + x_175 = x_169 = None + cat_2 = torch.cat([x_176, x2_2], dim=1) + x_176 = x2_2 = None + x_177 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_178 = torch.nn.functional.batch_norm( + x_177, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_177 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_179 = torch.nn.functional.silu(x_178, inplace=True) + x_178 = None + x_180 = torch.conv2d( + x_179, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_179 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_181 = torch.nn.functional.batch_norm( + x_180, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_180 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_182 = torch.nn.functional.silu(x_181, inplace=True) + x_181 = None + x_183 = torch.conv2d( + x_182, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_182 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_184 = torch.nn.functional.batch_norm( + x_183, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_183 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_185 = torch.nn.functional.silu(x_184, inplace=True) + x_184 = None + split_3 = x_185.split(640, dim=1) + x_185 = None + x1_3 = split_3[0] + x2_3 = split_3[1] + split_3 = None + x_186 = torch.conv2d( + x1_3, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_187 = torch.nn.functional.batch_norm( + x_186, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_186 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_188 = torch.nn.functional.silu(x_187, inplace=True) + x_187 = None + x_189 = torch.conv2d( + x_188, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_188 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_190 = torch.nn.functional.batch_norm( + x_189, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_189 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_191 = torch.nn.functional.silu(x_190, inplace=True) + x_190 = None + x_192 = x_191 + x1_3 + x_191 = x1_3 = None + x_193 = torch.conv2d( + x_192, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_194 = torch.nn.functional.batch_norm( + x_193, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_193 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_195 = torch.nn.functional.silu(x_194, inplace=True) + x_194 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_195 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_197 = torch.nn.functional.batch_norm( + x_196, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_196 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_198 = torch.nn.functional.silu(x_197, inplace=True) + x_197 = None + x_199 = x_198 + x_192 + x_198 = x_192 = None + x_200 = torch.conv2d( + x_199, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_201 = torch.nn.functional.batch_norm( + x_200, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_200 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_202 = torch.nn.functional.silu(x_201, inplace=True) + x_201 = None + x_203 = torch.conv2d( + x_202, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_202 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_204 = torch.nn.functional.batch_norm( + x_203, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_203 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_205 = torch.nn.functional.silu(x_204, inplace=True) + x_204 = None + x_206 = x_205 + x_199 + x_205 = x_199 = None + cat_3 = torch.cat([x_206, x2_3], dim=1) + x_206 = x2_3 = None + x_207 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_208 = torch.nn.functional.batch_norm( + x_207, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_207 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_209 = torch.nn.functional.silu(x_208, inplace=True) + x_208 = None + x_210 = torch.nn.functional.adaptive_avg_pool2d(x_209, 1) + x_209 = None + x_211 = x_210.flatten(1, -1) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_213 = torch._C._nn.linear( + x_212, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_212 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_213,) diff --git a/samples/timm/cs3darknet_x.c2ns_in1k/weight_meta.py b/samples/timm/cs3darknet_x.c2ns_in1k/weight_meta.py new file mode 100644 index 0000000000..c0621a0609 --- /dev/null +++ b/samples/timm/cs3darknet_x.c2ns_in1k/weight_meta.py @@ -0,0 +1,3162 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [40, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.077 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.231 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 40, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_" + shape = [80, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.052 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [160, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.113 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [80, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.157 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [80, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.160 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [80, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.157 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [320, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.113 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.113 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [640, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [640, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [640, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [1280, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [640, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [640, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [640, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [640, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [640, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [640, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [1280, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cs3edgenet_x.c2_in1k/graph_hash.txt b/samples/timm/cs3edgenet_x.c2_in1k/graph_hash.txt new file mode 100644 index 0000000000..d019171a0a --- /dev/null +++ b/samples/timm/cs3edgenet_x.c2_in1k/graph_hash.txt @@ -0,0 +1 @@ +7a5a8bd289abb7ffad2c153c67f5dfa9567d4efcf89bf2788d22265ba9030dcd \ No newline at end of file diff --git a/samples/timm/cs3edgenet_x.c2_in1k/graph_net.json b/samples/timm/cs3edgenet_x.c2_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/cs3edgenet_x.c2_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cs3edgenet_x.c2_in1k/input_meta.py b/samples/timm/cs3edgenet_x.c2_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cs3edgenet_x.c2_in1k/input_tensor_constraints.py b/samples/timm/cs3edgenet_x.c2_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cs3edgenet_x.c2_in1k/model.py b/samples/timm/cs3edgenet_x.c2_in1k/model.py new file mode 100644 index 0000000000..611d6b7882 --- /dev/null +++ b/samples/timm/cs3edgenet_x.c2_in1k/model.py @@ -0,0 +1,2251 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.silu(x_1, inplace=True) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = None + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = None + x_5 = torch.nn.functional.silu(x_4, inplace=True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_5 = l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_8 = torch.nn.functional.silu(x_7, inplace=True) + x_7 = None + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_8 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_11 = torch.nn.functional.silu(x_10, inplace=True) + x_10 = None + split = x_11.split(80, dim=1) + x_11 = None + x1 = split[0] + x2 = split[1] + split = None + x_12 = torch.conv2d( + x1, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_13 = torch.nn.functional.batch_norm( + x_12, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_14 = torch.nn.functional.silu(x_13, inplace=True) + x_13 = None + x_15 = torch.conv2d( + x_14, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_14 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_16 = torch.nn.functional.batch_norm( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_17 = torch.nn.functional.silu(x_16, inplace=True) + x_16 = None + x_18 = x_17 + x1 + x_17 = x1 = None + x_19 = torch.conv2d( + x_18, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_20 = torch.nn.functional.batch_norm( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_19 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_21 = torch.nn.functional.silu(x_20, inplace=True) + x_20 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_23 = torch.nn.functional.batch_norm( + x_22, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_22 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_24 = torch.nn.functional.silu(x_23, inplace=True) + x_23 = None + x_25 = x_24 + x_18 + x_24 = x_18 = None + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_27 = torch.nn.functional.batch_norm( + x_26, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_26 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_28 = torch.nn.functional.silu(x_27, inplace=True) + x_27 = None + x_29 = torch.conv2d( + x_28, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_28 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_30 = torch.nn.functional.batch_norm( + x_29, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_29 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_31 = torch.nn.functional.silu(x_30, inplace=True) + x_30 = None + x_32 = x_31 + x_25 + x_31 = x_25 = None + cat = torch.cat([x_32, x2], dim=1) + x_32 = x2 = None + x_33 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_34 = torch.nn.functional.batch_norm( + x_33, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_33 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_35 = torch.nn.functional.silu(x_34, inplace=True) + x_34 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_37 = torch.nn.functional.batch_norm( + x_36, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_36 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_38 = torch.nn.functional.silu(x_37, inplace=True) + x_37 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_40 = torch.nn.functional.batch_norm( + x_39, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_39 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_41 = torch.nn.functional.silu(x_40, inplace=True) + x_40 = None + split_1 = x_41.split(160, dim=1) + x_41 = None + x1_1 = split_1[0] + x2_1 = split_1[1] + split_1 = None + x_42 = torch.conv2d( + x1_1, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_43 = torch.nn.functional.batch_norm( + x_42, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_42 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_44 = torch.nn.functional.silu(x_43, inplace=True) + x_43 = None + x_45 = torch.conv2d( + x_44, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_44 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_46 = torch.nn.functional.batch_norm( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_47 = torch.nn.functional.silu(x_46, inplace=True) + x_46 = None + x_48 = x_47 + x1_1 + x_47 = x1_1 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_50 = torch.nn.functional.batch_norm( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_49 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_51 = torch.nn.functional.silu(x_50, inplace=True) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_53 = torch.nn.functional.batch_norm( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_54 = torch.nn.functional.silu(x_53, inplace=True) + x_53 = None + x_55 = x_54 + x_48 + x_54 = x_48 = None + x_56 = torch.conv2d( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_57 = torch.nn.functional.batch_norm( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_58 = torch.nn.functional.silu(x_57, inplace=True) + x_57 = None + x_59 = torch.conv2d( + x_58, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_58 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_60 = torch.nn.functional.batch_norm( + x_59, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_61 = torch.nn.functional.silu(x_60, inplace=True) + x_60 = None + x_62 = x_61 + x_55 + x_61 = x_55 = None + x_63 = torch.conv2d( + x_62, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_64 = torch.nn.functional.batch_norm( + x_63, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_63 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_65 = torch.nn.functional.silu(x_64, inplace=True) + x_64 = None + x_66 = torch.conv2d( + x_65, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_65 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_67 = torch.nn.functional.batch_norm( + x_66, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_66 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_68 = torch.nn.functional.silu(x_67, inplace=True) + x_67 = None + x_69 = x_68 + x_62 + x_68 = x_62 = None + x_70 = torch.conv2d( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_71 = torch.nn.functional.batch_norm( + x_70, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_70 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_72 = torch.nn.functional.silu(x_71, inplace=True) + x_71 = None + x_73 = torch.conv2d( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_74 = torch.nn.functional.batch_norm( + x_73, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_73 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_75 = torch.nn.functional.silu(x_74, inplace=True) + x_74 = None + x_76 = x_75 + x_69 + x_75 = x_69 = None + x_77 = torch.conv2d( + x_76, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_78 = torch.nn.functional.batch_norm( + x_77, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_77 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_79 = torch.nn.functional.silu(x_78, inplace=True) + x_78 = None + x_80 = torch.conv2d( + x_79, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_79 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_81 = torch.nn.functional.batch_norm( + x_80, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_82 = torch.nn.functional.silu(x_81, inplace=True) + x_81 = None + x_83 = x_82 + x_76 + x_82 = x_76 = None + x_84 = torch.conv2d( + x_83, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_85 = torch.nn.functional.batch_norm( + x_84, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_84 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_86 = torch.nn.functional.silu(x_85, inplace=True) + x_85 = None + x_87 = torch.conv2d( + x_86, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_86 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_88 = torch.nn.functional.batch_norm( + x_87, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_87 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_89 = torch.nn.functional.silu(x_88, inplace=True) + x_88 = None + x_90 = x_89 + x_83 + x_89 = x_83 = None + cat_1 = torch.cat([x_90, x2_1], dim=1) + x_90 = x2_1 = None + x_91 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_92 = torch.nn.functional.batch_norm( + x_91, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_91 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_93 = torch.nn.functional.silu(x_92, inplace=True) + x_92 = None + x_94 = torch.conv2d( + x_93, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_93 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_95 = torch.nn.functional.batch_norm( + x_94, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_94 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_96 = torch.nn.functional.silu(x_95, inplace=True) + x_95 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_96 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_98 = torch.nn.functional.batch_norm( + x_97, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_97 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_99 = torch.nn.functional.silu(x_98, inplace=True) + x_98 = None + split_2 = x_99.split(320, dim=1) + x_99 = None + x1_2 = split_2[0] + x2_2 = split_2[1] + split_2 = None + x_100 = torch.conv2d( + x1_2, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_101 = torch.nn.functional.batch_norm( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_102 = torch.nn.functional.silu(x_101, inplace=True) + x_101 = None + x_103 = torch.conv2d( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_104 = torch.nn.functional.batch_norm( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_105 = torch.nn.functional.silu(x_104, inplace=True) + x_104 = None + x_106 = x_105 + x1_2 + x_105 = x1_2 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_108 = torch.nn.functional.batch_norm( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_107 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_109 = torch.nn.functional.silu(x_108, inplace=True) + x_108 = None + x_110 = torch.conv2d( + x_109, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_111 = torch.nn.functional.batch_norm( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_112 = torch.nn.functional.silu(x_111, inplace=True) + x_111 = None + x_113 = x_112 + x_106 + x_112 = x_106 = None + x_114 = torch.conv2d( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_115 = torch.nn.functional.batch_norm( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_116 = torch.nn.functional.silu(x_115, inplace=True) + x_115 = None + x_117 = torch.conv2d( + x_116, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_116 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_118 = torch.nn.functional.batch_norm( + x_117, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_117 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_119 = torch.nn.functional.silu(x_118, inplace=True) + x_118 = None + x_120 = x_119 + x_113 + x_119 = x_113 = None + x_121 = torch.conv2d( + x_120, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_122 = torch.nn.functional.batch_norm( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_123 = torch.nn.functional.silu(x_122, inplace=True) + x_122 = None + x_124 = torch.conv2d( + x_123, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_123 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_125 = torch.nn.functional.batch_norm( + x_124, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_124 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_126 = torch.nn.functional.silu(x_125, inplace=True) + x_125 = None + x_127 = x_126 + x_120 + x_126 = x_120 = None + x_128 = torch.conv2d( + x_127, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_129 = torch.nn.functional.batch_norm( + x_128, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_128 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_130 = torch.nn.functional.silu(x_129, inplace=True) + x_129 = None + x_131 = torch.conv2d( + x_130, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_130 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_132 = torch.nn.functional.batch_norm( + x_131, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_131 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_133 = torch.nn.functional.silu(x_132, inplace=True) + x_132 = None + x_134 = x_133 + x_127 + x_133 = x_127 = None + x_135 = torch.conv2d( + x_134, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_136 = torch.nn.functional.batch_norm( + x_135, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_135 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_137 = torch.nn.functional.silu(x_136, inplace=True) + x_136 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_139 = torch.nn.functional.batch_norm( + x_138, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_138 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_140 = torch.nn.functional.silu(x_139, inplace=True) + x_139 = None + x_141 = x_140 + x_134 + x_140 = x_134 = None + x_142 = torch.conv2d( + x_141, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_143 = torch.nn.functional.batch_norm( + x_142, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_144 = torch.nn.functional.silu(x_143, inplace=True) + x_143 = None + x_145 = torch.conv2d( + x_144, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_144 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_146 = torch.nn.functional.batch_norm( + x_145, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_145 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_147 = torch.nn.functional.silu(x_146, inplace=True) + x_146 = None + x_148 = x_147 + x_141 + x_147 = x_141 = None + x_149 = torch.conv2d( + x_148, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_150 = torch.nn.functional.batch_norm( + x_149, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_149 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_151 = torch.nn.functional.silu(x_150, inplace=True) + x_150 = None + x_152 = torch.conv2d( + x_151, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_151 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_153 = torch.nn.functional.batch_norm( + x_152, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_152 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_154 = torch.nn.functional.silu(x_153, inplace=True) + x_153 = None + x_155 = x_154 + x_148 + x_154 = x_148 = None + x_156 = torch.conv2d( + x_155, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_157 = torch.nn.functional.batch_norm( + x_156, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_156 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = (None) + x_158 = torch.nn.functional.silu(x_157, inplace=True) + x_157 = None + x_159 = torch.conv2d( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = (None) + x_160 = torch.nn.functional.batch_norm( + x_159, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_159 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = (None) + x_161 = torch.nn.functional.silu(x_160, inplace=True) + x_160 = None + x_162 = x_161 + x_155 + x_161 = x_155 = None + x_163 = torch.conv2d( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_164 = torch.nn.functional.batch_norm( + x_163, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_163 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_ = (None) + x_165 = torch.nn.functional.silu(x_164, inplace=True) + x_164 = None + x_166 = torch.conv2d( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_165 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_ = (None) + x_167 = torch.nn.functional.batch_norm( + x_166, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_166 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_ = (None) + x_168 = torch.nn.functional.silu(x_167, inplace=True) + x_167 = None + x_169 = x_168 + x_162 + x_168 = x_162 = None + x_170 = torch.conv2d( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_171 = torch.nn.functional.batch_norm( + x_170, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_170 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_ = (None) + x_172 = torch.nn.functional.silu(x_171, inplace=True) + x_171 = None + x_173 = torch.conv2d( + x_172, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_172 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_ = (None) + x_174 = torch.nn.functional.batch_norm( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_173 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_ = (None) + x_175 = torch.nn.functional.silu(x_174, inplace=True) + x_174 = None + x_176 = x_175 + x_169 + x_175 = x_169 = None + cat_2 = torch.cat([x_176, x2_2], dim=1) + x_176 = x2_2 = None + x_177 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_178 = torch.nn.functional.batch_norm( + x_177, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_177 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_179 = torch.nn.functional.silu(x_178, inplace=True) + x_178 = None + x_180 = torch.conv2d( + x_179, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_179 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_181 = torch.nn.functional.batch_norm( + x_180, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_180 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_182 = torch.nn.functional.silu(x_181, inplace=True) + x_181 = None + x_183 = torch.conv2d( + x_182, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_182 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_184 = torch.nn.functional.batch_norm( + x_183, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_183 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_185 = torch.nn.functional.silu(x_184, inplace=True) + x_184 = None + split_3 = x_185.split(640, dim=1) + x_185 = None + x1_3 = split_3[0] + x2_3 = split_3[1] + split_3 = None + x_186 = torch.conv2d( + x1_3, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_187 = torch.nn.functional.batch_norm( + x_186, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_186 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_188 = torch.nn.functional.silu(x_187, inplace=True) + x_187 = None + x_189 = torch.conv2d( + x_188, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_188 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_190 = torch.nn.functional.batch_norm( + x_189, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_189 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_191 = torch.nn.functional.silu(x_190, inplace=True) + x_190 = None + x_192 = x_191 + x1_3 + x_191 = x1_3 = None + x_193 = torch.conv2d( + x_192, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_194 = torch.nn.functional.batch_norm( + x_193, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_193 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_195 = torch.nn.functional.silu(x_194, inplace=True) + x_194 = None + x_196 = torch.conv2d( + x_195, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_195 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_197 = torch.nn.functional.batch_norm( + x_196, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_196 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_198 = torch.nn.functional.silu(x_197, inplace=True) + x_197 = None + x_199 = x_198 + x_192 + x_198 = x_192 = None + x_200 = torch.conv2d( + x_199, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_201 = torch.nn.functional.batch_norm( + x_200, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_200 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_202 = torch.nn.functional.silu(x_201, inplace=True) + x_201 = None + x_203 = torch.conv2d( + x_202, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_202 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_204 = torch.nn.functional.batch_norm( + x_203, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_203 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_205 = torch.nn.functional.silu(x_204, inplace=True) + x_204 = None + x_206 = x_205 + x_199 + x_205 = x_199 = None + cat_3 = torch.cat([x_206, x2_3], dim=1) + x_206 = x2_3 = None + x_207 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_208 = torch.nn.functional.batch_norm( + x_207, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_207 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_209 = torch.nn.functional.silu(x_208, inplace=True) + x_208 = None + x_210 = torch.nn.functional.adaptive_avg_pool2d(x_209, 1) + x_209 = None + x_211 = x_210.flatten(1, -1) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_213 = torch._C._nn.linear( + x_212, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_212 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_213,) diff --git a/samples/timm/cs3edgenet_x.c2_in1k/weight_meta.py b/samples/timm/cs3edgenet_x.c2_in1k/weight_meta.py new file mode 100644 index 0000000000..a2264641ae --- /dev/null +++ b/samples/timm/cs3edgenet_x.c2_in1k/weight_meta.py @@ -0,0 +1,3162 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [40, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.075 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.230 + std = 1.283 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 40, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_" + shape = [80, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [160, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [120, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.159 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [120, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.158 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [120, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.158 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [320, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [640, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [640, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [640, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [1280, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [960, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [640, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [960, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [640, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [960, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [640, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [1280, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cs3se_edgenet_x.c2ns_in1k/graph_hash.txt b/samples/timm/cs3se_edgenet_x.c2ns_in1k/graph_hash.txt new file mode 100644 index 0000000000..89a760bcfe --- /dev/null +++ b/samples/timm/cs3se_edgenet_x.c2ns_in1k/graph_hash.txt @@ -0,0 +1 @@ +72fd84a46be8697b4d0df714a13ffa325c87f75dd13f8b9b1fa1e3f20421a218 \ No newline at end of file diff --git a/samples/timm/cs3se_edgenet_x.c2ns_in1k/graph_net.json b/samples/timm/cs3se_edgenet_x.c2ns_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/cs3se_edgenet_x.c2ns_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cs3se_edgenet_x.c2ns_in1k/input_meta.py b/samples/timm/cs3se_edgenet_x.c2ns_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cs3se_edgenet_x.c2ns_in1k/input_tensor_constraints.py b/samples/timm/cs3se_edgenet_x.c2ns_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cs3se_edgenet_x.c2ns_in1k/model.py b/samples/timm/cs3se_edgenet_x.c2ns_in1k/model.py new file mode 100644 index 0000000000..4482575d50 --- /dev/null +++ b/samples/timm/cs3se_edgenet_x.c2ns_in1k/model.py @@ -0,0 +1,3091 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.silu(x_1, inplace=True) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = None + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = None + x_5 = torch.nn.functional.silu(x_4, inplace=True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_5 = l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_8 = torch.nn.functional.silu(x_7, inplace=True) + x_7 = None + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_8 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_11 = torch.nn.functional.silu(x_10, inplace=True) + x_10 = None + split = x_11.split(80, dim=1) + x_11 = None + x1 = split[0] + x2 = split[1] + split = None + x_12 = torch.conv2d( + x1, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_13 = torch.nn.functional.batch_norm( + x_12, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_14 = torch.nn.functional.silu(x_13, inplace=True) + x_13 = None + x_se = x_14.mean((2, 3), keepdim=True) + x_se_1 = torch.conv2d( + x_se, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_2 = torch.nn.functional.silu(x_se_1, inplace=True) + x_se_1 = None + x_se_3 = torch.conv2d( + x_se_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_2 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid = x_se_3.sigmoid() + x_se_3 = None + x_15 = x_14 * sigmoid + x_14 = sigmoid = None + x_16 = torch.conv2d( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_17 = torch.nn.functional.batch_norm( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_18 = torch.nn.functional.silu(x_17, inplace=True) + x_17 = None + x_19 = x_18 + x1 + x_18 = x1 = None + x_20 = torch.conv2d( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_21 = torch.nn.functional.batch_norm( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_22 = torch.nn.functional.silu(x_21, inplace=True) + x_21 = None + x_se_4 = x_22.mean((2, 3), keepdim=True) + x_se_5 = torch.conv2d( + x_se_4, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_4 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_6 = torch.nn.functional.silu(x_se_5, inplace=True) + x_se_5 = None + x_se_7 = torch.conv2d( + x_se_6, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_6 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_1 = x_se_7.sigmoid() + x_se_7 = None + x_23 = x_22 * sigmoid_1 + x_22 = sigmoid_1 = None + x_24 = torch.conv2d( + x_23, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_25 = torch.nn.functional.batch_norm( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_24 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_26 = torch.nn.functional.silu(x_25, inplace=True) + x_25 = None + x_27 = x_26 + x_19 + x_26 = x_19 = None + x_28 = torch.conv2d( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_29 = torch.nn.functional.batch_norm( + x_28, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_28 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_30 = torch.nn.functional.silu(x_29, inplace=True) + x_29 = None + x_se_8 = x_30.mean((2, 3), keepdim=True) + x_se_9 = torch.conv2d( + x_se_8, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_8 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_10 = torch.nn.functional.silu(x_se_9, inplace=True) + x_se_9 = None + x_se_11 = torch.conv2d( + x_se_10, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_10 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_2 = x_se_11.sigmoid() + x_se_11 = None + x_31 = x_30 * sigmoid_2 + x_30 = sigmoid_2 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_33 = torch.nn.functional.batch_norm( + x_32, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_32 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_34 = torch.nn.functional.silu(x_33, inplace=True) + x_33 = None + x_35 = x_34 + x_27 + x_34 = x_27 = None + cat = torch.cat([x_35, x2], dim=1) + x_35 = x2 = None + x_36 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_37 = torch.nn.functional.batch_norm( + x_36, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_36 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_38 = torch.nn.functional.silu(x_37, inplace=True) + x_37 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_40 = torch.nn.functional.batch_norm( + x_39, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_39 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_41 = torch.nn.functional.silu(x_40, inplace=True) + x_40 = None + x_42 = torch.conv2d( + x_41, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_41 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_43 = torch.nn.functional.batch_norm( + x_42, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_42 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_44 = torch.nn.functional.silu(x_43, inplace=True) + x_43 = None + split_1 = x_44.split(160, dim=1) + x_44 = None + x1_1 = split_1[0] + x2_1 = split_1[1] + split_1 = None + x_45 = torch.conv2d( + x1_1, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_46 = torch.nn.functional.batch_norm( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_47 = torch.nn.functional.silu(x_46, inplace=True) + x_46 = None + x_se_12 = x_47.mean((2, 3), keepdim=True) + x_se_13 = torch.conv2d( + x_se_12, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_12 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_14 = torch.nn.functional.silu(x_se_13, inplace=True) + x_se_13 = None + x_se_15 = torch.conv2d( + x_se_14, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_14 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_3 = x_se_15.sigmoid() + x_se_15 = None + x_48 = x_47 * sigmoid_3 + x_47 = sigmoid_3 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_50 = torch.nn.functional.batch_norm( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_49 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_51 = torch.nn.functional.silu(x_50, inplace=True) + x_50 = None + x_52 = x_51 + x1_1 + x_51 = x1_1 = None + x_53 = torch.conv2d( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_54 = torch.nn.functional.batch_norm( + x_53, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_53 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_55 = torch.nn.functional.silu(x_54, inplace=True) + x_54 = None + x_se_16 = x_55.mean((2, 3), keepdim=True) + x_se_17 = torch.conv2d( + x_se_16, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_16 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_18 = torch.nn.functional.silu(x_se_17, inplace=True) + x_se_17 = None + x_se_19 = torch.conv2d( + x_se_18, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_18 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_4 = x_se_19.sigmoid() + x_se_19 = None + x_56 = x_55 * sigmoid_4 + x_55 = sigmoid_4 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_58 = torch.nn.functional.batch_norm( + x_57, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_57 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_59 = torch.nn.functional.silu(x_58, inplace=True) + x_58 = None + x_60 = x_59 + x_52 + x_59 = x_52 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_62 = torch.nn.functional.batch_norm( + x_61, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_63 = torch.nn.functional.silu(x_62, inplace=True) + x_62 = None + x_se_20 = x_63.mean((2, 3), keepdim=True) + x_se_21 = torch.conv2d( + x_se_20, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_20 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_22 = torch.nn.functional.silu(x_se_21, inplace=True) + x_se_21 = None + x_se_23 = torch.conv2d( + x_se_22, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_22 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_5 = x_se_23.sigmoid() + x_se_23 = None + x_64 = x_63 * sigmoid_5 + x_63 = sigmoid_5 = None + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_66 = torch.nn.functional.batch_norm( + x_65, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_65 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_67 = torch.nn.functional.silu(x_66, inplace=True) + x_66 = None + x_68 = x_67 + x_60 + x_67 = x_60 = None + x_69 = torch.conv2d( + x_68, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_70 = torch.nn.functional.batch_norm( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_71 = torch.nn.functional.silu(x_70, inplace=True) + x_70 = None + x_se_24 = x_71.mean((2, 3), keepdim=True) + x_se_25 = torch.conv2d( + x_se_24, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_24 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_26 = torch.nn.functional.silu(x_se_25, inplace=True) + x_se_25 = None + x_se_27 = torch.conv2d( + x_se_26, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_26 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_6 = x_se_27.sigmoid() + x_se_27 = None + x_72 = x_71 * sigmoid_6 + x_71 = sigmoid_6 = None + x_73 = torch.conv2d( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_74 = torch.nn.functional.batch_norm( + x_73, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_73 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_75 = torch.nn.functional.silu(x_74, inplace=True) + x_74 = None + x_76 = x_75 + x_68 + x_75 = x_68 = None + x_77 = torch.conv2d( + x_76, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_78 = torch.nn.functional.batch_norm( + x_77, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_77 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_79 = torch.nn.functional.silu(x_78, inplace=True) + x_78 = None + x_se_28 = x_79.mean((2, 3), keepdim=True) + x_se_29 = torch.conv2d( + x_se_28, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_28 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_30 = torch.nn.functional.silu(x_se_29, inplace=True) + x_se_29 = None + x_se_31 = torch.conv2d( + x_se_30, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_30 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_7 = x_se_31.sigmoid() + x_se_31 = None + x_80 = x_79 * sigmoid_7 + x_79 = sigmoid_7 = None + x_81 = torch.conv2d( + x_80, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_82 = torch.nn.functional.batch_norm( + x_81, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_83 = torch.nn.functional.silu(x_82, inplace=True) + x_82 = None + x_84 = x_83 + x_76 + x_83 = x_76 = None + x_85 = torch.conv2d( + x_84, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_86 = torch.nn.functional.batch_norm( + x_85, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_85 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_87 = torch.nn.functional.silu(x_86, inplace=True) + x_86 = None + x_se_32 = x_87.mean((2, 3), keepdim=True) + x_se_33 = torch.conv2d( + x_se_32, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_32 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_34 = torch.nn.functional.silu(x_se_33, inplace=True) + x_se_33 = None + x_se_35 = torch.conv2d( + x_se_34, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_34 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_8 = x_se_35.sigmoid() + x_se_35 = None + x_88 = x_87 * sigmoid_8 + x_87 = sigmoid_8 = None + x_89 = torch.conv2d( + x_88, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_88 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_90 = torch.nn.functional.batch_norm( + x_89, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_91 = torch.nn.functional.silu(x_90, inplace=True) + x_90 = None + x_92 = x_91 + x_84 + x_91 = x_84 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_94 = torch.nn.functional.batch_norm( + x_93, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_95 = torch.nn.functional.silu(x_94, inplace=True) + x_94 = None + x_se_36 = x_95.mean((2, 3), keepdim=True) + x_se_37 = torch.conv2d( + x_se_36, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_36 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_38 = torch.nn.functional.silu(x_se_37, inplace=True) + x_se_37 = None + x_se_39 = torch.conv2d( + x_se_38, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_38 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_9 = x_se_39.sigmoid() + x_se_39 = None + x_96 = x_95 * sigmoid_9 + x_95 = sigmoid_9 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_96 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_98 = torch.nn.functional.batch_norm( + x_97, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_99 = torch.nn.functional.silu(x_98, inplace=True) + x_98 = None + x_100 = x_99 + x_92 + x_99 = x_92 = None + cat_1 = torch.cat([x_100, x2_1], dim=1) + x_100 = x2_1 = None + x_101 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_102 = torch.nn.functional.batch_norm( + x_101, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_101 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_103 = torch.nn.functional.silu(x_102, inplace=True) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_105 = torch.nn.functional.batch_norm( + x_104, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_104 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_106 = torch.nn.functional.silu(x_105, inplace=True) + x_105 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_106 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_108 = torch.nn.functional.batch_norm( + x_107, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_107 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_109 = torch.nn.functional.silu(x_108, inplace=True) + x_108 = None + split_2 = x_109.split(320, dim=1) + x_109 = None + x1_2 = split_2[0] + x2_2 = split_2[1] + split_2 = None + x_110 = torch.conv2d( + x1_2, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_111 = torch.nn.functional.batch_norm( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_112 = torch.nn.functional.silu(x_111, inplace=True) + x_111 = None + x_se_40 = x_112.mean((2, 3), keepdim=True) + x_se_41 = torch.conv2d( + x_se_40, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_40 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_42 = torch.nn.functional.silu(x_se_41, inplace=True) + x_se_41 = None + x_se_43 = torch.conv2d( + x_se_42, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_42 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_10 = x_se_43.sigmoid() + x_se_43 = None + x_113 = x_112 * sigmoid_10 + x_112 = sigmoid_10 = None + x_114 = torch.conv2d( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_113 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_115 = torch.nn.functional.batch_norm( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_116 = torch.nn.functional.silu(x_115, inplace=True) + x_115 = None + x_117 = x_116 + x1_2 + x_116 = x1_2 = None + x_118 = torch.conv2d( + x_117, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_119 = torch.nn.functional.batch_norm( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_118 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_120 = torch.nn.functional.silu(x_119, inplace=True) + x_119 = None + x_se_44 = x_120.mean((2, 3), keepdim=True) + x_se_45 = torch.conv2d( + x_se_44, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_44 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_46 = torch.nn.functional.silu(x_se_45, inplace=True) + x_se_45 = None + x_se_47 = torch.conv2d( + x_se_46, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_46 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_11 = x_se_47.sigmoid() + x_se_47 = None + x_121 = x_120 * sigmoid_11 + x_120 = sigmoid_11 = None + x_122 = torch.conv2d( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_123 = torch.nn.functional.batch_norm( + x_122, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_124 = torch.nn.functional.silu(x_123, inplace=True) + x_123 = None + x_125 = x_124 + x_117 + x_124 = x_117 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_127 = torch.nn.functional.batch_norm( + x_126, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_126 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_128 = torch.nn.functional.silu(x_127, inplace=True) + x_127 = None + x_se_48 = x_128.mean((2, 3), keepdim=True) + x_se_49 = torch.conv2d( + x_se_48, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_48 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_50 = torch.nn.functional.silu(x_se_49, inplace=True) + x_se_49 = None + x_se_51 = torch.conv2d( + x_se_50, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_50 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_12 = x_se_51.sigmoid() + x_se_51 = None + x_129 = x_128 * sigmoid_12 + x_128 = sigmoid_12 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_131 = torch.nn.functional.batch_norm( + x_130, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_130 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_132 = torch.nn.functional.silu(x_131, inplace=True) + x_131 = None + x_133 = x_132 + x_125 + x_132 = x_125 = None + x_134 = torch.conv2d( + x_133, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_135 = torch.nn.functional.batch_norm( + x_134, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_134 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_136 = torch.nn.functional.silu(x_135, inplace=True) + x_135 = None + x_se_52 = x_136.mean((2, 3), keepdim=True) + x_se_53 = torch.conv2d( + x_se_52, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_52 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_54 = torch.nn.functional.silu(x_se_53, inplace=True) + x_se_53 = None + x_se_55 = torch.conv2d( + x_se_54, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_54 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_13 = x_se_55.sigmoid() + x_se_55 = None + x_137 = x_136 * sigmoid_13 + x_136 = sigmoid_13 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_139 = torch.nn.functional.batch_norm( + x_138, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_138 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_140 = torch.nn.functional.silu(x_139, inplace=True) + x_139 = None + x_141 = x_140 + x_133 + x_140 = x_133 = None + x_142 = torch.conv2d( + x_141, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_143 = torch.nn.functional.batch_norm( + x_142, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_144 = torch.nn.functional.silu(x_143, inplace=True) + x_143 = None + x_se_56 = x_144.mean((2, 3), keepdim=True) + x_se_57 = torch.conv2d( + x_se_56, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_56 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_58 = torch.nn.functional.silu(x_se_57, inplace=True) + x_se_57 = None + x_se_59 = torch.conv2d( + x_se_58, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_58 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_14 = x_se_59.sigmoid() + x_se_59 = None + x_145 = x_144 * sigmoid_14 + x_144 = sigmoid_14 = None + x_146 = torch.conv2d( + x_145, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_145 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_147 = torch.nn.functional.batch_norm( + x_146, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_146 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_148 = torch.nn.functional.silu(x_147, inplace=True) + x_147 = None + x_149 = x_148 + x_141 + x_148 = x_141 = None + x_150 = torch.conv2d( + x_149, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_151 = torch.nn.functional.batch_norm( + x_150, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_150 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_152 = torch.nn.functional.silu(x_151, inplace=True) + x_151 = None + x_se_60 = x_152.mean((2, 3), keepdim=True) + x_se_61 = torch.conv2d( + x_se_60, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_60 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_62 = torch.nn.functional.silu(x_se_61, inplace=True) + x_se_61 = None + x_se_63 = torch.conv2d( + x_se_62, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_62 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_15 = x_se_63.sigmoid() + x_se_63 = None + x_153 = x_152 * sigmoid_15 + x_152 = sigmoid_15 = None + x_154 = torch.conv2d( + x_153, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_155 = torch.nn.functional.batch_norm( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_156 = torch.nn.functional.silu(x_155, inplace=True) + x_155 = None + x_157 = x_156 + x_149 + x_156 = x_149 = None + x_158 = torch.conv2d( + x_157, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_159 = torch.nn.functional.batch_norm( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_160 = torch.nn.functional.silu(x_159, inplace=True) + x_159 = None + x_se_64 = x_160.mean((2, 3), keepdim=True) + x_se_65 = torch.conv2d( + x_se_64, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_64 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_66 = torch.nn.functional.silu(x_se_65, inplace=True) + x_se_65 = None + x_se_67 = torch.conv2d( + x_se_66, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_66 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_16 = x_se_67.sigmoid() + x_se_67 = None + x_161 = x_160 * sigmoid_16 + x_160 = sigmoid_16 = None + x_162 = torch.conv2d( + x_161, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_161 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_163 = torch.nn.functional.batch_norm( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_162 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_164 = torch.nn.functional.silu(x_163, inplace=True) + x_163 = None + x_165 = x_164 + x_157 + x_164 = x_157 = None + x_166 = torch.conv2d( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_167 = torch.nn.functional.batch_norm( + x_166, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_166 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_168 = torch.nn.functional.silu(x_167, inplace=True) + x_167 = None + x_se_68 = x_168.mean((2, 3), keepdim=True) + x_se_69 = torch.conv2d( + x_se_68, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_68 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_70 = torch.nn.functional.silu(x_se_69, inplace=True) + x_se_69 = None + x_se_71 = torch.conv2d( + x_se_70, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_70 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_17 = x_se_71.sigmoid() + x_se_71 = None + x_169 = x_168 * sigmoid_17 + x_168 = sigmoid_17 = None + x_170 = torch.conv2d( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_171 = torch.nn.functional.batch_norm( + x_170, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_170 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_172 = torch.nn.functional.silu(x_171, inplace=True) + x_171 = None + x_173 = x_172 + x_165 + x_172 = x_165 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_175 = torch.nn.functional.batch_norm( + x_174, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_174 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = (None) + x_176 = torch.nn.functional.silu(x_175, inplace=True) + x_175 = None + x_se_72 = x_176.mean((2, 3), keepdim=True) + x_se_73 = torch.conv2d( + x_se_72, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_72 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_74 = torch.nn.functional.silu(x_se_73, inplace=True) + x_se_73 = None + x_se_75 = torch.conv2d( + x_se_74, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_74 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_18 = x_se_75.sigmoid() + x_se_75 = None + x_177 = x_176 * sigmoid_18 + x_176 = sigmoid_18 = None + x_178 = torch.conv2d( + x_177, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_177 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = (None) + x_179 = torch.nn.functional.batch_norm( + x_178, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_178 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = (None) + x_180 = torch.nn.functional.silu(x_179, inplace=True) + x_179 = None + x_181 = x_180 + x_173 + x_180 = x_173 = None + x_182 = torch.conv2d( + x_181, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_183 = torch.nn.functional.batch_norm( + x_182, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_182 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_ = (None) + x_184 = torch.nn.functional.silu(x_183, inplace=True) + x_183 = None + x_se_76 = x_184.mean((2, 3), keepdim=True) + x_se_77 = torch.conv2d( + x_se_76, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_76 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_78 = torch.nn.functional.silu(x_se_77, inplace=True) + x_se_77 = None + x_se_79 = torch.conv2d( + x_se_78, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_78 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_19 = x_se_79.sigmoid() + x_se_79 = None + x_185 = x_184 * sigmoid_19 + x_184 = sigmoid_19 = None + x_186 = torch.conv2d( + x_185, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_185 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_ = (None) + x_187 = torch.nn.functional.batch_norm( + x_186, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_ = (None) + x_188 = torch.nn.functional.silu(x_187, inplace=True) + x_187 = None + x_189 = x_188 + x_181 + x_188 = x_181 = None + x_190 = torch.conv2d( + x_189, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_191 = torch.nn.functional.batch_norm( + x_190, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_190 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_ = (None) + x_192 = torch.nn.functional.silu(x_191, inplace=True) + x_191 = None + x_se_80 = x_192.mean((2, 3), keepdim=True) + x_se_81 = torch.conv2d( + x_se_80, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_80 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_82 = torch.nn.functional.silu(x_se_81, inplace=True) + x_se_81 = None + x_se_83 = torch.conv2d( + x_se_82, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_82 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_20 = x_se_83.sigmoid() + x_se_83 = None + x_193 = x_192 * sigmoid_20 + x_192 = sigmoid_20 = None + x_194 = torch.conv2d( + x_193, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_193 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_ = (None) + x_195 = torch.nn.functional.batch_norm( + x_194, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_194 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_ = (None) + x_196 = torch.nn.functional.silu(x_195, inplace=True) + x_195 = None + x_197 = x_196 + x_189 + x_196 = x_189 = None + cat_2 = torch.cat([x_197, x2_2], dim=1) + x_197 = x2_2 = None + x_198 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_199 = torch.nn.functional.batch_norm( + x_198, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_198 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_200 = torch.nn.functional.silu(x_199, inplace=True) + x_199 = None + x_201 = torch.conv2d( + x_200, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_200 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_202 = torch.nn.functional.batch_norm( + x_201, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_201 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_203 = torch.nn.functional.silu(x_202, inplace=True) + x_202 = None + x_204 = torch.conv2d( + x_203, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_203 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_205 = torch.nn.functional.batch_norm( + x_204, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_204 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_206 = torch.nn.functional.silu(x_205, inplace=True) + x_205 = None + split_3 = x_206.split(640, dim=1) + x_206 = None + x1_3 = split_3[0] + x2_3 = split_3[1] + split_3 = None + x_207 = torch.conv2d( + x1_3, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_208 = torch.nn.functional.batch_norm( + x_207, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_207 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_209 = torch.nn.functional.silu(x_208, inplace=True) + x_208 = None + x_se_84 = x_209.mean((2, 3), keepdim=True) + x_se_85 = torch.conv2d( + x_se_84, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_84 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_86 = torch.nn.functional.silu(x_se_85, inplace=True) + x_se_85 = None + x_se_87 = torch.conv2d( + x_se_86, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_86 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_21 = x_se_87.sigmoid() + x_se_87 = None + x_210 = x_209 * sigmoid_21 + x_209 = sigmoid_21 = None + x_211 = torch.conv2d( + x_210, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_210 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_212 = torch.nn.functional.batch_norm( + x_211, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_211 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_213 = torch.nn.functional.silu(x_212, inplace=True) + x_212 = None + x_214 = x_213 + x1_3 + x_213 = x1_3 = None + x_215 = torch.conv2d( + x_214, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_216 = torch.nn.functional.batch_norm( + x_215, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_215 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_217 = torch.nn.functional.silu(x_216, inplace=True) + x_216 = None + x_se_88 = x_217.mean((2, 3), keepdim=True) + x_se_89 = torch.conv2d( + x_se_88, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_88 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_90 = torch.nn.functional.silu(x_se_89, inplace=True) + x_se_89 = None + x_se_91 = torch.conv2d( + x_se_90, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_90 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_22 = x_se_91.sigmoid() + x_se_91 = None + x_218 = x_217 * sigmoid_22 + x_217 = sigmoid_22 = None + x_219 = torch.conv2d( + x_218, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_218 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_220 = torch.nn.functional.batch_norm( + x_219, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_219 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_221 = torch.nn.functional.silu(x_220, inplace=True) + x_220 = None + x_222 = x_221 + x_214 + x_221 = x_214 = None + x_223 = torch.conv2d( + x_222, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_224 = torch.nn.functional.batch_norm( + x_223, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_223 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_225 = torch.nn.functional.silu(x_224, inplace=True) + x_224 = None + x_se_92 = x_225.mean((2, 3), keepdim=True) + x_se_93 = torch.conv2d( + x_se_92, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_92 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_94 = torch.nn.functional.silu(x_se_93, inplace=True) + x_se_93 = None + x_se_95 = torch.conv2d( + x_se_94, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_94 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_23 = x_se_95.sigmoid() + x_se_95 = None + x_226 = x_225 * sigmoid_23 + x_225 = sigmoid_23 = None + x_227 = torch.conv2d( + x_226, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_226 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_228 = torch.nn.functional.batch_norm( + x_227, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_227 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_229 = torch.nn.functional.silu(x_228, inplace=True) + x_228 = None + x_230 = x_229 + x_222 + x_229 = x_222 = None + cat_3 = torch.cat([x_230, x2_3], dim=1) + x_230 = x2_3 = None + x_231 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_232 = torch.nn.functional.batch_norm( + x_231, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_231 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_233 = torch.nn.functional.silu(x_232, inplace=True) + x_232 = None + x_234 = torch.nn.functional.adaptive_avg_pool2d(x_233, 1) + x_233 = None + x_235 = x_234.flatten(1, -1) + x_234 = None + x_236 = torch.nn.functional.dropout(x_235, 0.0, False, False) + x_235 = None + x_237 = torch._C._nn.linear( + x_236, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_236 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_237,) diff --git a/samples/timm/cs3se_edgenet_x.c2ns_in1k/weight_meta.py b/samples/timm/cs3se_edgenet_x.c2ns_in1k/weight_meta.py new file mode 100644 index 0000000000..a0f4220105 --- /dev/null +++ b/samples/timm/cs3se_edgenet_x.c2ns_in1k/weight_meta.py @@ -0,0 +1,4122 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [40, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.075 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.221 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 40, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_" + shape = [80, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [160, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [120, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_" + shape = [32, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.245 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_" + shape = [120, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.132 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.157 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [120, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_" + shape = [32, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.248 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_" + shape = [120, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.129 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.159 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [120, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_" + shape = [32, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.245 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_" + shape = [120, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.129 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.158 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [320, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_" + shape = [240, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.091 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.175 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_" + shape = [240, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.092 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_" + shape = [240, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.091 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_" + shape = [240, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.092 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_" + shape = [240, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.092 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_" + shape = [240, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.091 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [240, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_" + shape = [240, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.092 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [640, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_" + shape = [120, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.129 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_" + shape = [480, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_" + shape = [120, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.129 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_" + shape = [480, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_" + shape = [120, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.129 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_" + shape = [480, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_" + shape = [120, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.129 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_" + shape = [480, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_" + shape = [120, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.129 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_" + shape = [480, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_" + shape = [120, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.129 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_" + shape = [480, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_" + shape = [120, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.128 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_" + shape = [480, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_" + shape = [120, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.129 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_" + shape = [480, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_" + shape = [120, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.128 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_" + shape = [480, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_" + shape = [120, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.129 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_" + shape = [480, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_" + shape = [480, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.022 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_" + shape = [120, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.129 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_" + shape = [120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_" + shape = [480, 120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [640, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [640, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [1280, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [960, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_" + shape = [240, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.091 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_" + shape = [960, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [640, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [960, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_" + shape = [240, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.091 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_" + shape = [960, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [640, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [960, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_" + shape = [240, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.092 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_" + shape = [960, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [640, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [1280, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cs3sedarknet_l.c2ns_in1k/graph_hash.txt b/samples/timm/cs3sedarknet_l.c2ns_in1k/graph_hash.txt new file mode 100644 index 0000000000..4fd012443c --- /dev/null +++ b/samples/timm/cs3sedarknet_l.c2ns_in1k/graph_hash.txt @@ -0,0 +1 @@ +9968dcd4e6249cddec435975f4335d38d057d1e3be27cb9ab2f6f4cb2dbc677d \ No newline at end of file diff --git a/samples/timm/cs3sedarknet_l.c2ns_in1k/graph_net.json b/samples/timm/cs3sedarknet_l.c2ns_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/cs3sedarknet_l.c2ns_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cs3sedarknet_l.c2ns_in1k/input_meta.py b/samples/timm/cs3sedarknet_l.c2ns_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cs3sedarknet_l.c2ns_in1k/input_tensor_constraints.py b/samples/timm/cs3sedarknet_l.c2ns_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cs3sedarknet_l.c2ns_in1k/model.py b/samples/timm/cs3sedarknet_l.c2ns_in1k/model.py new file mode 100644 index 0000000000..b3d8537226 --- /dev/null +++ b/samples/timm/cs3sedarknet_l.c2ns_in1k/model.py @@ -0,0 +1,2776 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.silu(x_1, inplace=True) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = None + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = None + x_5 = torch.nn.functional.silu(x_4, inplace=True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_5 = l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_8 = torch.nn.functional.silu(x_7, inplace=True) + x_7 = None + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_8 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_11 = torch.nn.functional.silu(x_10, inplace=True) + x_10 = None + split = x_11.split(64, dim=1) + x_11 = None + x1 = split[0] + x2 = split[1] + split = None + x_12 = torch.conv2d( + x1, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_13 = torch.nn.functional.batch_norm( + x_12, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_14 = torch.nn.functional.silu(x_13, inplace=True) + x_13 = None + x_se = x_14.mean((2, 3), keepdim=True) + x_se_1 = torch.conv2d( + x_se, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_2 = torch.nn.functional.silu(x_se_1, inplace=True) + x_se_1 = None + x_se_3 = torch.conv2d( + x_se_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_2 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid = x_se_3.sigmoid() + x_se_3 = None + x_15 = x_14 * sigmoid + x_14 = sigmoid = None + x_16 = torch.conv2d( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_17 = torch.nn.functional.batch_norm( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_18 = torch.nn.functional.silu(x_17, inplace=True) + x_17 = None + x_19 = x_18 + x1 + x_18 = x1 = None + x_20 = torch.conv2d( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_21 = torch.nn.functional.batch_norm( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_22 = torch.nn.functional.silu(x_21, inplace=True) + x_21 = None + x_se_4 = x_22.mean((2, 3), keepdim=True) + x_se_5 = torch.conv2d( + x_se_4, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_4 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_6 = torch.nn.functional.silu(x_se_5, inplace=True) + x_se_5 = None + x_se_7 = torch.conv2d( + x_se_6, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_6 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_1 = x_se_7.sigmoid() + x_se_7 = None + x_23 = x_22 * sigmoid_1 + x_22 = sigmoid_1 = None + x_24 = torch.conv2d( + x_23, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_25 = torch.nn.functional.batch_norm( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_24 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_26 = torch.nn.functional.silu(x_25, inplace=True) + x_25 = None + x_27 = x_26 + x_19 + x_26 = x_19 = None + x_28 = torch.conv2d( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_29 = torch.nn.functional.batch_norm( + x_28, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_28 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_30 = torch.nn.functional.silu(x_29, inplace=True) + x_29 = None + x_se_8 = x_30.mean((2, 3), keepdim=True) + x_se_9 = torch.conv2d( + x_se_8, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_8 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_10 = torch.nn.functional.silu(x_se_9, inplace=True) + x_se_9 = None + x_se_11 = torch.conv2d( + x_se_10, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_10 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_2 = x_se_11.sigmoid() + x_se_11 = None + x_31 = x_30 * sigmoid_2 + x_30 = sigmoid_2 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_33 = torch.nn.functional.batch_norm( + x_32, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_32 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_34 = torch.nn.functional.silu(x_33, inplace=True) + x_33 = None + x_35 = x_34 + x_27 + x_34 = x_27 = None + cat = torch.cat([x_35, x2], dim=1) + x_35 = x2 = None + x_36 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_37 = torch.nn.functional.batch_norm( + x_36, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_36 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_38 = torch.nn.functional.silu(x_37, inplace=True) + x_37 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_40 = torch.nn.functional.batch_norm( + x_39, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_39 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_41 = torch.nn.functional.silu(x_40, inplace=True) + x_40 = None + x_42 = torch.conv2d( + x_41, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_41 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_43 = torch.nn.functional.batch_norm( + x_42, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_42 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_44 = torch.nn.functional.silu(x_43, inplace=True) + x_43 = None + split_1 = x_44.split(128, dim=1) + x_44 = None + x1_1 = split_1[0] + x2_1 = split_1[1] + split_1 = None + x_45 = torch.conv2d( + x1_1, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_46 = torch.nn.functional.batch_norm( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_47 = torch.nn.functional.silu(x_46, inplace=True) + x_46 = None + x_se_12 = x_47.mean((2, 3), keepdim=True) + x_se_13 = torch.conv2d( + x_se_12, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_12 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_14 = torch.nn.functional.silu(x_se_13, inplace=True) + x_se_13 = None + x_se_15 = torch.conv2d( + x_se_14, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_14 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_3 = x_se_15.sigmoid() + x_se_15 = None + x_48 = x_47 * sigmoid_3 + x_47 = sigmoid_3 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_50 = torch.nn.functional.batch_norm( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_49 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_51 = torch.nn.functional.silu(x_50, inplace=True) + x_50 = None + x_52 = x_51 + x1_1 + x_51 = x1_1 = None + x_53 = torch.conv2d( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_54 = torch.nn.functional.batch_norm( + x_53, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_53 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_55 = torch.nn.functional.silu(x_54, inplace=True) + x_54 = None + x_se_16 = x_55.mean((2, 3), keepdim=True) + x_se_17 = torch.conv2d( + x_se_16, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_16 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_18 = torch.nn.functional.silu(x_se_17, inplace=True) + x_se_17 = None + x_se_19 = torch.conv2d( + x_se_18, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_18 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_4 = x_se_19.sigmoid() + x_se_19 = None + x_56 = x_55 * sigmoid_4 + x_55 = sigmoid_4 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_58 = torch.nn.functional.batch_norm( + x_57, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_57 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_59 = torch.nn.functional.silu(x_58, inplace=True) + x_58 = None + x_60 = x_59 + x_52 + x_59 = x_52 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_62 = torch.nn.functional.batch_norm( + x_61, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_63 = torch.nn.functional.silu(x_62, inplace=True) + x_62 = None + x_se_20 = x_63.mean((2, 3), keepdim=True) + x_se_21 = torch.conv2d( + x_se_20, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_20 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_22 = torch.nn.functional.silu(x_se_21, inplace=True) + x_se_21 = None + x_se_23 = torch.conv2d( + x_se_22, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_22 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_5 = x_se_23.sigmoid() + x_se_23 = None + x_64 = x_63 * sigmoid_5 + x_63 = sigmoid_5 = None + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_66 = torch.nn.functional.batch_norm( + x_65, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_65 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_67 = torch.nn.functional.silu(x_66, inplace=True) + x_66 = None + x_68 = x_67 + x_60 + x_67 = x_60 = None + x_69 = torch.conv2d( + x_68, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_70 = torch.nn.functional.batch_norm( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_71 = torch.nn.functional.silu(x_70, inplace=True) + x_70 = None + x_se_24 = x_71.mean((2, 3), keepdim=True) + x_se_25 = torch.conv2d( + x_se_24, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_24 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_26 = torch.nn.functional.silu(x_se_25, inplace=True) + x_se_25 = None + x_se_27 = torch.conv2d( + x_se_26, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_26 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_6 = x_se_27.sigmoid() + x_se_27 = None + x_72 = x_71 * sigmoid_6 + x_71 = sigmoid_6 = None + x_73 = torch.conv2d( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_74 = torch.nn.functional.batch_norm( + x_73, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_73 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_75 = torch.nn.functional.silu(x_74, inplace=True) + x_74 = None + x_76 = x_75 + x_68 + x_75 = x_68 = None + x_77 = torch.conv2d( + x_76, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_78 = torch.nn.functional.batch_norm( + x_77, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_77 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_79 = torch.nn.functional.silu(x_78, inplace=True) + x_78 = None + x_se_28 = x_79.mean((2, 3), keepdim=True) + x_se_29 = torch.conv2d( + x_se_28, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_28 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_30 = torch.nn.functional.silu(x_se_29, inplace=True) + x_se_29 = None + x_se_31 = torch.conv2d( + x_se_30, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_30 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_7 = x_se_31.sigmoid() + x_se_31 = None + x_80 = x_79 * sigmoid_7 + x_79 = sigmoid_7 = None + x_81 = torch.conv2d( + x_80, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_82 = torch.nn.functional.batch_norm( + x_81, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_83 = torch.nn.functional.silu(x_82, inplace=True) + x_82 = None + x_84 = x_83 + x_76 + x_83 = x_76 = None + x_85 = torch.conv2d( + x_84, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_86 = torch.nn.functional.batch_norm( + x_85, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_85 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_87 = torch.nn.functional.silu(x_86, inplace=True) + x_86 = None + x_se_32 = x_87.mean((2, 3), keepdim=True) + x_se_33 = torch.conv2d( + x_se_32, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_32 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_34 = torch.nn.functional.silu(x_se_33, inplace=True) + x_se_33 = None + x_se_35 = torch.conv2d( + x_se_34, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_34 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_8 = x_se_35.sigmoid() + x_se_35 = None + x_88 = x_87 * sigmoid_8 + x_87 = sigmoid_8 = None + x_89 = torch.conv2d( + x_88, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_88 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_90 = torch.nn.functional.batch_norm( + x_89, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_91 = torch.nn.functional.silu(x_90, inplace=True) + x_90 = None + x_92 = x_91 + x_84 + x_91 = x_84 = None + cat_1 = torch.cat([x_92, x2_1], dim=1) + x_92 = x2_1 = None + x_93 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_94 = torch.nn.functional.batch_norm( + x_93, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_93 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_95 = torch.nn.functional.silu(x_94, inplace=True) + x_94 = None + x_96 = torch.conv2d( + x_95, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_95 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_97 = torch.nn.functional.batch_norm( + x_96, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_96 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_98 = torch.nn.functional.silu(x_97, inplace=True) + x_97 = None + x_99 = torch.conv2d( + x_98, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_98 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_100 = torch.nn.functional.batch_norm( + x_99, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_99 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_101 = torch.nn.functional.silu(x_100, inplace=True) + x_100 = None + split_2 = x_101.split(256, dim=1) + x_101 = None + x1_2 = split_2[0] + x2_2 = split_2[1] + split_2 = None + x_102 = torch.conv2d( + x1_2, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_103 = torch.nn.functional.batch_norm( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_104 = torch.nn.functional.silu(x_103, inplace=True) + x_103 = None + x_se_36 = x_104.mean((2, 3), keepdim=True) + x_se_37 = torch.conv2d( + x_se_36, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_36 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_38 = torch.nn.functional.silu(x_se_37, inplace=True) + x_se_37 = None + x_se_39 = torch.conv2d( + x_se_38, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_38 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_9 = x_se_39.sigmoid() + x_se_39 = None + x_105 = x_104 * sigmoid_9 + x_104 = sigmoid_9 = None + x_106 = torch.conv2d( + x_105, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_105 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_107 = torch.nn.functional.batch_norm( + x_106, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_106 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_108 = torch.nn.functional.silu(x_107, inplace=True) + x_107 = None + x_109 = x_108 + x1_2 + x_108 = x1_2 = None + x_110 = torch.conv2d( + x_109, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_111 = torch.nn.functional.batch_norm( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_112 = torch.nn.functional.silu(x_111, inplace=True) + x_111 = None + x_se_40 = x_112.mean((2, 3), keepdim=True) + x_se_41 = torch.conv2d( + x_se_40, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_40 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_42 = torch.nn.functional.silu(x_se_41, inplace=True) + x_se_41 = None + x_se_43 = torch.conv2d( + x_se_42, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_42 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_10 = x_se_43.sigmoid() + x_se_43 = None + x_113 = x_112 * sigmoid_10 + x_112 = sigmoid_10 = None + x_114 = torch.conv2d( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_113 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_115 = torch.nn.functional.batch_norm( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_116 = torch.nn.functional.silu(x_115, inplace=True) + x_115 = None + x_117 = x_116 + x_109 + x_116 = x_109 = None + x_118 = torch.conv2d( + x_117, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_119 = torch.nn.functional.batch_norm( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_118 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_120 = torch.nn.functional.silu(x_119, inplace=True) + x_119 = None + x_se_44 = x_120.mean((2, 3), keepdim=True) + x_se_45 = torch.conv2d( + x_se_44, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_44 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_46 = torch.nn.functional.silu(x_se_45, inplace=True) + x_se_45 = None + x_se_47 = torch.conv2d( + x_se_46, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_46 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_11 = x_se_47.sigmoid() + x_se_47 = None + x_121 = x_120 * sigmoid_11 + x_120 = sigmoid_11 = None + x_122 = torch.conv2d( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_123 = torch.nn.functional.batch_norm( + x_122, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_124 = torch.nn.functional.silu(x_123, inplace=True) + x_123 = None + x_125 = x_124 + x_117 + x_124 = x_117 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_127 = torch.nn.functional.batch_norm( + x_126, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_126 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_128 = torch.nn.functional.silu(x_127, inplace=True) + x_127 = None + x_se_48 = x_128.mean((2, 3), keepdim=True) + x_se_49 = torch.conv2d( + x_se_48, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_48 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_50 = torch.nn.functional.silu(x_se_49, inplace=True) + x_se_49 = None + x_se_51 = torch.conv2d( + x_se_50, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_50 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_12 = x_se_51.sigmoid() + x_se_51 = None + x_129 = x_128 * sigmoid_12 + x_128 = sigmoid_12 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_131 = torch.nn.functional.batch_norm( + x_130, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_130 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_132 = torch.nn.functional.silu(x_131, inplace=True) + x_131 = None + x_133 = x_132 + x_125 + x_132 = x_125 = None + x_134 = torch.conv2d( + x_133, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_135 = torch.nn.functional.batch_norm( + x_134, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_134 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_136 = torch.nn.functional.silu(x_135, inplace=True) + x_135 = None + x_se_52 = x_136.mean((2, 3), keepdim=True) + x_se_53 = torch.conv2d( + x_se_52, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_52 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_54 = torch.nn.functional.silu(x_se_53, inplace=True) + x_se_53 = None + x_se_55 = torch.conv2d( + x_se_54, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_54 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_13 = x_se_55.sigmoid() + x_se_55 = None + x_137 = x_136 * sigmoid_13 + x_136 = sigmoid_13 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_139 = torch.nn.functional.batch_norm( + x_138, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_138 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_140 = torch.nn.functional.silu(x_139, inplace=True) + x_139 = None + x_141 = x_140 + x_133 + x_140 = x_133 = None + x_142 = torch.conv2d( + x_141, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_143 = torch.nn.functional.batch_norm( + x_142, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_144 = torch.nn.functional.silu(x_143, inplace=True) + x_143 = None + x_se_56 = x_144.mean((2, 3), keepdim=True) + x_se_57 = torch.conv2d( + x_se_56, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_56 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_58 = torch.nn.functional.silu(x_se_57, inplace=True) + x_se_57 = None + x_se_59 = torch.conv2d( + x_se_58, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_58 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_14 = x_se_59.sigmoid() + x_se_59 = None + x_145 = x_144 * sigmoid_14 + x_144 = sigmoid_14 = None + x_146 = torch.conv2d( + x_145, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_145 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_147 = torch.nn.functional.batch_norm( + x_146, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_146 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_148 = torch.nn.functional.silu(x_147, inplace=True) + x_147 = None + x_149 = x_148 + x_141 + x_148 = x_141 = None + x_150 = torch.conv2d( + x_149, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_151 = torch.nn.functional.batch_norm( + x_150, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_150 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_152 = torch.nn.functional.silu(x_151, inplace=True) + x_151 = None + x_se_60 = x_152.mean((2, 3), keepdim=True) + x_se_61 = torch.conv2d( + x_se_60, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_60 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_62 = torch.nn.functional.silu(x_se_61, inplace=True) + x_se_61 = None + x_se_63 = torch.conv2d( + x_se_62, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_62 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_15 = x_se_63.sigmoid() + x_se_63 = None + x_153 = x_152 * sigmoid_15 + x_152 = sigmoid_15 = None + x_154 = torch.conv2d( + x_153, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_155 = torch.nn.functional.batch_norm( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_156 = torch.nn.functional.silu(x_155, inplace=True) + x_155 = None + x_157 = x_156 + x_149 + x_156 = x_149 = None + x_158 = torch.conv2d( + x_157, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_159 = torch.nn.functional.batch_norm( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_160 = torch.nn.functional.silu(x_159, inplace=True) + x_159 = None + x_se_64 = x_160.mean((2, 3), keepdim=True) + x_se_65 = torch.conv2d( + x_se_64, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_64 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_66 = torch.nn.functional.silu(x_se_65, inplace=True) + x_se_65 = None + x_se_67 = torch.conv2d( + x_se_66, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_66 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_16 = x_se_67.sigmoid() + x_se_67 = None + x_161 = x_160 * sigmoid_16 + x_160 = sigmoid_16 = None + x_162 = torch.conv2d( + x_161, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_161 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_163 = torch.nn.functional.batch_norm( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_162 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_164 = torch.nn.functional.silu(x_163, inplace=True) + x_163 = None + x_165 = x_164 + x_157 + x_164 = x_157 = None + x_166 = torch.conv2d( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_167 = torch.nn.functional.batch_norm( + x_166, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_166 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = (None) + x_168 = torch.nn.functional.silu(x_167, inplace=True) + x_167 = None + x_se_68 = x_168.mean((2, 3), keepdim=True) + x_se_69 = torch.conv2d( + x_se_68, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_68 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_70 = torch.nn.functional.silu(x_se_69, inplace=True) + x_se_69 = None + x_se_71 = torch.conv2d( + x_se_70, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_70 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_17 = x_se_71.sigmoid() + x_se_71 = None + x_169 = x_168 * sigmoid_17 + x_168 = sigmoid_17 = None + x_170 = torch.conv2d( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = (None) + x_171 = torch.nn.functional.batch_norm( + x_170, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_170 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = (None) + x_172 = torch.nn.functional.silu(x_171, inplace=True) + x_171 = None + x_173 = x_172 + x_165 + x_172 = x_165 = None + cat_2 = torch.cat([x_173, x2_2], dim=1) + x_173 = x2_2 = None + x_174 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_175 = torch.nn.functional.batch_norm( + x_174, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_174 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_176 = torch.nn.functional.silu(x_175, inplace=True) + x_175 = None + x_177 = torch.conv2d( + x_176, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_176 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_178 = torch.nn.functional.batch_norm( + x_177, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_177 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_179 = torch.nn.functional.silu(x_178, inplace=True) + x_178 = None + x_180 = torch.conv2d( + x_179, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_179 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_181 = torch.nn.functional.batch_norm( + x_180, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_180 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_182 = torch.nn.functional.silu(x_181, inplace=True) + x_181 = None + split_3 = x_182.split(512, dim=1) + x_182 = None + x1_3 = split_3[0] + x2_3 = split_3[1] + split_3 = None + x_183 = torch.conv2d( + x1_3, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_184 = torch.nn.functional.batch_norm( + x_183, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_183 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_185 = torch.nn.functional.silu(x_184, inplace=True) + x_184 = None + x_se_72 = x_185.mean((2, 3), keepdim=True) + x_se_73 = torch.conv2d( + x_se_72, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_72 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_74 = torch.nn.functional.silu(x_se_73, inplace=True) + x_se_73 = None + x_se_75 = torch.conv2d( + x_se_74, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_74 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_18 = x_se_75.sigmoid() + x_se_75 = None + x_186 = x_185 * sigmoid_18 + x_185 = sigmoid_18 = None + x_187 = torch.conv2d( + x_186, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_186 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_188 = torch.nn.functional.batch_norm( + x_187, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_187 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_189 = torch.nn.functional.silu(x_188, inplace=True) + x_188 = None + x_190 = x_189 + x1_3 + x_189 = x1_3 = None + x_191 = torch.conv2d( + x_190, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_192 = torch.nn.functional.batch_norm( + x_191, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_191 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_193 = torch.nn.functional.silu(x_192, inplace=True) + x_192 = None + x_se_76 = x_193.mean((2, 3), keepdim=True) + x_se_77 = torch.conv2d( + x_se_76, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_76 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_78 = torch.nn.functional.silu(x_se_77, inplace=True) + x_se_77 = None + x_se_79 = torch.conv2d( + x_se_78, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_78 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_19 = x_se_79.sigmoid() + x_se_79 = None + x_194 = x_193 * sigmoid_19 + x_193 = sigmoid_19 = None + x_195 = torch.conv2d( + x_194, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_194 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_196 = torch.nn.functional.batch_norm( + x_195, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_195 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_197 = torch.nn.functional.silu(x_196, inplace=True) + x_196 = None + x_198 = x_197 + x_190 + x_197 = x_190 = None + x_199 = torch.conv2d( + x_198, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_200 = torch.nn.functional.batch_norm( + x_199, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_199 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_201 = torch.nn.functional.silu(x_200, inplace=True) + x_200 = None + x_se_80 = x_201.mean((2, 3), keepdim=True) + x_se_81 = torch.conv2d( + x_se_80, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_80 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_82 = torch.nn.functional.silu(x_se_81, inplace=True) + x_se_81 = None + x_se_83 = torch.conv2d( + x_se_82, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_82 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_20 = x_se_83.sigmoid() + x_se_83 = None + x_202 = x_201 * sigmoid_20 + x_201 = sigmoid_20 = None + x_203 = torch.conv2d( + x_202, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_202 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_204 = torch.nn.functional.batch_norm( + x_203, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_203 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_205 = torch.nn.functional.silu(x_204, inplace=True) + x_204 = None + x_206 = x_205 + x_198 + x_205 = x_198 = None + cat_3 = torch.cat([x_206, x2_3], dim=1) + x_206 = x2_3 = None + x_207 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_208 = torch.nn.functional.batch_norm( + x_207, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_207 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_209 = torch.nn.functional.silu(x_208, inplace=True) + x_208 = None + x_210 = torch.nn.functional.adaptive_avg_pool2d(x_209, 1) + x_209 = None + x_211 = x_210.flatten(1, -1) + x_210 = None + x_212 = torch.nn.functional.dropout(x_211, 0.0, False, False) + x_211 = None + x_213 = torch._C._nn.linear( + x_212, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_212 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_213,) diff --git a/samples/timm/cs3sedarknet_l.c2ns_in1k/weight_meta.py b/samples/timm/cs3sedarknet_l.c2ns_in1k/weight_meta.py new file mode 100644 index 0000000000..9727385988 --- /dev/null +++ b/samples/timm/cs3sedarknet_l.c2ns_in1k/weight_meta.py @@ -0,0 +1,3702 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [32, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.224 + std = 1.284 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.175 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_" + shape = [16, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.006 + std = 0.363 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_" + shape = [16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_" + shape = [64, 16, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.005 + std = 0.180 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.004 + std = 0.174 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_" + shape = [16, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.016 + std = 0.358 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_" + shape = [16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_" + shape = [64, 16, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.007 + std = 0.173 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_" + shape = [16, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.012 + std = 0.373 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_" + shape = [16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_" + shape = [64, 16, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_" + shape = [32, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.242 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_" + shape = [128, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_" + shape = [32, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.254 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_" + shape = [128, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.123 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_" + shape = [32, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.252 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_" + shape = [128, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.122 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_" + shape = [32, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.247 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_" + shape = [128, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.124 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_" + shape = [32, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.008 + std = 0.249 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_" + shape = [128, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.123 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_" + shape = [32, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.250 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_" + shape = [128, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.178 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_" + shape = [64, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [1024, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cs3sedarknet_x.c2ns_in1k/graph_hash.txt b/samples/timm/cs3sedarknet_x.c2ns_in1k/graph_hash.txt new file mode 100644 index 0000000000..18bda71715 --- /dev/null +++ b/samples/timm/cs3sedarknet_x.c2ns_in1k/graph_hash.txt @@ -0,0 +1 @@ +1b92ad067f29b181b3dc7b6bff02d3202367a7fe6d248d63413fad08d6f86697 \ No newline at end of file diff --git a/samples/timm/cs3sedarknet_x.c2ns_in1k/graph_net.json b/samples/timm/cs3sedarknet_x.c2ns_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/cs3sedarknet_x.c2ns_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cs3sedarknet_x.c2ns_in1k/input_meta.py b/samples/timm/cs3sedarknet_x.c2ns_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cs3sedarknet_x.c2ns_in1k/input_tensor_constraints.py b/samples/timm/cs3sedarknet_x.c2ns_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cs3sedarknet_x.c2ns_in1k/model.py b/samples/timm/cs3sedarknet_x.c2ns_in1k/model.py new file mode 100644 index 0000000000..8091febe41 --- /dev/null +++ b/samples/timm/cs3sedarknet_x.c2ns_in1k/model.py @@ -0,0 +1,3091 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.silu(x_1, inplace=True) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_stem_modules_conv2_modules_conv_parameters_weight_ = None + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv2_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv2_modules_bn_parameters_bias_ = None + x_5 = torch.nn.functional.silu(x_4, inplace=True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_5 = l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_8 = torch.nn.functional.silu(x_7, inplace=True) + x_7 = None + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_8 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_11 = torch.nn.functional.silu(x_10, inplace=True) + x_10 = None + split = x_11.split(80, dim=1) + x_11 = None + x1 = split[0] + x2 = split[1] + split = None + x_12 = torch.conv2d( + x1, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_13 = torch.nn.functional.batch_norm( + x_12, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_14 = torch.nn.functional.silu(x_13, inplace=True) + x_13 = None + x_se = x_14.mean((2, 3), keepdim=True) + x_se_1 = torch.conv2d( + x_se, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_2 = torch.nn.functional.silu(x_se_1, inplace=True) + x_se_1 = None + x_se_3 = torch.conv2d( + x_se_2, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_2 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid = x_se_3.sigmoid() + x_se_3 = None + x_15 = x_14 * sigmoid + x_14 = sigmoid = None + x_16 = torch.conv2d( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_17 = torch.nn.functional.batch_norm( + x_16, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_16 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_18 = torch.nn.functional.silu(x_17, inplace=True) + x_17 = None + x_19 = x_18 + x1 + x_18 = x1 = None + x_20 = torch.conv2d( + x_19, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_21 = torch.nn.functional.batch_norm( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_22 = torch.nn.functional.silu(x_21, inplace=True) + x_21 = None + x_se_4 = x_22.mean((2, 3), keepdim=True) + x_se_5 = torch.conv2d( + x_se_4, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_4 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_6 = torch.nn.functional.silu(x_se_5, inplace=True) + x_se_5 = None + x_se_7 = torch.conv2d( + x_se_6, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_6 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_1 = x_se_7.sigmoid() + x_se_7 = None + x_23 = x_22 * sigmoid_1 + x_22 = sigmoid_1 = None + x_24 = torch.conv2d( + x_23, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_23 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_25 = torch.nn.functional.batch_norm( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_24 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_26 = torch.nn.functional.silu(x_25, inplace=True) + x_25 = None + x_27 = x_26 + x_19 + x_26 = x_19 = None + x_28 = torch.conv2d( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_29 = torch.nn.functional.batch_norm( + x_28, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_28 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_30 = torch.nn.functional.silu(x_29, inplace=True) + x_29 = None + x_se_8 = x_30.mean((2, 3), keepdim=True) + x_se_9 = torch.conv2d( + x_se_8, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_8 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_10 = torch.nn.functional.silu(x_se_9, inplace=True) + x_se_9 = None + x_se_11 = torch.conv2d( + x_se_10, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_10 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_2 = x_se_11.sigmoid() + x_se_11 = None + x_31 = x_30 * sigmoid_2 + x_30 = sigmoid_2 = None + x_32 = torch.conv2d( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_33 = torch.nn.functional.batch_norm( + x_32, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_32 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_34 = torch.nn.functional.silu(x_33, inplace=True) + x_33 = None + x_35 = x_34 + x_27 + x_34 = x_27 = None + cat = torch.cat([x_35, x2], dim=1) + x_35 = x2 = None + x_36 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_37 = torch.nn.functional.batch_norm( + x_36, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_36 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_38 = torch.nn.functional.silu(x_37, inplace=True) + x_37 = None + x_39 = torch.conv2d( + x_38, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_38 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_40 = torch.nn.functional.batch_norm( + x_39, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_39 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_41 = torch.nn.functional.silu(x_40, inplace=True) + x_40 = None + x_42 = torch.conv2d( + x_41, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_41 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_43 = torch.nn.functional.batch_norm( + x_42, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_42 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_44 = torch.nn.functional.silu(x_43, inplace=True) + x_43 = None + split_1 = x_44.split(160, dim=1) + x_44 = None + x1_1 = split_1[0] + x2_1 = split_1[1] + split_1 = None + x_45 = torch.conv2d( + x1_1, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_46 = torch.nn.functional.batch_norm( + x_45, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_45 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_47 = torch.nn.functional.silu(x_46, inplace=True) + x_46 = None + x_se_12 = x_47.mean((2, 3), keepdim=True) + x_se_13 = torch.conv2d( + x_se_12, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_12 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_14 = torch.nn.functional.silu(x_se_13, inplace=True) + x_se_13 = None + x_se_15 = torch.conv2d( + x_se_14, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_14 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_3 = x_se_15.sigmoid() + x_se_15 = None + x_48 = x_47 * sigmoid_3 + x_47 = sigmoid_3 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_50 = torch.nn.functional.batch_norm( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_49 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_51 = torch.nn.functional.silu(x_50, inplace=True) + x_50 = None + x_52 = x_51 + x1_1 + x_51 = x1_1 = None + x_53 = torch.conv2d( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_54 = torch.nn.functional.batch_norm( + x_53, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_53 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_55 = torch.nn.functional.silu(x_54, inplace=True) + x_54 = None + x_se_16 = x_55.mean((2, 3), keepdim=True) + x_se_17 = torch.conv2d( + x_se_16, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_16 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_18 = torch.nn.functional.silu(x_se_17, inplace=True) + x_se_17 = None + x_se_19 = torch.conv2d( + x_se_18, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_18 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_4 = x_se_19.sigmoid() + x_se_19 = None + x_56 = x_55 * sigmoid_4 + x_55 = sigmoid_4 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_58 = torch.nn.functional.batch_norm( + x_57, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_57 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_59 = torch.nn.functional.silu(x_58, inplace=True) + x_58 = None + x_60 = x_59 + x_52 + x_59 = x_52 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_62 = torch.nn.functional.batch_norm( + x_61, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_63 = torch.nn.functional.silu(x_62, inplace=True) + x_62 = None + x_se_20 = x_63.mean((2, 3), keepdim=True) + x_se_21 = torch.conv2d( + x_se_20, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_20 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_22 = torch.nn.functional.silu(x_se_21, inplace=True) + x_se_21 = None + x_se_23 = torch.conv2d( + x_se_22, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_22 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_5 = x_se_23.sigmoid() + x_se_23 = None + x_64 = x_63 * sigmoid_5 + x_63 = sigmoid_5 = None + x_65 = torch.conv2d( + x_64, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_64 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_66 = torch.nn.functional.batch_norm( + x_65, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_65 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_67 = torch.nn.functional.silu(x_66, inplace=True) + x_66 = None + x_68 = x_67 + x_60 + x_67 = x_60 = None + x_69 = torch.conv2d( + x_68, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_70 = torch.nn.functional.batch_norm( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_71 = torch.nn.functional.silu(x_70, inplace=True) + x_70 = None + x_se_24 = x_71.mean((2, 3), keepdim=True) + x_se_25 = torch.conv2d( + x_se_24, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_24 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_26 = torch.nn.functional.silu(x_se_25, inplace=True) + x_se_25 = None + x_se_27 = torch.conv2d( + x_se_26, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_26 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_6 = x_se_27.sigmoid() + x_se_27 = None + x_72 = x_71 * sigmoid_6 + x_71 = sigmoid_6 = None + x_73 = torch.conv2d( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_74 = torch.nn.functional.batch_norm( + x_73, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_73 = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_75 = torch.nn.functional.silu(x_74, inplace=True) + x_74 = None + x_76 = x_75 + x_68 + x_75 = x_68 = None + x_77 = torch.conv2d( + x_76, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_78 = torch.nn.functional.batch_norm( + x_77, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_77 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_79 = torch.nn.functional.silu(x_78, inplace=True) + x_78 = None + x_se_28 = x_79.mean((2, 3), keepdim=True) + x_se_29 = torch.conv2d( + x_se_28, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_28 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_30 = torch.nn.functional.silu(x_se_29, inplace=True) + x_se_29 = None + x_se_31 = torch.conv2d( + x_se_30, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_30 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_7 = x_se_31.sigmoid() + x_se_31 = None + x_80 = x_79 * sigmoid_7 + x_79 = sigmoid_7 = None + x_81 = torch.conv2d( + x_80, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_80 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_82 = torch.nn.functional.batch_norm( + x_81, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_81 = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_83 = torch.nn.functional.silu(x_82, inplace=True) + x_82 = None + x_84 = x_83 + x_76 + x_83 = x_76 = None + x_85 = torch.conv2d( + x_84, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_86 = torch.nn.functional.batch_norm( + x_85, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_85 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_87 = torch.nn.functional.silu(x_86, inplace=True) + x_86 = None + x_se_32 = x_87.mean((2, 3), keepdim=True) + x_se_33 = torch.conv2d( + x_se_32, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_32 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_34 = torch.nn.functional.silu(x_se_33, inplace=True) + x_se_33 = None + x_se_35 = torch.conv2d( + x_se_34, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_34 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_8 = x_se_35.sigmoid() + x_se_35 = None + x_88 = x_87 * sigmoid_8 + x_87 = sigmoid_8 = None + x_89 = torch.conv2d( + x_88, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_88 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_90 = torch.nn.functional.batch_norm( + x_89, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_89 = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_91 = torch.nn.functional.silu(x_90, inplace=True) + x_90 = None + x_92 = x_91 + x_84 + x_91 = x_84 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_94 = torch.nn.functional.batch_norm( + x_93, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_93 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_95 = torch.nn.functional.silu(x_94, inplace=True) + x_94 = None + x_se_36 = x_95.mean((2, 3), keepdim=True) + x_se_37 = torch.conv2d( + x_se_36, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_36 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_38 = torch.nn.functional.silu(x_se_37, inplace=True) + x_se_37 = None + x_se_39 = torch.conv2d( + x_se_38, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_38 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_9 = x_se_39.sigmoid() + x_se_39 = None + x_96 = x_95 * sigmoid_9 + x_95 = sigmoid_9 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_96 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_98 = torch.nn.functional.batch_norm( + x_97, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_97 = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_99 = torch.nn.functional.silu(x_98, inplace=True) + x_98 = None + x_100 = x_99 + x_92 + x_99 = x_92 = None + cat_1 = torch.cat([x_100, x2_1], dim=1) + x_100 = x2_1 = None + x_101 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_102 = torch.nn.functional.batch_norm( + x_101, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_101 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_103 = torch.nn.functional.silu(x_102, inplace=True) + x_102 = None + x_104 = torch.conv2d( + x_103, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_103 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_105 = torch.nn.functional.batch_norm( + x_104, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_104 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_106 = torch.nn.functional.silu(x_105, inplace=True) + x_105 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_106 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_108 = torch.nn.functional.batch_norm( + x_107, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_107 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_109 = torch.nn.functional.silu(x_108, inplace=True) + x_108 = None + split_2 = x_109.split(320, dim=1) + x_109 = None + x1_2 = split_2[0] + x2_2 = split_2[1] + split_2 = None + x_110 = torch.conv2d( + x1_2, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_111 = torch.nn.functional.batch_norm( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_112 = torch.nn.functional.silu(x_111, inplace=True) + x_111 = None + x_se_40 = x_112.mean((2, 3), keepdim=True) + x_se_41 = torch.conv2d( + x_se_40, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_40 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_42 = torch.nn.functional.silu(x_se_41, inplace=True) + x_se_41 = None + x_se_43 = torch.conv2d( + x_se_42, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_42 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_10 = x_se_43.sigmoid() + x_se_43 = None + x_113 = x_112 * sigmoid_10 + x_112 = sigmoid_10 = None + x_114 = torch.conv2d( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_113 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_115 = torch.nn.functional.batch_norm( + x_114, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_114 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_116 = torch.nn.functional.silu(x_115, inplace=True) + x_115 = None + x_117 = x_116 + x1_2 + x_116 = x1_2 = None + x_118 = torch.conv2d( + x_117, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_119 = torch.nn.functional.batch_norm( + x_118, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_118 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_120 = torch.nn.functional.silu(x_119, inplace=True) + x_119 = None + x_se_44 = x_120.mean((2, 3), keepdim=True) + x_se_45 = torch.conv2d( + x_se_44, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_44 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_46 = torch.nn.functional.silu(x_se_45, inplace=True) + x_se_45 = None + x_se_47 = torch.conv2d( + x_se_46, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_46 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_11 = x_se_47.sigmoid() + x_se_47 = None + x_121 = x_120 * sigmoid_11 + x_120 = sigmoid_11 = None + x_122 = torch.conv2d( + x_121, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_121 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_123 = torch.nn.functional.batch_norm( + x_122, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_124 = torch.nn.functional.silu(x_123, inplace=True) + x_123 = None + x_125 = x_124 + x_117 + x_124 = x_117 = None + x_126 = torch.conv2d( + x_125, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_127 = torch.nn.functional.batch_norm( + x_126, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_126 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_128 = torch.nn.functional.silu(x_127, inplace=True) + x_127 = None + x_se_48 = x_128.mean((2, 3), keepdim=True) + x_se_49 = torch.conv2d( + x_se_48, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_48 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_50 = torch.nn.functional.silu(x_se_49, inplace=True) + x_se_49 = None + x_se_51 = torch.conv2d( + x_se_50, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_50 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_12 = x_se_51.sigmoid() + x_se_51 = None + x_129 = x_128 * sigmoid_12 + x_128 = sigmoid_12 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_131 = torch.nn.functional.batch_norm( + x_130, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_130 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_132 = torch.nn.functional.silu(x_131, inplace=True) + x_131 = None + x_133 = x_132 + x_125 + x_132 = x_125 = None + x_134 = torch.conv2d( + x_133, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_135 = torch.nn.functional.batch_norm( + x_134, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_134 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_136 = torch.nn.functional.silu(x_135, inplace=True) + x_135 = None + x_se_52 = x_136.mean((2, 3), keepdim=True) + x_se_53 = torch.conv2d( + x_se_52, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_52 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_54 = torch.nn.functional.silu(x_se_53, inplace=True) + x_se_53 = None + x_se_55 = torch.conv2d( + x_se_54, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_54 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_13 = x_se_55.sigmoid() + x_se_55 = None + x_137 = x_136 * sigmoid_13 + x_136 = sigmoid_13 = None + x_138 = torch.conv2d( + x_137, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_137 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_139 = torch.nn.functional.batch_norm( + x_138, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_138 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_140 = torch.nn.functional.silu(x_139, inplace=True) + x_139 = None + x_141 = x_140 + x_133 + x_140 = x_133 = None + x_142 = torch.conv2d( + x_141, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_143 = torch.nn.functional.batch_norm( + x_142, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_142 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_144 = torch.nn.functional.silu(x_143, inplace=True) + x_143 = None + x_se_56 = x_144.mean((2, 3), keepdim=True) + x_se_57 = torch.conv2d( + x_se_56, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_56 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_58 = torch.nn.functional.silu(x_se_57, inplace=True) + x_se_57 = None + x_se_59 = torch.conv2d( + x_se_58, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_58 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_14 = x_se_59.sigmoid() + x_se_59 = None + x_145 = x_144 * sigmoid_14 + x_144 = sigmoid_14 = None + x_146 = torch.conv2d( + x_145, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_145 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_147 = torch.nn.functional.batch_norm( + x_146, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_146 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_148 = torch.nn.functional.silu(x_147, inplace=True) + x_147 = None + x_149 = x_148 + x_141 + x_148 = x_141 = None + x_150 = torch.conv2d( + x_149, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_151 = torch.nn.functional.batch_norm( + x_150, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_150 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_152 = torch.nn.functional.silu(x_151, inplace=True) + x_151 = None + x_se_60 = x_152.mean((2, 3), keepdim=True) + x_se_61 = torch.conv2d( + x_se_60, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_60 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_62 = torch.nn.functional.silu(x_se_61, inplace=True) + x_se_61 = None + x_se_63 = torch.conv2d( + x_se_62, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_62 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_15 = x_se_63.sigmoid() + x_se_63 = None + x_153 = x_152 * sigmoid_15 + x_152 = sigmoid_15 = None + x_154 = torch.conv2d( + x_153, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_153 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_155 = torch.nn.functional.batch_norm( + x_154, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_154 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_156 = torch.nn.functional.silu(x_155, inplace=True) + x_155 = None + x_157 = x_156 + x_149 + x_156 = x_149 = None + x_158 = torch.conv2d( + x_157, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_159 = torch.nn.functional.batch_norm( + x_158, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_158 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_160 = torch.nn.functional.silu(x_159, inplace=True) + x_159 = None + x_se_64 = x_160.mean((2, 3), keepdim=True) + x_se_65 = torch.conv2d( + x_se_64, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_64 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_66 = torch.nn.functional.silu(x_se_65, inplace=True) + x_se_65 = None + x_se_67 = torch.conv2d( + x_se_66, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_66 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_16 = x_se_67.sigmoid() + x_se_67 = None + x_161 = x_160 * sigmoid_16 + x_160 = sigmoid_16 = None + x_162 = torch.conv2d( + x_161, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_161 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_163 = torch.nn.functional.batch_norm( + x_162, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_162 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_164 = torch.nn.functional.silu(x_163, inplace=True) + x_163 = None + x_165 = x_164 + x_157 + x_164 = x_157 = None + x_166 = torch.conv2d( + x_165, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_167 = torch.nn.functional.batch_norm( + x_166, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_166 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_168 = torch.nn.functional.silu(x_167, inplace=True) + x_167 = None + x_se_68 = x_168.mean((2, 3), keepdim=True) + x_se_69 = torch.conv2d( + x_se_68, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_68 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_70 = torch.nn.functional.silu(x_se_69, inplace=True) + x_se_69 = None + x_se_71 = torch.conv2d( + x_se_70, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_70 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_17 = x_se_71.sigmoid() + x_se_71 = None + x_169 = x_168 * sigmoid_17 + x_168 = sigmoid_17 = None + x_170 = torch.conv2d( + x_169, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_169 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_171 = torch.nn.functional.batch_norm( + x_170, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_170 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_172 = torch.nn.functional.silu(x_171, inplace=True) + x_171 = None + x_173 = x_172 + x_165 + x_172 = x_165 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_175 = torch.nn.functional.batch_norm( + x_174, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_174 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_ = (None) + x_176 = torch.nn.functional.silu(x_175, inplace=True) + x_175 = None + x_se_72 = x_176.mean((2, 3), keepdim=True) + x_se_73 = torch.conv2d( + x_se_72, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_72 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_74 = torch.nn.functional.silu(x_se_73, inplace=True) + x_se_73 = None + x_se_75 = torch.conv2d( + x_se_74, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_74 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_18 = x_se_75.sigmoid() + x_se_75 = None + x_177 = x_176 * sigmoid_18 + x_176 = sigmoid_18 = None + x_178 = torch.conv2d( + x_177, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_177 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_ = (None) + x_179 = torch.nn.functional.batch_norm( + x_178, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_178 = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_ = (None) + x_180 = torch.nn.functional.silu(x_179, inplace=True) + x_179 = None + x_181 = x_180 + x_173 + x_180 = x_173 = None + x_182 = torch.conv2d( + x_181, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_183 = torch.nn.functional.batch_norm( + x_182, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_182 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_ = (None) + x_184 = torch.nn.functional.silu(x_183, inplace=True) + x_183 = None + x_se_76 = x_184.mean((2, 3), keepdim=True) + x_se_77 = torch.conv2d( + x_se_76, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_76 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_78 = torch.nn.functional.silu(x_se_77, inplace=True) + x_se_77 = None + x_se_79 = torch.conv2d( + x_se_78, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_78 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_19 = x_se_79.sigmoid() + x_se_79 = None + x_185 = x_184 * sigmoid_19 + x_184 = sigmoid_19 = None + x_186 = torch.conv2d( + x_185, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_185 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_ = (None) + x_187 = torch.nn.functional.batch_norm( + x_186, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_186 = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_ = (None) + x_188 = torch.nn.functional.silu(x_187, inplace=True) + x_187 = None + x_189 = x_188 + x_181 + x_188 = x_181 = None + x_190 = torch.conv2d( + x_189, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_191 = torch.nn.functional.batch_norm( + x_190, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_190 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_ = (None) + x_192 = torch.nn.functional.silu(x_191, inplace=True) + x_191 = None + x_se_80 = x_192.mean((2, 3), keepdim=True) + x_se_81 = torch.conv2d( + x_se_80, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_80 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_82 = torch.nn.functional.silu(x_se_81, inplace=True) + x_se_81 = None + x_se_83 = torch.conv2d( + x_se_82, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_82 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_20 = x_se_83.sigmoid() + x_se_83 = None + x_193 = x_192 * sigmoid_20 + x_192 = sigmoid_20 = None + x_194 = torch.conv2d( + x_193, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_193 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_ = (None) + x_195 = torch.nn.functional.batch_norm( + x_194, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_194 = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_ = (None) + x_196 = torch.nn.functional.silu(x_195, inplace=True) + x_195 = None + x_197 = x_196 + x_189 + x_196 = x_189 = None + cat_2 = torch.cat([x_197, x2_2], dim=1) + x_197 = x2_2 = None + x_198 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_199 = torch.nn.functional.batch_norm( + x_198, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_198 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_200 = torch.nn.functional.silu(x_199, inplace=True) + x_199 = None + x_201 = torch.conv2d( + x_200, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_200 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_202 = torch.nn.functional.batch_norm( + x_201, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_201 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_203 = torch.nn.functional.silu(x_202, inplace=True) + x_202 = None + x_204 = torch.conv2d( + x_203, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_203 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_205 = torch.nn.functional.batch_norm( + x_204, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_204 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_206 = torch.nn.functional.silu(x_205, inplace=True) + x_205 = None + split_3 = x_206.split(640, dim=1) + x_206 = None + x1_3 = split_3[0] + x2_3 = split_3[1] + split_3 = None + x_207 = torch.conv2d( + x1_3, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_208 = torch.nn.functional.batch_norm( + x_207, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_207 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_209 = torch.nn.functional.silu(x_208, inplace=True) + x_208 = None + x_se_84 = x_209.mean((2, 3), keepdim=True) + x_se_85 = torch.conv2d( + x_se_84, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_84 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_86 = torch.nn.functional.silu(x_se_85, inplace=True) + x_se_85 = None + x_se_87 = torch.conv2d( + x_se_86, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_86 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_21 = x_se_87.sigmoid() + x_se_87 = None + x_210 = x_209 * sigmoid_21 + x_209 = sigmoid_21 = None + x_211 = torch.conv2d( + x_210, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_210 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_212 = torch.nn.functional.batch_norm( + x_211, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_211 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_213 = torch.nn.functional.silu(x_212, inplace=True) + x_212 = None + x_214 = x_213 + x1_3 + x_213 = x1_3 = None + x_215 = torch.conv2d( + x_214, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_216 = torch.nn.functional.batch_norm( + x_215, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_215 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_217 = torch.nn.functional.silu(x_216, inplace=True) + x_216 = None + x_se_88 = x_217.mean((2, 3), keepdim=True) + x_se_89 = torch.conv2d( + x_se_88, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_88 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_90 = torch.nn.functional.silu(x_se_89, inplace=True) + x_se_89 = None + x_se_91 = torch.conv2d( + x_se_90, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_90 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_22 = x_se_91.sigmoid() + x_se_91 = None + x_218 = x_217 * sigmoid_22 + x_217 = sigmoid_22 = None + x_219 = torch.conv2d( + x_218, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_218 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_220 = torch.nn.functional.batch_norm( + x_219, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_219 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_221 = torch.nn.functional.silu(x_220, inplace=True) + x_220 = None + x_222 = x_221 + x_214 + x_221 = x_214 = None + x_223 = torch.conv2d( + x_222, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_224 = torch.nn.functional.batch_norm( + x_223, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_223 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_225 = torch.nn.functional.silu(x_224, inplace=True) + x_224 = None + x_se_92 = x_225.mean((2, 3), keepdim=True) + x_se_93 = torch.conv2d( + x_se_92, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_92 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_ = (None) + x_se_94 = torch.nn.functional.silu(x_se_93, inplace=True) + x_se_93 = None + x_se_95 = torch.conv2d( + x_se_94, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_se_94 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_ = (None) + sigmoid_23 = x_se_95.sigmoid() + x_se_95 = None + x_226 = x_225 * sigmoid_23 + x_225 = sigmoid_23 = None + x_227 = torch.conv2d( + x_226, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_226 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_228 = torch.nn.functional.batch_norm( + x_227, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_227 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_229 = torch.nn.functional.silu(x_228, inplace=True) + x_228 = None + x_230 = x_229 + x_222 + x_229 = x_222 = None + cat_3 = torch.cat([x_230, x2_3], dim=1) + x_230 = x2_3 = None + x_231 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_232 = torch.nn.functional.batch_norm( + x_231, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_231 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_233 = torch.nn.functional.silu(x_232, inplace=True) + x_232 = None + x_234 = torch.nn.functional.adaptive_avg_pool2d(x_233, 1) + x_233 = None + x_235 = x_234.flatten(1, -1) + x_234 = None + x_236 = torch.nn.functional.dropout(x_235, 0.0, False, False) + x_235 = None + x_237 = torch._C._nn.linear( + x_236, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_236 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_237,) diff --git a/samples/timm/cs3sedarknet_x.c2ns_in1k/weight_meta.py b/samples/timm/cs3sedarknet_x.c2ns_in1k/weight_meta.py new file mode 100644 index 0000000000..4a7c009e89 --- /dev/null +++ b/samples/timm/cs3sedarknet_x.c2ns_in1k/weight_meta.py @@ -0,0 +1,4122 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [40, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.075 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.221 + std = 1.284 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 40, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_" + shape = [80, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.052 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [160, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [80, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.159 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_" + shape = [8, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.505 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_" + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_" + shape = [80, 8, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.008 + std = 0.150 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [80, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.156 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_" + shape = [8, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.020 + std = 0.498 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_" + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_" + shape = [80, 8, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.008 + std = 0.155 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [80, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.158 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_" + shape = [8, 80, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.020 + std = 0.503 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_" + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_" + shape = [80, 8, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.006 + std = 0.154 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [80, 80, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [80] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.113 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [320, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_" + shape = [8, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.029 + std = 0.484 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_" + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_" + shape = [160, 8, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.113 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_" + shape = [8, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.013 + std = 0.499 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_" + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_" + shape = [160, 8, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.114 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_" + shape = [8, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.486 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_" + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_" + shape = [160, 8, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.113 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_" + shape = [8, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.013 + std = 0.503 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_" + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_" + shape = [160, 8, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.109 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_" + shape = [8, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.491 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_" + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_" + shape = [160, 8, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.114 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_" + shape = [8, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.018 + std = 0.501 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_" + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_" + shape = [160, 8, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.113 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [160, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_" + shape = [8, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.013 + std = 0.484 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_" + shape = [8] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_" + shape = [160, 8, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.110 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [160, 160, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [640, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_" + shape = [24, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.289 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_" + shape = [320, 24, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.080 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_" + shape = [24, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_" + shape = [320, 24, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.078 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_" + shape = [24, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.292 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_" + shape = [320, 24, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_weight_" + shape = [24, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.290 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc1_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_weight_" + shape = [320, 24, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.080 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_attn_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_weight_" + shape = [24, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc1_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_weight_" + shape = [320, 24, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_attn_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_weight_" + shape = [24, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.006 + std = 0.288 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc1_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_weight_" + shape = [320, 24, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_attn_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_weight_" + shape = [24, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.289 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc1_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_weight_" + shape = [320, 24, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_attn_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_weight_" + shape = [24, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.290 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc1_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_weight_" + shape = [320, 24, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_attn_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_weight_" + shape = [24, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.291 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc1_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_weight_" + shape = [320, 24, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_attn_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_8_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_weight_" + shape = [24, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.292 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc1_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_weight_" + shape = [320, 24, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.080 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_attn_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_9_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_conv_parameters_weight_" + shape = [320, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv1_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_weight_" + shape = [24, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc1_parameters_bias_" + shape = [24] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_weight_" + shape = [320, 24, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.080 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_attn_modules_fc2_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_conv_parameters_weight_" + shape = [320, 320, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.026 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_10_modules_conv2_modules_bn_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [640, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [640, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [1280, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [640, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_weight_" + shape = [40, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.223 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc1_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_weight_" + shape = [640, 40, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_attn_modules_fc2_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [640, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [640, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_weight_" + shape = [40, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.224 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc1_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_weight_" + shape = [640, 40, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_attn_modules_fc2_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [640, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [640, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_weight_" + shape = [40, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.225 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc1_parameters_bias_" + shape = [40] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_weight_" + shape = [640, 40, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_attn_modules_fc2_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [640, 640, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.019 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [1280, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cspdarknet53.ra_in1k/graph_hash.txt b/samples/timm/cspdarknet53.ra_in1k/graph_hash.txt new file mode 100644 index 0000000000..1a7ee185dd --- /dev/null +++ b/samples/timm/cspdarknet53.ra_in1k/graph_hash.txt @@ -0,0 +1 @@ +4510f5a457c81963c81e8519198a97e4628f0f5671dd4fe01bce02082834404a \ No newline at end of file diff --git a/samples/timm/cspdarknet53.ra_in1k/graph_net.json b/samples/timm/cspdarknet53.ra_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/cspdarknet53.ra_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cspdarknet53.ra_in1k/input_meta.py b/samples/timm/cspdarknet53.ra_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cspdarknet53.ra_in1k/input_tensor_constraints.py b/samples/timm/cspdarknet53.ra_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cspdarknet53.ra_in1k/model.py b/samples/timm/cspdarknet53.ra_in1k/model.py new file mode 100644 index 0000000000..28566124bc --- /dev/null +++ b/samples/timm/cspdarknet53.ra_in1k/model.py @@ -0,0 +1,2417 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.leaky_relu(x_1, 0.01, True) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_5 = torch.nn.functional.leaky_relu(x_4, 0.01, True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_5 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_8 = torch.nn.functional.leaky_relu(x_7, 0.01, True) + x_7 = None + split = x_8.split(64, dim=1) + x_8 = None + xs = split[0] + xb = split[1] + split = None + x_9 = torch.conv2d( + xb, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_11 = torch.nn.functional.leaky_relu(x_10, 0.01, True) + x_10 = None + x_12 = torch.conv2d( + x_11, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_11 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_13 = torch.nn.functional.batch_norm( + x_12, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_12 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_14 = torch.nn.functional.leaky_relu(x_13, 0.01, True) + x_13 = None + x_15 = x_14 + xb + x_14 = xb = None + x_16 = torch.conv2d( + x_15, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_15 = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_17 = torch.nn.functional.batch_norm( + x_16, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_16 = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_18 = torch.nn.functional.leaky_relu(x_17, 0.01, True) + x_17 = None + xb_1 = x_18.contiguous() + x_18 = None + cat = torch.cat([xs, xb_1], dim=1) + xs = xb_1 = None + x_19 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_20 = torch.nn.functional.batch_norm( + x_19, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_19 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_21 = torch.nn.functional.leaky_relu(x_20, 0.01, True) + x_20 = None + x_22 = torch.conv2d( + x_21, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_21 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_23 = torch.nn.functional.batch_norm( + x_22, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_22 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_24 = torch.nn.functional.leaky_relu(x_23, 0.01, True) + x_23 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_24 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_26 = torch.nn.functional.batch_norm( + x_25, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_25 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_27 = torch.nn.functional.leaky_relu(x_26, 0.01, True) + x_26 = None + split_1 = x_27.split(64, dim=1) + x_27 = None + xs_1 = split_1[0] + xb_2 = split_1[1] + split_1 = None + x_28 = torch.conv2d( + xb_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_29 = torch.nn.functional.batch_norm( + x_28, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_28 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_30 = torch.nn.functional.leaky_relu(x_29, 0.01, True) + x_29 = None + x_31 = torch.conv2d( + x_30, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_30 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_32 = torch.nn.functional.batch_norm( + x_31, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_31 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_33 = torch.nn.functional.leaky_relu(x_32, 0.01, True) + x_32 = None + x_34 = x_33 + xb_2 + x_33 = xb_2 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_36 = torch.nn.functional.batch_norm( + x_35, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_35 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_37 = torch.nn.functional.leaky_relu(x_36, 0.01, True) + x_36 = None + x_38 = torch.conv2d( + x_37, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_37 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_39 = torch.nn.functional.batch_norm( + x_38, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_38 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_40 = torch.nn.functional.leaky_relu(x_39, 0.01, True) + x_39 = None + x_41 = x_40 + x_34 + x_40 = x_34 = None + x_42 = torch.conv2d( + x_41, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_41 = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_43 = torch.nn.functional.batch_norm( + x_42, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_42 = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_44 = torch.nn.functional.leaky_relu(x_43, 0.01, True) + x_43 = None + xb_3 = x_44.contiguous() + x_44 = None + cat_1 = torch.cat([xs_1, xb_3], dim=1) + xs_1 = xb_3 = None + x_45 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_46 = torch.nn.functional.batch_norm( + x_45, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_45 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_47 = torch.nn.functional.leaky_relu(x_46, 0.01, True) + x_46 = None + x_48 = torch.conv2d( + x_47, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_47 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_49 = torch.nn.functional.batch_norm( + x_48, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_48 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_50 = torch.nn.functional.leaky_relu(x_49, 0.01, True) + x_49 = None + x_51 = torch.conv2d( + x_50, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_50 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_52 = torch.nn.functional.batch_norm( + x_51, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_51 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_53 = torch.nn.functional.leaky_relu(x_52, 0.01, True) + x_52 = None + split_2 = x_53.split(128, dim=1) + x_53 = None + xs_2 = split_2[0] + xb_4 = split_2[1] + split_2 = None + x_54 = torch.conv2d( + xb_4, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_55 = torch.nn.functional.batch_norm( + x_54, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_56 = torch.nn.functional.leaky_relu(x_55, 0.01, True) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_58 = torch.nn.functional.batch_norm( + x_57, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_57 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_59 = torch.nn.functional.leaky_relu(x_58, 0.01, True) + x_58 = None + x_60 = x_59 + xb_4 + x_59 = xb_4 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_62 = torch.nn.functional.batch_norm( + x_61, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_61 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_63 = torch.nn.functional.leaky_relu(x_62, 0.01, True) + x_62 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_63 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_65 = torch.nn.functional.batch_norm( + x_64, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_64 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_66 = torch.nn.functional.leaky_relu(x_65, 0.01, True) + x_65 = None + x_67 = x_66 + x_60 + x_66 = x_60 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_69 = torch.nn.functional.batch_norm( + x_68, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_68 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_70 = torch.nn.functional.leaky_relu(x_69, 0.01, True) + x_69 = None + x_71 = torch.conv2d( + x_70, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_70 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_72 = torch.nn.functional.batch_norm( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_73 = torch.nn.functional.leaky_relu(x_72, 0.01, True) + x_72 = None + x_74 = x_73 + x_67 + x_73 = x_67 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_76 = torch.nn.functional.batch_norm( + x_75, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_75 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_77 = torch.nn.functional.leaky_relu(x_76, 0.01, True) + x_76 = None + x_78 = torch.conv2d( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_79 = torch.nn.functional.batch_norm( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_80 = torch.nn.functional.leaky_relu(x_79, 0.01, True) + x_79 = None + x_81 = x_80 + x_74 + x_80 = x_74 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_83 = torch.nn.functional.batch_norm( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_82 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_84 = torch.nn.functional.leaky_relu(x_83, 0.01, True) + x_83 = None + x_85 = torch.conv2d( + x_84, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_84 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_86 = torch.nn.functional.batch_norm( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_85 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_87 = torch.nn.functional.leaky_relu(x_86, 0.01, True) + x_86 = None + x_88 = x_87 + x_81 + x_87 = x_81 = None + x_89 = torch.conv2d( + x_88, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_90 = torch.nn.functional.batch_norm( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_91 = torch.nn.functional.leaky_relu(x_90, 0.01, True) + x_90 = None + x_92 = torch.conv2d( + x_91, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_91 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_93 = torch.nn.functional.batch_norm( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_94 = torch.nn.functional.leaky_relu(x_93, 0.01, True) + x_93 = None + x_95 = x_94 + x_88 + x_94 = x_88 = None + x_96 = torch.conv2d( + x_95, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_97 = torch.nn.functional.batch_norm( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_96 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_98 = torch.nn.functional.leaky_relu(x_97, 0.01, True) + x_97 = None + x_99 = torch.conv2d( + x_98, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_98 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_100 = torch.nn.functional.batch_norm( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_101 = torch.nn.functional.leaky_relu(x_100, 0.01, True) + x_100 = None + x_102 = x_101 + x_95 + x_101 = x_95 = None + x_103 = torch.conv2d( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_104 = torch.nn.functional.batch_norm( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_105 = torch.nn.functional.leaky_relu(x_104, 0.01, True) + x_104 = None + x_106 = torch.conv2d( + x_105, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_105 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_107 = torch.nn.functional.batch_norm( + x_106, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_106 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_108 = torch.nn.functional.leaky_relu(x_107, 0.01, True) + x_107 = None + x_109 = x_108 + x_102 + x_108 = x_102 = None + x_110 = torch.conv2d( + x_109, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_109 = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_111 = torch.nn.functional.batch_norm( + x_110, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_110 = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_112 = torch.nn.functional.leaky_relu(x_111, 0.01, True) + x_111 = None + xb_5 = x_112.contiguous() + x_112 = None + cat_2 = torch.cat([xs_2, xb_5], dim=1) + xs_2 = xb_5 = None + x_113 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_114 = torch.nn.functional.batch_norm( + x_113, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_113 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_115 = torch.nn.functional.leaky_relu(x_114, 0.01, True) + x_114 = None + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_117 = torch.nn.functional.batch_norm( + x_116, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_116 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_118 = torch.nn.functional.leaky_relu(x_117, 0.01, True) + x_117 = None + x_119 = torch.conv2d( + x_118, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_118 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_120 = torch.nn.functional.batch_norm( + x_119, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_119 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_121 = torch.nn.functional.leaky_relu(x_120, 0.01, True) + x_120 = None + split_3 = x_121.split(256, dim=1) + x_121 = None + xs_3 = split_3[0] + xb_6 = split_3[1] + split_3 = None + x_122 = torch.conv2d( + xb_6, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_123 = torch.nn.functional.batch_norm( + x_122, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_122 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_124 = torch.nn.functional.leaky_relu(x_123, 0.01, True) + x_123 = None + x_125 = torch.conv2d( + x_124, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_124 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_126 = torch.nn.functional.batch_norm( + x_125, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_125 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_127 = torch.nn.functional.leaky_relu(x_126, 0.01, True) + x_126 = None + x_128 = x_127 + xb_6 + x_127 = xb_6 = None + x_129 = torch.conv2d( + x_128, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_130 = torch.nn.functional.batch_norm( + x_129, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_129 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_131 = torch.nn.functional.leaky_relu(x_130, 0.01, True) + x_130 = None + x_132 = torch.conv2d( + x_131, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_131 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_133 = torch.nn.functional.batch_norm( + x_132, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_132 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_134 = torch.nn.functional.leaky_relu(x_133, 0.01, True) + x_133 = None + x_135 = x_134 + x_128 + x_134 = x_128 = None + x_136 = torch.conv2d( + x_135, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_137 = torch.nn.functional.batch_norm( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_138 = torch.nn.functional.leaky_relu(x_137, 0.01, True) + x_137 = None + x_139 = torch.conv2d( + x_138, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_138 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_140 = torch.nn.functional.batch_norm( + x_139, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_139 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_141 = torch.nn.functional.leaky_relu(x_140, 0.01, True) + x_140 = None + x_142 = x_141 + x_135 + x_141 = x_135 = None + x_143 = torch.conv2d( + x_142, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_144 = torch.nn.functional.batch_norm( + x_143, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_143 = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_145 = torch.nn.functional.leaky_relu(x_144, 0.01, True) + x_144 = None + x_146 = torch.conv2d( + x_145, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_145 = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_147 = torch.nn.functional.batch_norm( + x_146, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_146 = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_148 = torch.nn.functional.leaky_relu(x_147, 0.01, True) + x_147 = None + x_149 = x_148 + x_142 + x_148 = x_142 = None + x_150 = torch.conv2d( + x_149, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_151 = torch.nn.functional.batch_norm( + x_150, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_150 = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_152 = torch.nn.functional.leaky_relu(x_151, 0.01, True) + x_151 = None + x_153 = torch.conv2d( + x_152, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_152 = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_154 = torch.nn.functional.batch_norm( + x_153, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_153 = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_155 = torch.nn.functional.leaky_relu(x_154, 0.01, True) + x_154 = None + x_156 = x_155 + x_149 + x_155 = x_149 = None + x_157 = torch.conv2d( + x_156, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_158 = torch.nn.functional.batch_norm( + x_157, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_157 = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_159 = torch.nn.functional.leaky_relu(x_158, 0.01, True) + x_158 = None + x_160 = torch.conv2d( + x_159, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_159 = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_161 = torch.nn.functional.batch_norm( + x_160, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_160 = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_162 = torch.nn.functional.leaky_relu(x_161, 0.01, True) + x_161 = None + x_163 = x_162 + x_156 + x_162 = x_156 = None + x_164 = torch.conv2d( + x_163, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_165 = torch.nn.functional.batch_norm( + x_164, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_164 = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_166 = torch.nn.functional.leaky_relu(x_165, 0.01, True) + x_165 = None + x_167 = torch.conv2d( + x_166, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_166 = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_168 = torch.nn.functional.batch_norm( + x_167, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_167 = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_169 = torch.nn.functional.leaky_relu(x_168, 0.01, True) + x_168 = None + x_170 = x_169 + x_163 + x_169 = x_163 = None + x_171 = torch.conv2d( + x_170, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_172 = torch.nn.functional.batch_norm( + x_171, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_171 = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_173 = torch.nn.functional.leaky_relu(x_172, 0.01, True) + x_172 = None + x_174 = torch.conv2d( + x_173, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_173 = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_175 = torch.nn.functional.batch_norm( + x_174, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_174 = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_176 = torch.nn.functional.leaky_relu(x_175, 0.01, True) + x_175 = None + x_177 = x_176 + x_170 + x_176 = x_170 = None + x_178 = torch.conv2d( + x_177, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_177 = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_179 = torch.nn.functional.batch_norm( + x_178, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_178 = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_180 = torch.nn.functional.leaky_relu(x_179, 0.01, True) + x_179 = None + xb_7 = x_180.contiguous() + x_180 = None + cat_3 = torch.cat([xs_3, xb_7], dim=1) + xs_3 = xb_7 = None + x_181 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_182 = torch.nn.functional.batch_norm( + x_181, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_181 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_183 = torch.nn.functional.leaky_relu(x_182, 0.01, True) + x_182 = None + x_184 = torch.conv2d( + x_183, + l_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_183 = l_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_185 = torch.nn.functional.batch_norm( + x_184, + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_184 = l_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_186 = torch.nn.functional.leaky_relu(x_185, 0.01, True) + x_185 = None + x_187 = torch.conv2d( + x_186, + l_self_modules_stages_modules_4_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_186 = l_self_modules_stages_modules_4_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_188 = torch.nn.functional.batch_norm( + x_187, + l_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_187 = l_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + x_189 = torch.nn.functional.leaky_relu(x_188, 0.01, True) + x_188 = None + split_4 = x_189.split(512, dim=1) + x_189 = None + xs_4 = split_4[0] + xb_8 = split_4[1] + split_4 = None + x_190 = torch.conv2d( + xb_8, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_191 = torch.nn.functional.batch_norm( + x_190, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_190 = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_192 = torch.nn.functional.leaky_relu(x_191, 0.01, True) + x_191 = None + x_193 = torch.conv2d( + x_192, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_192 = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_194 = torch.nn.functional.batch_norm( + x_193, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_193 = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_195 = torch.nn.functional.leaky_relu(x_194, 0.01, True) + x_194 = None + x_196 = x_195 + xb_8 + x_195 = xb_8 = None + x_197 = torch.conv2d( + x_196, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_198 = torch.nn.functional.batch_norm( + x_197, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_197 = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_199 = torch.nn.functional.leaky_relu(x_198, 0.01, True) + x_198 = None + x_200 = torch.conv2d( + x_199, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_199 = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_201 = torch.nn.functional.batch_norm( + x_200, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_200 = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_202 = torch.nn.functional.leaky_relu(x_201, 0.01, True) + x_201 = None + x_203 = x_202 + x_196 + x_202 = x_196 = None + x_204 = torch.conv2d( + x_203, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_205 = torch.nn.functional.batch_norm( + x_204, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_204 = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_206 = torch.nn.functional.leaky_relu(x_205, 0.01, True) + x_205 = None + x_207 = torch.conv2d( + x_206, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_206 = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_208 = torch.nn.functional.batch_norm( + x_207, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_207 = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_209 = torch.nn.functional.leaky_relu(x_208, 0.01, True) + x_208 = None + x_210 = x_209 + x_203 + x_209 = x_203 = None + x_211 = torch.conv2d( + x_210, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_212 = torch.nn.functional.batch_norm( + x_211, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_211 = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_213 = torch.nn.functional.leaky_relu(x_212, 0.01, True) + x_212 = None + x_214 = torch.conv2d( + x_213, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_213 = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_215 = torch.nn.functional.batch_norm( + x_214, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_214 = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_216 = torch.nn.functional.leaky_relu(x_215, 0.01, True) + x_215 = None + x_217 = x_216 + x_210 + x_216 = x_210 = None + x_218 = torch.conv2d( + x_217, + l_self_modules_stages_modules_4_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_217 = l_self_modules_stages_modules_4_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_219 = torch.nn.functional.batch_norm( + x_218, + l_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_218 = l_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_220 = torch.nn.functional.leaky_relu(x_219, 0.01, True) + x_219 = None + xb_9 = x_220.contiguous() + x_220 = None + cat_4 = torch.cat([xs_4, xb_9], dim=1) + xs_4 = xb_9 = None + x_221 = torch.conv2d( + cat_4, + l_self_modules_stages_modules_4_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_4 = l_self_modules_stages_modules_4_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_222 = torch.nn.functional.batch_norm( + x_221, + l_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_221 = l_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_223 = torch.nn.functional.leaky_relu(x_222, 0.01, True) + x_222 = None + x_224 = torch.nn.functional.adaptive_avg_pool2d(x_223, 1) + x_223 = None + x_225 = x_224.flatten(1, -1) + x_224 = None + x_226 = torch.nn.functional.dropout(x_225, 0.0, False, False) + x_225 = None + x_227 = torch._C._nn.linear( + x_226, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_226 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_227,) diff --git a/samples/timm/cspdarknet53.ra_in1k/weight_meta.py b/samples/timm/cspdarknet53.ra_in1k/weight_meta.py new file mode 100644 index 0000000000..4a1ed38820 --- /dev/null +++ b/samples/timm/cspdarknet53.ra_in1k/weight_meta.py @@ -0,0 +1,3418 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [32, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.222 + std = 1.284 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_" + shape = [64, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [32, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.006 + std = 0.252 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [128, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.124 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.179 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [64, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.003 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.124 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_" + shape = [1024, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_conv_exp_modules_conv_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_4_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_conv_transition_modules_conv_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_conv_transition_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cspresnet50.ra_in1k/graph_hash.txt b/samples/timm/cspresnet50.ra_in1k/graph_hash.txt new file mode 100644 index 0000000000..5d28f39ffc --- /dev/null +++ b/samples/timm/cspresnet50.ra_in1k/graph_hash.txt @@ -0,0 +1 @@ +d7eb07ed68b44726e3fb9f25d944a7607852e83651394761699a126e70a4755e \ No newline at end of file diff --git a/samples/timm/cspresnet50.ra_in1k/graph_net.json b/samples/timm/cspresnet50.ra_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/cspresnet50.ra_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cspresnet50.ra_in1k/input_meta.py b/samples/timm/cspresnet50.ra_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cspresnet50.ra_in1k/input_tensor_constraints.py b/samples/timm/cspresnet50.ra_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cspresnet50.ra_in1k/model.py b/samples/timm/cspresnet50.ra_in1k/model.py new file mode 100644 index 0000000000..34bd70bbe6 --- /dev/null +++ b/samples/timm/cspresnet50.ra_in1k/model.py @@ -0,0 +1,1964 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (2, 2), + (3, 3), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.leaky_relu(x_1, 0.01, True) + x_1 = None + input_1 = torch.nn.functional.max_pool2d( + x_2, 3, 2, 1, 1, ceil_mode=False, return_indices=False + ) + x_2 = None + x_3 = torch.conv2d( + input_1, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_1 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + split = x_4.split(128, dim=1) + x_4 = None + xs = split[0] + xb = split[1] + split = None + x_5 = torch.conv2d( + xb, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_6 = torch.nn.functional.batch_norm( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_7 = torch.nn.functional.leaky_relu(x_6, 0.01, True) + x_6 = None + x_8 = torch.conv2d( + x_7, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_7 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_9 = torch.nn.functional.batch_norm( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_10 = torch.nn.functional.leaky_relu(x_9, 0.01, True) + x_9 = None + x_11 = torch.conv2d( + x_10, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_10 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = (None) + x_12 = torch.nn.functional.batch_norm( + x_11, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_11 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = (None) + x_13 = x_12 + xb + x_12 = xb = None + x_14 = torch.nn.functional.leaky_relu(x_13, 0.01, False) + x_13 = None + x_15 = torch.conv2d( + x_14, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_16 = torch.nn.functional.batch_norm( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_17 = torch.nn.functional.leaky_relu(x_16, 0.01, True) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_19 = torch.nn.functional.batch_norm( + x_18, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_18 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_20 = torch.nn.functional.leaky_relu(x_19, 0.01, True) + x_19 = None + x_21 = torch.conv2d( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = (None) + x_22 = torch.nn.functional.batch_norm( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = (None) + x_23 = x_22 + x_14 + x_22 = x_14 = None + x_24 = torch.nn.functional.leaky_relu(x_23, 0.01, False) + x_23 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_26 = torch.nn.functional.batch_norm( + x_25, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_27 = torch.nn.functional.leaky_relu(x_26, 0.01, True) + x_26 = None + x_28 = torch.conv2d( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_29 = torch.nn.functional.batch_norm( + x_28, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_28 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_30 = torch.nn.functional.leaky_relu(x_29, 0.01, True) + x_29 = None + x_31 = torch.conv2d( + x_30, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_30 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ = (None) + x_32 = torch.nn.functional.batch_norm( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ = (None) + x_33 = x_32 + x_24 + x_32 = x_24 = None + x_34 = torch.nn.functional.leaky_relu(x_33, 0.01, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_36 = torch.nn.functional.batch_norm( + x_35, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_35 = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_37 = torch.nn.functional.leaky_relu(x_36, 0.01, True) + x_36 = None + xb_1 = x_37.contiguous() + x_37 = None + cat = torch.cat([xs, xb_1], dim=1) + xs = xb_1 = None + x_38 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_39 = torch.nn.functional.batch_norm( + x_38, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_38 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_40 = torch.nn.functional.leaky_relu(x_39, 0.01, True) + x_39 = None + x_41 = torch.conv2d( + x_40, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_40 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_42 = torch.nn.functional.batch_norm( + x_41, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_41 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_43 = torch.nn.functional.leaky_relu(x_42, 0.01, True) + x_42 = None + x_44 = torch.conv2d( + x_43, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_43 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_45 = torch.nn.functional.batch_norm( + x_44, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_44 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + split_1 = x_45.split(256, dim=1) + x_45 = None + xs_1 = split_1[0] + xb_2 = split_1[1] + split_1 = None + x_46 = torch.conv2d( + xb_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_47 = torch.nn.functional.batch_norm( + x_46, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_48 = torch.nn.functional.leaky_relu(x_47, 0.01, True) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_48 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_50 = torch.nn.functional.batch_norm( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_49 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_51 = torch.nn.functional.leaky_relu(x_50, 0.01, True) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = (None) + x_53 = torch.nn.functional.batch_norm( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = (None) + x_54 = x_53 + xb_2 + x_53 = xb_2 = None + x_55 = torch.nn.functional.leaky_relu(x_54, 0.01, False) + x_54 = None + x_56 = torch.conv2d( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_57 = torch.nn.functional.batch_norm( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_58 = torch.nn.functional.leaky_relu(x_57, 0.01, True) + x_57 = None + x_59 = torch.conv2d( + x_58, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_58 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_60 = torch.nn.functional.batch_norm( + x_59, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_61 = torch.nn.functional.leaky_relu(x_60, 0.01, True) + x_60 = None + x_62 = torch.conv2d( + x_61, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = (None) + x_63 = torch.nn.functional.batch_norm( + x_62, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = (None) + x_64 = x_63 + x_55 + x_63 = x_55 = None + x_65 = torch.nn.functional.leaky_relu(x_64, 0.01, False) + x_64 = None + x_66 = torch.conv2d( + x_65, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_67 = torch.nn.functional.batch_norm( + x_66, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_66 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_68 = torch.nn.functional.leaky_relu(x_67, 0.01, True) + x_67 = None + x_69 = torch.conv2d( + x_68, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_68 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_70 = torch.nn.functional.batch_norm( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_71 = torch.nn.functional.leaky_relu(x_70, 0.01, True) + x_70 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_71 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ = (None) + x_73 = torch.nn.functional.batch_norm( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ = (None) + x_74 = x_73 + x_65 + x_73 = x_65 = None + x_75 = torch.nn.functional.leaky_relu(x_74, 0.01, False) + x_74 = None + x_76 = torch.conv2d( + x_75, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_75 = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_77 = torch.nn.functional.batch_norm( + x_76, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_76 = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_78 = torch.nn.functional.leaky_relu(x_77, 0.01, True) + x_77 = None + xb_3 = x_78.contiguous() + x_78 = None + cat_1 = torch.cat([xs_1, xb_3], dim=1) + xs_1 = xb_3 = None + x_79 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_80 = torch.nn.functional.batch_norm( + x_79, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_79 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_81 = torch.nn.functional.leaky_relu(x_80, 0.01, True) + x_80 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_81 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_83 = torch.nn.functional.batch_norm( + x_82, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_82 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_84 = torch.nn.functional.leaky_relu(x_83, 0.01, True) + x_83 = None + x_85 = torch.conv2d( + x_84, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_84 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_86 = torch.nn.functional.batch_norm( + x_85, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_85 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + split_2 = x_86.split(512, dim=1) + x_86 = None + xs_2 = split_2[0] + xb_4 = split_2[1] + split_2 = None + x_87 = torch.conv2d( + xb_4, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_88 = torch.nn.functional.batch_norm( + x_87, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_89 = torch.nn.functional.leaky_relu(x_88, 0.01, True) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_91 = torch.nn.functional.batch_norm( + x_90, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_90 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_92 = torch.nn.functional.leaky_relu(x_91, 0.01, True) + x_91 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = (None) + x_94 = torch.nn.functional.batch_norm( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = (None) + x_95 = x_94 + xb_4 + x_94 = xb_4 = None + x_96 = torch.nn.functional.leaky_relu(x_95, 0.01, False) + x_95 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_98 = torch.nn.functional.batch_norm( + x_97, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_97 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_99 = torch.nn.functional.leaky_relu(x_98, 0.01, True) + x_98 = None + x_100 = torch.conv2d( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_101 = torch.nn.functional.batch_norm( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_102 = torch.nn.functional.leaky_relu(x_101, 0.01, True) + x_101 = None + x_103 = torch.conv2d( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = (None) + x_104 = torch.nn.functional.batch_norm( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = (None) + x_105 = x_104 + x_96 + x_104 = x_96 = None + x_106 = torch.nn.functional.leaky_relu(x_105, 0.01, False) + x_105 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_108 = torch.nn.functional.batch_norm( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_107 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_109 = torch.nn.functional.leaky_relu(x_108, 0.01, True) + x_108 = None + x_110 = torch.conv2d( + x_109, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_111 = torch.nn.functional.batch_norm( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_112 = torch.nn.functional.leaky_relu(x_111, 0.01, True) + x_111 = None + x_113 = torch.conv2d( + x_112, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_112 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ = (None) + x_114 = torch.nn.functional.batch_norm( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_113 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ = (None) + x_115 = x_114 + x_106 + x_114 = x_106 = None + x_116 = torch.nn.functional.leaky_relu(x_115, 0.01, False) + x_115 = None + x_117 = torch.conv2d( + x_116, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_118 = torch.nn.functional.batch_norm( + x_117, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_117 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_119 = torch.nn.functional.leaky_relu(x_118, 0.01, True) + x_118 = None + x_120 = torch.conv2d( + x_119, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_119 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_121 = torch.nn.functional.batch_norm( + x_120, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_122 = torch.nn.functional.leaky_relu(x_121, 0.01, True) + x_121 = None + x_123 = torch.conv2d( + x_122, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_ = (None) + x_124 = torch.nn.functional.batch_norm( + x_123, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_123 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_ = (None) + x_125 = x_124 + x_116 + x_124 = x_116 = None + x_126 = torch.nn.functional.leaky_relu(x_125, 0.01, False) + x_125 = None + x_127 = torch.conv2d( + x_126, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_128 = torch.nn.functional.batch_norm( + x_127, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_127 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_129 = torch.nn.functional.leaky_relu(x_128, 0.01, True) + x_128 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_131 = torch.nn.functional.batch_norm( + x_130, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_130 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_132 = torch.nn.functional.leaky_relu(x_131, 0.01, True) + x_131 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_ = (None) + x_134 = torch.nn.functional.batch_norm( + x_133, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_133 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_ = (None) + x_135 = x_134 + x_126 + x_134 = x_126 = None + x_136 = torch.nn.functional.leaky_relu(x_135, 0.01, False) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_138 = torch.nn.functional.batch_norm( + x_137, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_137 = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_139 = torch.nn.functional.leaky_relu(x_138, 0.01, True) + x_138 = None + xb_5 = x_139.contiguous() + x_139 = None + cat_2 = torch.cat([xs_2, xb_5], dim=1) + xs_2 = xb_5 = None + x_140 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_141 = torch.nn.functional.batch_norm( + x_140, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_140 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_142 = torch.nn.functional.leaky_relu(x_141, 0.01, True) + x_141 = None + x_143 = torch.conv2d( + x_142, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_142 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_144 = torch.nn.functional.batch_norm( + x_143, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_143 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_145 = torch.nn.functional.leaky_relu(x_144, 0.01, True) + x_144 = None + x_146 = torch.conv2d( + x_145, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_145 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_147 = torch.nn.functional.batch_norm( + x_146, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_146 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + split_3 = x_147.split(1024, dim=1) + x_147 = None + xs_3 = split_3[0] + xb_6 = split_3[1] + split_3 = None + x_148 = torch.conv2d( + xb_6, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_149 = torch.nn.functional.batch_norm( + x_148, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_148 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_150 = torch.nn.functional.leaky_relu(x_149, 0.01, True) + x_149 = None + x_151 = torch.conv2d( + x_150, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_150 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_152 = torch.nn.functional.batch_norm( + x_151, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_151 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_153 = torch.nn.functional.leaky_relu(x_152, 0.01, True) + x_152 = None + x_154 = torch.conv2d( + x_153, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_153 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = (None) + x_155 = torch.nn.functional.batch_norm( + x_154, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_154 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = (None) + x_156 = x_155 + xb_6 + x_155 = xb_6 = None + x_157 = torch.nn.functional.leaky_relu(x_156, 0.01, False) + x_156 = None + x_158 = torch.conv2d( + x_157, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_159 = torch.nn.functional.batch_norm( + x_158, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_158 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_160 = torch.nn.functional.leaky_relu(x_159, 0.01, True) + x_159 = None + x_161 = torch.conv2d( + x_160, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_160 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_162 = torch.nn.functional.batch_norm( + x_161, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_161 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_163 = torch.nn.functional.leaky_relu(x_162, 0.01, True) + x_162 = None + x_164 = torch.conv2d( + x_163, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_163 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = (None) + x_165 = torch.nn.functional.batch_norm( + x_164, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_164 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = (None) + x_166 = x_165 + x_157 + x_165 = x_157 = None + x_167 = torch.nn.functional.leaky_relu(x_166, 0.01, False) + x_166 = None + x_168 = torch.conv2d( + x_167, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_167 = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_169 = torch.nn.functional.batch_norm( + x_168, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_168 = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_170 = torch.nn.functional.leaky_relu(x_169, 0.01, True) + x_169 = None + xb_7 = x_170.contiguous() + x_170 = None + cat_3 = torch.cat([xs_3, xb_7], dim=1) + xs_3 = xb_7 = None + x_171 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_172 = torch.nn.functional.batch_norm( + x_171, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_171 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_173 = torch.nn.functional.leaky_relu(x_172, 0.01, True) + x_172 = None + x_174 = torch.nn.functional.adaptive_avg_pool2d(x_173, 1) + x_173 = None + x_175 = x_174.flatten(1, -1) + x_174 = None + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_176 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_177,) diff --git a/samples/timm/cspresnet50.ra_in1k/weight_meta.py b/samples/timm/cspresnet50.ra_in1k/weight_meta.py new file mode 100644 index 0000000000..48413f25f5 --- /dev/null +++ b/samples/timm/cspresnet50.ra_in1k/weight_meta.py @@ -0,0 +1,2810 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.025 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.229 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.005 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.179 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [512, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [1024, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [2048, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.031 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_" + shape = [1024, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_" + shape = [1024, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [1024, 2048, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/cspresnext50.ra_in1k/graph_hash.txt b/samples/timm/cspresnext50.ra_in1k/graph_hash.txt new file mode 100644 index 0000000000..99b0ca8e43 --- /dev/null +++ b/samples/timm/cspresnext50.ra_in1k/graph_hash.txt @@ -0,0 +1 @@ +828c24b6529b054ebd739d9b467e9b368a14d6bd6582a8549964e3483aee93df \ No newline at end of file diff --git a/samples/timm/cspresnext50.ra_in1k/graph_net.json b/samples/timm/cspresnext50.ra_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/cspresnext50.ra_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/cspresnext50.ra_in1k/input_meta.py b/samples/timm/cspresnext50.ra_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cspresnext50.ra_in1k/input_tensor_constraints.py b/samples/timm/cspresnext50.ra_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/cspresnext50.ra_in1k/model.py b/samples/timm/cspresnext50.ra_in1k/model.py new file mode 100644 index 0000000000..6ca320a889 --- /dev/null +++ b/samples/timm/cspresnext50.ra_in1k/model.py @@ -0,0 +1,1964 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ = ( + L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (2, 2), + (3, 3), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.leaky_relu(x_1, 0.01, True) + x_1 = None + input_1 = torch.nn.functional.max_pool2d( + x_2, 3, 2, 1, 1, ceil_mode=False, return_indices=False + ) + x_2 = None + x_3 = torch.conv2d( + input_1, + l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_1 = l_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + split = x_4.split(128, dim=1) + x_4 = None + xs = split[0] + xb = split[1] + split = None + x_5 = torch.conv2d( + xb, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_6 = torch.nn.functional.batch_norm( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_5 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_7 = torch.nn.functional.leaky_relu(x_6, 0.01, True) + x_6 = None + x_8 = torch.conv2d( + x_7, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_7 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_9 = torch.nn.functional.batch_norm( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_10 = torch.nn.functional.leaky_relu(x_9, 0.01, True) + x_9 = None + x_11 = torch.conv2d( + x_10, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_10 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = (None) + x_12 = torch.nn.functional.batch_norm( + x_11, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_11 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = (None) + x_13 = x_12 + xb + x_12 = xb = None + x_14 = torch.nn.functional.leaky_relu(x_13, 0.01, False) + x_13 = None + x_15 = torch.conv2d( + x_14, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_16 = torch.nn.functional.batch_norm( + x_15, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_15 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_17 = torch.nn.functional.leaky_relu(x_16, 0.01, True) + x_16 = None + x_18 = torch.conv2d( + x_17, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_17 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_19 = torch.nn.functional.batch_norm( + x_18, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_18 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_20 = torch.nn.functional.leaky_relu(x_19, 0.01, True) + x_19 = None + x_21 = torch.conv2d( + x_20, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_20 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = (None) + x_22 = torch.nn.functional.batch_norm( + x_21, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_21 = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = (None) + x_23 = x_22 + x_14 + x_22 = x_14 = None + x_24 = torch.nn.functional.leaky_relu(x_23, 0.01, False) + x_23 = None + x_25 = torch.conv2d( + x_24, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_26 = torch.nn.functional.batch_norm( + x_25, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_25 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_27 = torch.nn.functional.leaky_relu(x_26, 0.01, True) + x_26 = None + x_28 = torch.conv2d( + x_27, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_27 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_29 = torch.nn.functional.batch_norm( + x_28, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_28 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_30 = torch.nn.functional.leaky_relu(x_29, 0.01, True) + x_29 = None + x_31 = torch.conv2d( + x_30, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_30 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ = (None) + x_32 = torch.nn.functional.batch_norm( + x_31, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_31 = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ = (None) + x_33 = x_32 + x_24 + x_32 = x_24 = None + x_34 = torch.nn.functional.leaky_relu(x_33, 0.01, False) + x_33 = None + x_35 = torch.conv2d( + x_34, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_36 = torch.nn.functional.batch_norm( + x_35, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_35 = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_37 = torch.nn.functional.leaky_relu(x_36, 0.01, True) + x_36 = None + xb_1 = x_37.contiguous() + x_37 = None + cat = torch.cat([xs, xb_1], dim=1) + xs = xb_1 = None + x_38 = torch.conv2d( + cat, + l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_39 = torch.nn.functional.batch_norm( + x_38, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_38 = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_40 = torch.nn.functional.leaky_relu(x_39, 0.01, True) + x_39 = None + x_41 = torch.conv2d( + x_40, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 32, + ) + x_40 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_42 = torch.nn.functional.batch_norm( + x_41, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_41 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_43 = torch.nn.functional.leaky_relu(x_42, 0.01, True) + x_42 = None + x_44 = torch.conv2d( + x_43, + l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_43 = l_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_45 = torch.nn.functional.batch_norm( + x_44, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_44 = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + split_1 = x_45.split(256, dim=1) + x_45 = None + xs_1 = split_1[0] + xb_2 = split_1[1] + split_1 = None + x_46 = torch.conv2d( + xb_2, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_47 = torch.nn.functional.batch_norm( + x_46, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_46 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_48 = torch.nn.functional.leaky_relu(x_47, 0.01, True) + x_47 = None + x_49 = torch.conv2d( + x_48, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_48 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_50 = torch.nn.functional.batch_norm( + x_49, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_49 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_51 = torch.nn.functional.leaky_relu(x_50, 0.01, True) + x_50 = None + x_52 = torch.conv2d( + x_51, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_51 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = (None) + x_53 = torch.nn.functional.batch_norm( + x_52, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_52 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = (None) + x_54 = x_53 + xb_2 + x_53 = xb_2 = None + x_55 = torch.nn.functional.leaky_relu(x_54, 0.01, False) + x_54 = None + x_56 = torch.conv2d( + x_55, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_57 = torch.nn.functional.batch_norm( + x_56, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_56 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_58 = torch.nn.functional.leaky_relu(x_57, 0.01, True) + x_57 = None + x_59 = torch.conv2d( + x_58, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_58 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_60 = torch.nn.functional.batch_norm( + x_59, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_59 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_61 = torch.nn.functional.leaky_relu(x_60, 0.01, True) + x_60 = None + x_62 = torch.conv2d( + x_61, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_61 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = (None) + x_63 = torch.nn.functional.batch_norm( + x_62, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_62 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = (None) + x_64 = x_63 + x_55 + x_63 = x_55 = None + x_65 = torch.nn.functional.leaky_relu(x_64, 0.01, False) + x_64 = None + x_66 = torch.conv2d( + x_65, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_67 = torch.nn.functional.batch_norm( + x_66, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_66 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_68 = torch.nn.functional.leaky_relu(x_67, 0.01, True) + x_67 = None + x_69 = torch.conv2d( + x_68, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_68 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_70 = torch.nn.functional.batch_norm( + x_69, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_69 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_71 = torch.nn.functional.leaky_relu(x_70, 0.01, True) + x_70 = None + x_72 = torch.conv2d( + x_71, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_71 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ = (None) + x_73 = torch.nn.functional.batch_norm( + x_72, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_72 = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ = (None) + x_74 = x_73 + x_65 + x_73 = x_65 = None + x_75 = torch.nn.functional.leaky_relu(x_74, 0.01, False) + x_74 = None + x_76 = torch.conv2d( + x_75, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_75 = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_77 = torch.nn.functional.batch_norm( + x_76, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_76 = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_78 = torch.nn.functional.leaky_relu(x_77, 0.01, True) + x_77 = None + xb_3 = x_78.contiguous() + x_78 = None + cat_1 = torch.cat([xs_1, xb_3], dim=1) + xs_1 = xb_3 = None + x_79 = torch.conv2d( + cat_1, + l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_80 = torch.nn.functional.batch_norm( + x_79, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_79 = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_81 = torch.nn.functional.leaky_relu(x_80, 0.01, True) + x_80 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 32, + ) + x_81 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_83 = torch.nn.functional.batch_norm( + x_82, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_82 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_84 = torch.nn.functional.leaky_relu(x_83, 0.01, True) + x_83 = None + x_85 = torch.conv2d( + x_84, + l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_84 = l_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_86 = torch.nn.functional.batch_norm( + x_85, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_85 = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + split_2 = x_86.split(512, dim=1) + x_86 = None + xs_2 = split_2[0] + xb_4 = split_2[1] + split_2 = None + x_87 = torch.conv2d( + xb_4, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_88 = torch.nn.functional.batch_norm( + x_87, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_87 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_89 = torch.nn.functional.leaky_relu(x_88, 0.01, True) + x_88 = None + x_90 = torch.conv2d( + x_89, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_89 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_91 = torch.nn.functional.batch_norm( + x_90, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_90 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_92 = torch.nn.functional.leaky_relu(x_91, 0.01, True) + x_91 = None + x_93 = torch.conv2d( + x_92, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_92 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = (None) + x_94 = torch.nn.functional.batch_norm( + x_93, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_93 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = (None) + x_95 = x_94 + xb_4 + x_94 = xb_4 = None + x_96 = torch.nn.functional.leaky_relu(x_95, 0.01, False) + x_95 = None + x_97 = torch.conv2d( + x_96, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_98 = torch.nn.functional.batch_norm( + x_97, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_97 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_99 = torch.nn.functional.leaky_relu(x_98, 0.01, True) + x_98 = None + x_100 = torch.conv2d( + x_99, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_99 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_101 = torch.nn.functional.batch_norm( + x_100, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_100 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_102 = torch.nn.functional.leaky_relu(x_101, 0.01, True) + x_101 = None + x_103 = torch.conv2d( + x_102, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_102 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = (None) + x_104 = torch.nn.functional.batch_norm( + x_103, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_103 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = (None) + x_105 = x_104 + x_96 + x_104 = x_96 = None + x_106 = torch.nn.functional.leaky_relu(x_105, 0.01, False) + x_105 = None + x_107 = torch.conv2d( + x_106, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_108 = torch.nn.functional.batch_norm( + x_107, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_107 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_109 = torch.nn.functional.leaky_relu(x_108, 0.01, True) + x_108 = None + x_110 = torch.conv2d( + x_109, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_109 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_111 = torch.nn.functional.batch_norm( + x_110, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_110 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_112 = torch.nn.functional.leaky_relu(x_111, 0.01, True) + x_111 = None + x_113 = torch.conv2d( + x_112, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_112 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_ = (None) + x_114 = torch.nn.functional.batch_norm( + x_113, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_113 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_ = (None) + x_115 = x_114 + x_106 + x_114 = x_106 = None + x_116 = torch.nn.functional.leaky_relu(x_115, 0.01, False) + x_115 = None + x_117 = torch.conv2d( + x_116, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_118 = torch.nn.functional.batch_norm( + x_117, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_117 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_119 = torch.nn.functional.leaky_relu(x_118, 0.01, True) + x_118 = None + x_120 = torch.conv2d( + x_119, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_119 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_121 = torch.nn.functional.batch_norm( + x_120, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_120 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_122 = torch.nn.functional.leaky_relu(x_121, 0.01, True) + x_121 = None + x_123 = torch.conv2d( + x_122, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_122 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_ = (None) + x_124 = torch.nn.functional.batch_norm( + x_123, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_123 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_ = (None) + x_125 = x_124 + x_116 + x_124 = x_116 = None + x_126 = torch.nn.functional.leaky_relu(x_125, 0.01, False) + x_125 = None + x_127 = torch.conv2d( + x_126, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_128 = torch.nn.functional.batch_norm( + x_127, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_127 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_129 = torch.nn.functional.leaky_relu(x_128, 0.01, True) + x_128 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_129 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_131 = torch.nn.functional.batch_norm( + x_130, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_130 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_132 = torch.nn.functional.leaky_relu(x_131, 0.01, True) + x_131 = None + x_133 = torch.conv2d( + x_132, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_132 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_ = (None) + x_134 = torch.nn.functional.batch_norm( + x_133, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_133 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_ = (None) + x_135 = x_134 + x_126 + x_134 = x_126 = None + x_136 = torch.nn.functional.leaky_relu(x_135, 0.01, False) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_138 = torch.nn.functional.batch_norm( + x_137, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_137 = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_139 = torch.nn.functional.leaky_relu(x_138, 0.01, True) + x_138 = None + xb_5 = x_139.contiguous() + x_139 = None + cat_2 = torch.cat([xs_2, xb_5], dim=1) + xs_2 = xb_5 = None + x_140 = torch.conv2d( + cat_2, + l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_141 = torch.nn.functional.batch_norm( + x_140, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_140 = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_142 = torch.nn.functional.leaky_relu(x_141, 0.01, True) + x_141 = None + x_143 = torch.conv2d( + x_142, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 32, + ) + x_142 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_144 = torch.nn.functional.batch_norm( + x_143, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_143 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_145 = torch.nn.functional.leaky_relu(x_144, 0.01, True) + x_144 = None + x_146 = torch.conv2d( + x_145, + l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_145 = l_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_ = (None) + x_147 = torch.nn.functional.batch_norm( + x_146, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_146 = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_ = ( + l_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_ + ) = None + split_3 = x_147.split(1024, dim=1) + x_147 = None + xs_3 = split_3[0] + xb_6 = split_3[1] + split_3 = None + x_148 = torch.conv2d( + xb_6, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_149 = torch.nn.functional.batch_norm( + x_148, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_148 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_150 = torch.nn.functional.leaky_relu(x_149, 0.01, True) + x_149 = None + x_151 = torch.conv2d( + x_150, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_150 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_152 = torch.nn.functional.batch_norm( + x_151, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_151 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_153 = torch.nn.functional.leaky_relu(x_152, 0.01, True) + x_152 = None + x_154 = torch.conv2d( + x_153, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_153 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_ = (None) + x_155 = torch.nn.functional.batch_norm( + x_154, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_154 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_ = (None) + x_156 = x_155 + xb_6 + x_155 = xb_6 = None + x_157 = torch.nn.functional.leaky_relu(x_156, 0.01, False) + x_156 = None + x_158 = torch.conv2d( + x_157, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_159 = torch.nn.functional.batch_norm( + x_158, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_158 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_160 = torch.nn.functional.leaky_relu(x_159, 0.01, True) + x_159 = None + x_161 = torch.conv2d( + x_160, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 32, + ) + x_160 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_162 = torch.nn.functional.batch_norm( + x_161, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_161 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_163 = torch.nn.functional.leaky_relu(x_162, 0.01, True) + x_162 = None + x_164 = torch.conv2d( + x_163, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_163 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_ = (None) + x_165 = torch.nn.functional.batch_norm( + x_164, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_164 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_ = (None) + x_166 = x_165 + x_157 + x_165 = x_157 = None + x_167 = torch.nn.functional.leaky_relu(x_166, 0.01, False) + x_166 = None + x_168 = torch.conv2d( + x_167, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_167 = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_ = (None) + x_169 = torch.nn.functional.batch_norm( + x_168, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_168 = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_ = (None) + x_170 = torch.nn.functional.leaky_relu(x_169, 0.01, True) + x_169 = None + xb_7 = x_170.contiguous() + x_170 = None + cat_3 = torch.cat([xs_3, xb_7], dim=1) + xs_3 = xb_7 = None + x_171 = torch.conv2d( + cat_3, + l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_ = (None) + x_172 = torch.nn.functional.batch_norm( + x_171, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_171 = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_ = (None) + x_173 = torch.nn.functional.leaky_relu(x_172, 0.01, True) + x_172 = None + x_174 = torch.nn.functional.adaptive_avg_pool2d(x_173, 1) + x_173 = None + x_175 = x_174.flatten(1, -1) + x_174 = None + x_176 = torch.nn.functional.dropout(x_175, 0.0, False, False) + x_175 = None + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_176 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_177,) diff --git a/samples/timm/cspresnext50.ra_in1k/weight_meta.py b/samples/timm/cspresnext50.ra_in1k/weight_meta.py new file mode 100644 index 0000000000..ffe972c49d --- /dev/null +++ b/samples/timm/cspresnext50.ra_in1k/weight_meta.py @@ -0,0 +1,2810 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.025 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.222 + std = 1.289 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_conv_parameters_weight_" + shape = [256, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.087 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 4, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 4, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.124 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 4, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_transition_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [256, 8, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_conv_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 8, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 8, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 8, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [256, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_transition_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [512, 16, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_conv_parameters_weight_" + shape = [1024, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 16, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 16, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 16, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv3_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 16, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv3_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 16, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv3_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [512, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_conv_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_transition_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [1024, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_conv_parameters_weight_" + shape = [2048, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.031 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_mean_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_buffers_running_var_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_weight_" + ) + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_exp_modules_bn_parameters_bias_" + ) + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [1024, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_conv_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv3_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [1024, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_conv_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv3_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_conv_parameters_weight_" + shape = [1024, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_b_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_conv_parameters_weight_" + shape = [2048, 2048, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.031 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_mean_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_buffers_running_var_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_weight_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_transition_modules_bn_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/darknet53.c2ns_in1k/graph_hash.txt b/samples/timm/darknet53.c2ns_in1k/graph_hash.txt new file mode 100644 index 0000000000..9cdb76c82e --- /dev/null +++ b/samples/timm/darknet53.c2ns_in1k/graph_hash.txt @@ -0,0 +1 @@ +1c50cabf9a25ff8ea22ef875728acf4e6412d099b4dfaf536614bf27e8a11e73 \ No newline at end of file diff --git a/samples/timm/darknet53.c2ns_in1k/graph_net.json b/samples/timm/darknet53.c2ns_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/darknet53.c2ns_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/darknet53.c2ns_in1k/input_meta.py b/samples/timm/darknet53.c2ns_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/darknet53.c2ns_in1k/input_tensor_constraints.py b/samples/timm/darknet53.c2ns_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/darknet53.c2ns_in1k/model.py b/samples/timm/darknet53.c2ns_in1k/model.py new file mode 100644 index 0000000000..3245af82e4 --- /dev/null +++ b/samples/timm/darknet53.c2ns_in1k/model.py @@ -0,0 +1,1857 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.leaky_relu(x_1, 0.01, True) + x_1 = None + x_3 = torch.conv2d( + x_2, + l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_5 = torch.nn.functional.leaky_relu(x_4, 0.01, True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_8 = torch.nn.functional.leaky_relu(x_7, 0.01, True) + x_7 = None + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_11 = torch.nn.functional.leaky_relu(x_10, 0.01, True) + x_10 = None + x_12 = x_11 + x_5 + x_11 = x_5 = None + x_13 = torch.conv2d( + x_12, + l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_12 = l_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_14 = torch.nn.functional.batch_norm( + x_13, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_13 = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_15 = torch.nn.functional.leaky_relu(x_14, 0.01, True) + x_14 = None + x_16 = torch.conv2d( + x_15, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_17 = torch.nn.functional.batch_norm( + x_16, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_16 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_18 = torch.nn.functional.leaky_relu(x_17, 0.01, True) + x_17 = None + x_19 = torch.conv2d( + x_18, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_18 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_20 = torch.nn.functional.batch_norm( + x_19, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_19 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_21 = torch.nn.functional.leaky_relu(x_20, 0.01, True) + x_20 = None + x_22 = x_21 + x_15 + x_21 = x_15 = None + x_23 = torch.conv2d( + x_22, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_24 = torch.nn.functional.batch_norm( + x_23, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_23 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_25 = torch.nn.functional.leaky_relu(x_24, 0.01, True) + x_24 = None + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_25 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_27 = torch.nn.functional.batch_norm( + x_26, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_26 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_28 = torch.nn.functional.leaky_relu(x_27, 0.01, True) + x_27 = None + x_29 = x_28 + x_22 + x_28 = x_22 = None + x_30 = torch.conv2d( + x_29, + l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_29 = l_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_31 = torch.nn.functional.batch_norm( + x_30, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_30 = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_32 = torch.nn.functional.leaky_relu(x_31, 0.01, True) + x_31 = None + x_33 = torch.conv2d( + x_32, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_34 = torch.nn.functional.batch_norm( + x_33, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_33 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_35 = torch.nn.functional.leaky_relu(x_34, 0.01, True) + x_34 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_37 = torch.nn.functional.batch_norm( + x_36, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_36 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_38 = torch.nn.functional.leaky_relu(x_37, 0.01, True) + x_37 = None + x_39 = x_38 + x_32 + x_38 = x_32 = None + x_40 = torch.conv2d( + x_39, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_41 = torch.nn.functional.batch_norm( + x_40, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_40 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_42 = torch.nn.functional.leaky_relu(x_41, 0.01, True) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_44 = torch.nn.functional.batch_norm( + x_43, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_43 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_45 = torch.nn.functional.leaky_relu(x_44, 0.01, True) + x_44 = None + x_46 = x_45 + x_39 + x_45 = x_39 = None + x_47 = torch.conv2d( + x_46, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_48 = torch.nn.functional.batch_norm( + x_47, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_47 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_49 = torch.nn.functional.leaky_relu(x_48, 0.01, True) + x_48 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_49 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_51 = torch.nn.functional.batch_norm( + x_50, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_50 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_52 = torch.nn.functional.leaky_relu(x_51, 0.01, True) + x_51 = None + x_53 = x_52 + x_46 + x_52 = x_46 = None + x_54 = torch.conv2d( + x_53, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_55 = torch.nn.functional.batch_norm( + x_54, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_56 = torch.nn.functional.leaky_relu(x_55, 0.01, True) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_58 = torch.nn.functional.batch_norm( + x_57, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_57 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_59 = torch.nn.functional.leaky_relu(x_58, 0.01, True) + x_58 = None + x_60 = x_59 + x_53 + x_59 = x_53 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_62 = torch.nn.functional.batch_norm( + x_61, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_61 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_63 = torch.nn.functional.leaky_relu(x_62, 0.01, True) + x_62 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_63 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_65 = torch.nn.functional.batch_norm( + x_64, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_64 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_66 = torch.nn.functional.leaky_relu(x_65, 0.01, True) + x_65 = None + x_67 = x_66 + x_60 + x_66 = x_60 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_69 = torch.nn.functional.batch_norm( + x_68, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_68 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_70 = torch.nn.functional.leaky_relu(x_69, 0.01, True) + x_69 = None + x_71 = torch.conv2d( + x_70, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_70 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_72 = torch.nn.functional.batch_norm( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_73 = torch.nn.functional.leaky_relu(x_72, 0.01, True) + x_72 = None + x_74 = x_73 + x_67 + x_73 = x_67 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_76 = torch.nn.functional.batch_norm( + x_75, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_75 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_77 = torch.nn.functional.leaky_relu(x_76, 0.01, True) + x_76 = None + x_78 = torch.conv2d( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_79 = torch.nn.functional.batch_norm( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_80 = torch.nn.functional.leaky_relu(x_79, 0.01, True) + x_79 = None + x_81 = x_80 + x_74 + x_80 = x_74 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_83 = torch.nn.functional.batch_norm( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_82 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_84 = torch.nn.functional.leaky_relu(x_83, 0.01, True) + x_83 = None + x_85 = torch.conv2d( + x_84, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_84 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_86 = torch.nn.functional.batch_norm( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_85 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_87 = torch.nn.functional.leaky_relu(x_86, 0.01, True) + x_86 = None + x_88 = x_87 + x_81 + x_87 = x_81 = None + x_89 = torch.conv2d( + x_88, + l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_88 = l_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_90 = torch.nn.functional.batch_norm( + x_89, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_89 = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_91 = torch.nn.functional.leaky_relu(x_90, 0.01, True) + x_90 = None + x_92 = torch.conv2d( + x_91, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_93 = torch.nn.functional.batch_norm( + x_92, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_92 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_94 = torch.nn.functional.leaky_relu(x_93, 0.01, True) + x_93 = None + x_95 = torch.conv2d( + x_94, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_94 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_96 = torch.nn.functional.batch_norm( + x_95, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_95 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_97 = torch.nn.functional.leaky_relu(x_96, 0.01, True) + x_96 = None + x_98 = x_97 + x_91 + x_97 = x_91 = None + x_99 = torch.conv2d( + x_98, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_100 = torch.nn.functional.batch_norm( + x_99, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_99 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_101 = torch.nn.functional.leaky_relu(x_100, 0.01, True) + x_100 = None + x_102 = torch.conv2d( + x_101, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_101 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_103 = torch.nn.functional.batch_norm( + x_102, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_102 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_104 = torch.nn.functional.leaky_relu(x_103, 0.01, True) + x_103 = None + x_105 = x_104 + x_98 + x_104 = x_98 = None + x_106 = torch.conv2d( + x_105, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_107 = torch.nn.functional.batch_norm( + x_106, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_106 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_108 = torch.nn.functional.leaky_relu(x_107, 0.01, True) + x_107 = None + x_109 = torch.conv2d( + x_108, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_108 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_110 = torch.nn.functional.batch_norm( + x_109, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_109 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_111 = torch.nn.functional.leaky_relu(x_110, 0.01, True) + x_110 = None + x_112 = x_111 + x_105 + x_111 = x_105 = None + x_113 = torch.conv2d( + x_112, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_114 = torch.nn.functional.batch_norm( + x_113, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_113 = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_115 = torch.nn.functional.leaky_relu(x_114, 0.01, True) + x_114 = None + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_117 = torch.nn.functional.batch_norm( + x_116, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_116 = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_118 = torch.nn.functional.leaky_relu(x_117, 0.01, True) + x_117 = None + x_119 = x_118 + x_112 + x_118 = x_112 = None + x_120 = torch.conv2d( + x_119, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_121 = torch.nn.functional.batch_norm( + x_120, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_120 = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_122 = torch.nn.functional.leaky_relu(x_121, 0.01, True) + x_121 = None + x_123 = torch.conv2d( + x_122, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_122 = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_124 = torch.nn.functional.batch_norm( + x_123, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_125 = torch.nn.functional.leaky_relu(x_124, 0.01, True) + x_124 = None + x_126 = x_125 + x_119 + x_125 = x_119 = None + x_127 = torch.conv2d( + x_126, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_128 = torch.nn.functional.batch_norm( + x_127, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_127 = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_129 = torch.nn.functional.leaky_relu(x_128, 0.01, True) + x_128 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_131 = torch.nn.functional.batch_norm( + x_130, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_130 = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_132 = torch.nn.functional.leaky_relu(x_131, 0.01, True) + x_131 = None + x_133 = x_132 + x_126 + x_132 = x_126 = None + x_134 = torch.conv2d( + x_133, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_135 = torch.nn.functional.batch_norm( + x_134, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_134 = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_136 = torch.nn.functional.leaky_relu(x_135, 0.01, True) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_138 = torch.nn.functional.batch_norm( + x_137, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_137 = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_139 = torch.nn.functional.leaky_relu(x_138, 0.01, True) + x_138 = None + x_140 = x_139 + x_133 + x_139 = x_133 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_142 = torch.nn.functional.batch_norm( + x_141, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_141 = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_143 = torch.nn.functional.leaky_relu(x_142, 0.01, True) + x_142 = None + x_144 = torch.conv2d( + x_143, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_143 = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_145 = torch.nn.functional.batch_norm( + x_144, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_144 = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_146 = torch.nn.functional.leaky_relu(x_145, 0.01, True) + x_145 = None + x_147 = x_146 + x_140 + x_146 = x_140 = None + x_148 = torch.conv2d( + x_147, + l_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_147 = l_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_ = (None) + x_149 = torch.nn.functional.batch_norm( + x_148, + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_148 = l_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_ = (None) + x_150 = torch.nn.functional.leaky_relu(x_149, 0.01, True) + x_149 = None + x_151 = torch.conv2d( + x_150, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_152 = torch.nn.functional.batch_norm( + x_151, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_151 = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_153 = torch.nn.functional.leaky_relu(x_152, 0.01, True) + x_152 = None + x_154 = torch.conv2d( + x_153, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_153 = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_155 = torch.nn.functional.batch_norm( + x_154, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_154 = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_156 = torch.nn.functional.leaky_relu(x_155, 0.01, True) + x_155 = None + x_157 = x_156 + x_150 + x_156 = x_150 = None + x_158 = torch.conv2d( + x_157, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_159 = torch.nn.functional.batch_norm( + x_158, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_158 = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_160 = torch.nn.functional.leaky_relu(x_159, 0.01, True) + x_159 = None + x_161 = torch.conv2d( + x_160, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_160 = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_162 = torch.nn.functional.batch_norm( + x_161, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_161 = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_163 = torch.nn.functional.leaky_relu(x_162, 0.01, True) + x_162 = None + x_164 = x_163 + x_157 + x_163 = x_157 = None + x_165 = torch.conv2d( + x_164, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_166 = torch.nn.functional.batch_norm( + x_165, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_165 = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_167 = torch.nn.functional.leaky_relu(x_166, 0.01, True) + x_166 = None + x_168 = torch.conv2d( + x_167, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_167 = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_169 = torch.nn.functional.batch_norm( + x_168, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_168 = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_170 = torch.nn.functional.leaky_relu(x_169, 0.01, True) + x_169 = None + x_171 = x_170 + x_164 + x_170 = x_164 = None + x_172 = torch.conv2d( + x_171, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_173 = torch.nn.functional.batch_norm( + x_172, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_172 = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_174 = torch.nn.functional.leaky_relu(x_173, 0.01, True) + x_173 = None + x_175 = torch.conv2d( + x_174, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_174 = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_176 = torch.nn.functional.batch_norm( + x_175, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_175 = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_177 = torch.nn.functional.leaky_relu(x_176, 0.01, True) + x_176 = None + x_178 = x_177 + x_171 + x_177 = x_171 = None + x_179 = torch.nn.functional.adaptive_avg_pool2d(x_178, 1) + x_178 = None + x_180 = x_179.flatten(1, -1) + x_179 = None + x_181 = torch.nn.functional.dropout(x_180, 0.0, False, False) + x_180 = None + x_182 = torch._C._nn.linear( + x_181, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_181 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_182,) diff --git a/samples/timm/darknet53.c2ns_in1k/weight_meta.py b/samples/timm/darknet53.c2ns_in1k/weight_meta.py new file mode 100644 index 0000000000..10302cd699 --- /dev/null +++ b/samples/timm/darknet53.c2ns_in1k/weight_meta.py @@ -0,0 +1,2648 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [32, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.004 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.224 + std = 1.283 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_conv_parameters_weight_" + shape = [64, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_0_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [32, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.254 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_conv_parameters_weight_" + shape = [128, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_1_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.178 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_2_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.124 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_3_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_conv_parameters_weight_" + shape = [1024, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_: + name = ( + "L_self_modules_stages_modules_4_modules_conv_down_modules_bn_parameters_bias_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [1024, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [1024, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [1024, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [1024, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/darknetaa53.c2ns_in1k/graph_hash.txt b/samples/timm/darknetaa53.c2ns_in1k/graph_hash.txt new file mode 100644 index 0000000000..6449c61d97 --- /dev/null +++ b/samples/timm/darknetaa53.c2ns_in1k/graph_hash.txt @@ -0,0 +1 @@ +c2ae90f0c65938bc0418ed7fc311e7f62365ebea543cf67c922c46d0878fab01 \ No newline at end of file diff --git a/samples/timm/darknetaa53.c2ns_in1k/graph_net.json b/samples/timm/darknetaa53.c2ns_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/darknetaa53.c2ns_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/darknetaa53.c2ns_in1k/input_meta.py b/samples/timm/darknetaa53.c2ns_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/darknetaa53.c2ns_in1k/input_tensor_constraints.py b/samples/timm/darknetaa53.c2ns_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/darknetaa53.c2ns_in1k/model.py b/samples/timm/darknetaa53.c2ns_in1k/model.py new file mode 100644 index 0000000000..67732f4a67 --- /dev/null +++ b/samples/timm/darknetaa53.c2ns_in1k/model.py @@ -0,0 +1,1867 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ = ( + L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = ( + L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ + ) + l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ + l_self_modules_head_modules_fc_parameters_weight_ = ( + L_self_modules_head_modules_fc_parameters_weight_ + ) + l_self_modules_head_modules_fc_parameters_bias_ = ( + L_self_modules_head_modules_fc_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_x_ = l_self_modules_stem_modules_conv1_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_stem_modules_conv1_modules_bn_parameters_weight_ + ) = l_self_modules_stem_modules_conv1_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.leaky_relu(x_1, 0.01, True) + x_1 = None + input_1 = torch._C._nn.avg_pool2d(x_2, 2, 2, 0, False, True, None) + x_2 = None + x_3 = torch.conv2d( + input_1, + l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_1 = l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_conv_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_bias_ = (None) + x_5 = torch.nn.functional.leaky_relu(x_4, 0.01, True) + x_4 = None + x_6 = torch.conv2d( + x_5, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_8 = torch.nn.functional.leaky_relu(x_7, 0.01, True) + x_7 = None + x_9 = torch.conv2d( + x_8, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_8 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_11 = torch.nn.functional.leaky_relu(x_10, 0.01, True) + x_10 = None + x_12 = x_11 + x_5 + x_11 = x_5 = None + input_2 = torch._C._nn.avg_pool2d(x_12, 2, 2, 0, False, True, None) + x_12 = None + x_13 = torch.conv2d( + input_2, + l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_2 = l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_conv_parameters_weight_ = (None) + x_14 = torch.nn.functional.batch_norm( + x_13, + l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_13 = l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_bias_ = (None) + x_15 = torch.nn.functional.leaky_relu(x_14, 0.01, True) + x_14 = None + x_16 = torch.conv2d( + x_15, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_17 = torch.nn.functional.batch_norm( + x_16, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_16 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_18 = torch.nn.functional.leaky_relu(x_17, 0.01, True) + x_17 = None + x_19 = torch.conv2d( + x_18, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_18 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_20 = torch.nn.functional.batch_norm( + x_19, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_19 = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_21 = torch.nn.functional.leaky_relu(x_20, 0.01, True) + x_20 = None + x_22 = x_21 + x_15 + x_21 = x_15 = None + x_23 = torch.conv2d( + x_22, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_24 = torch.nn.functional.batch_norm( + x_23, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_23 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_25 = torch.nn.functional.leaky_relu(x_24, 0.01, True) + x_24 = None + x_26 = torch.conv2d( + x_25, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_25 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_27 = torch.nn.functional.batch_norm( + x_26, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_26 = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_28 = torch.nn.functional.leaky_relu(x_27, 0.01, True) + x_27 = None + x_29 = x_28 + x_22 + x_28 = x_22 = None + input_3 = torch._C._nn.avg_pool2d(x_29, 2, 2, 0, False, True, None) + x_29 = None + x_30 = torch.conv2d( + input_3, + l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_3 = l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_conv_parameters_weight_ = (None) + x_31 = torch.nn.functional.batch_norm( + x_30, + l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_30 = l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_bias_ = (None) + x_32 = torch.nn.functional.leaky_relu(x_31, 0.01, True) + x_31 = None + x_33 = torch.conv2d( + x_32, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_34 = torch.nn.functional.batch_norm( + x_33, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_33 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_35 = torch.nn.functional.leaky_relu(x_34, 0.01, True) + x_34 = None + x_36 = torch.conv2d( + x_35, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_35 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_37 = torch.nn.functional.batch_norm( + x_36, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_36 = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_38 = torch.nn.functional.leaky_relu(x_37, 0.01, True) + x_37 = None + x_39 = x_38 + x_32 + x_38 = x_32 = None + x_40 = torch.conv2d( + x_39, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_41 = torch.nn.functional.batch_norm( + x_40, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_40 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_42 = torch.nn.functional.leaky_relu(x_41, 0.01, True) + x_41 = None + x_43 = torch.conv2d( + x_42, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_42 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_44 = torch.nn.functional.batch_norm( + x_43, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_43 = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_45 = torch.nn.functional.leaky_relu(x_44, 0.01, True) + x_44 = None + x_46 = x_45 + x_39 + x_45 = x_39 = None + x_47 = torch.conv2d( + x_46, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_48 = torch.nn.functional.batch_norm( + x_47, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_47 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_49 = torch.nn.functional.leaky_relu(x_48, 0.01, True) + x_48 = None + x_50 = torch.conv2d( + x_49, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_49 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_51 = torch.nn.functional.batch_norm( + x_50, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_50 = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_52 = torch.nn.functional.leaky_relu(x_51, 0.01, True) + x_51 = None + x_53 = x_52 + x_46 + x_52 = x_46 = None + x_54 = torch.conv2d( + x_53, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_55 = torch.nn.functional.batch_norm( + x_54, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_54 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_56 = torch.nn.functional.leaky_relu(x_55, 0.01, True) + x_55 = None + x_57 = torch.conv2d( + x_56, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_56 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_58 = torch.nn.functional.batch_norm( + x_57, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_57 = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_59 = torch.nn.functional.leaky_relu(x_58, 0.01, True) + x_58 = None + x_60 = x_59 + x_53 + x_59 = x_53 = None + x_61 = torch.conv2d( + x_60, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_62 = torch.nn.functional.batch_norm( + x_61, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_61 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_63 = torch.nn.functional.leaky_relu(x_62, 0.01, True) + x_62 = None + x_64 = torch.conv2d( + x_63, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_63 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_65 = torch.nn.functional.batch_norm( + x_64, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_64 = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_66 = torch.nn.functional.leaky_relu(x_65, 0.01, True) + x_65 = None + x_67 = x_66 + x_60 + x_66 = x_60 = None + x_68 = torch.conv2d( + x_67, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_69 = torch.nn.functional.batch_norm( + x_68, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_68 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_70 = torch.nn.functional.leaky_relu(x_69, 0.01, True) + x_69 = None + x_71 = torch.conv2d( + x_70, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_70 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_72 = torch.nn.functional.batch_norm( + x_71, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_71 = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_73 = torch.nn.functional.leaky_relu(x_72, 0.01, True) + x_72 = None + x_74 = x_73 + x_67 + x_73 = x_67 = None + x_75 = torch.conv2d( + x_74, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_76 = torch.nn.functional.batch_norm( + x_75, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_75 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_77 = torch.nn.functional.leaky_relu(x_76, 0.01, True) + x_76 = None + x_78 = torch.conv2d( + x_77, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_77 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_79 = torch.nn.functional.batch_norm( + x_78, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_78 = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_80 = torch.nn.functional.leaky_relu(x_79, 0.01, True) + x_79 = None + x_81 = x_80 + x_74 + x_80 = x_74 = None + x_82 = torch.conv2d( + x_81, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_83 = torch.nn.functional.batch_norm( + x_82, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_82 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_84 = torch.nn.functional.leaky_relu(x_83, 0.01, True) + x_83 = None + x_85 = torch.conv2d( + x_84, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_84 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_86 = torch.nn.functional.batch_norm( + x_85, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_85 = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_87 = torch.nn.functional.leaky_relu(x_86, 0.01, True) + x_86 = None + x_88 = x_87 + x_81 + x_87 = x_81 = None + input_4 = torch._C._nn.avg_pool2d(x_88, 2, 2, 0, False, True, None) + x_88 = None + x_89 = torch.conv2d( + input_4, + l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_4 = l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_conv_parameters_weight_ = (None) + x_90 = torch.nn.functional.batch_norm( + x_89, + l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_89 = l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_bias_ = (None) + x_91 = torch.nn.functional.leaky_relu(x_90, 0.01, True) + x_90 = None + x_92 = torch.conv2d( + x_91, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_93 = torch.nn.functional.batch_norm( + x_92, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_92 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_94 = torch.nn.functional.leaky_relu(x_93, 0.01, True) + x_93 = None + x_95 = torch.conv2d( + x_94, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_94 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_96 = torch.nn.functional.batch_norm( + x_95, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_95 = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_97 = torch.nn.functional.leaky_relu(x_96, 0.01, True) + x_96 = None + x_98 = x_97 + x_91 + x_97 = x_91 = None + x_99 = torch.conv2d( + x_98, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_100 = torch.nn.functional.batch_norm( + x_99, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_99 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_101 = torch.nn.functional.leaky_relu(x_100, 0.01, True) + x_100 = None + x_102 = torch.conv2d( + x_101, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_101 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_103 = torch.nn.functional.batch_norm( + x_102, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_102 = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_104 = torch.nn.functional.leaky_relu(x_103, 0.01, True) + x_103 = None + x_105 = x_104 + x_98 + x_104 = x_98 = None + x_106 = torch.conv2d( + x_105, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_107 = torch.nn.functional.batch_norm( + x_106, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_106 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_108 = torch.nn.functional.leaky_relu(x_107, 0.01, True) + x_107 = None + x_109 = torch.conv2d( + x_108, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_108 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_110 = torch.nn.functional.batch_norm( + x_109, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_109 = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_111 = torch.nn.functional.leaky_relu(x_110, 0.01, True) + x_110 = None + x_112 = x_111 + x_105 + x_111 = x_105 = None + x_113 = torch.conv2d( + x_112, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_114 = torch.nn.functional.batch_norm( + x_113, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_113 = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_115 = torch.nn.functional.leaky_relu(x_114, 0.01, True) + x_114 = None + x_116 = torch.conv2d( + x_115, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_115 = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_117 = torch.nn.functional.batch_norm( + x_116, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_116 = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_118 = torch.nn.functional.leaky_relu(x_117, 0.01, True) + x_117 = None + x_119 = x_118 + x_112 + x_118 = x_112 = None + x_120 = torch.conv2d( + x_119, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_121 = torch.nn.functional.batch_norm( + x_120, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_120 = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_ = (None) + x_122 = torch.nn.functional.leaky_relu(x_121, 0.01, True) + x_121 = None + x_123 = torch.conv2d( + x_122, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_122 = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_ = (None) + x_124 = torch.nn.functional.batch_norm( + x_123, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_123 = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_ = (None) + x_125 = torch.nn.functional.leaky_relu(x_124, 0.01, True) + x_124 = None + x_126 = x_125 + x_119 + x_125 = x_119 = None + x_127 = torch.conv2d( + x_126, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_128 = torch.nn.functional.batch_norm( + x_127, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_127 = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_ = (None) + x_129 = torch.nn.functional.leaky_relu(x_128, 0.01, True) + x_128 = None + x_130 = torch.conv2d( + x_129, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_129 = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_ = (None) + x_131 = torch.nn.functional.batch_norm( + x_130, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_130 = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_ = (None) + x_132 = torch.nn.functional.leaky_relu(x_131, 0.01, True) + x_131 = None + x_133 = x_132 + x_126 + x_132 = x_126 = None + x_134 = torch.conv2d( + x_133, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_135 = torch.nn.functional.batch_norm( + x_134, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_134 = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_ = (None) + x_136 = torch.nn.functional.leaky_relu(x_135, 0.01, True) + x_135 = None + x_137 = torch.conv2d( + x_136, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_136 = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_ = (None) + x_138 = torch.nn.functional.batch_norm( + x_137, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_137 = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_ = (None) + x_139 = torch.nn.functional.leaky_relu(x_138, 0.01, True) + x_138 = None + x_140 = x_139 + x_133 + x_139 = x_133 = None + x_141 = torch.conv2d( + x_140, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_142 = torch.nn.functional.batch_norm( + x_141, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_141 = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_ = (None) + x_143 = torch.nn.functional.leaky_relu(x_142, 0.01, True) + x_142 = None + x_144 = torch.conv2d( + x_143, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_143 = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_ = (None) + x_145 = torch.nn.functional.batch_norm( + x_144, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_144 = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_ = (None) + x_146 = torch.nn.functional.leaky_relu(x_145, 0.01, True) + x_145 = None + x_147 = x_146 + x_140 + x_146 = x_140 = None + input_5 = torch._C._nn.avg_pool2d(x_147, 2, 2, 0, False, True, None) + x_147 = None + x_148 = torch.conv2d( + input_5, + l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + input_5 = l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_conv_parameters_weight_ = (None) + x_149 = torch.nn.functional.batch_norm( + x_148, + l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_148 = l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_bias_ = (None) + x_150 = torch.nn.functional.leaky_relu(x_149, 0.01, True) + x_149 = None + x_151 = torch.conv2d( + x_150, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_152 = torch.nn.functional.batch_norm( + x_151, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_151 = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_ = (None) + x_153 = torch.nn.functional.leaky_relu(x_152, 0.01, True) + x_152 = None + x_154 = torch.conv2d( + x_153, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_153 = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_ = (None) + x_155 = torch.nn.functional.batch_norm( + x_154, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_154 = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_ = (None) + x_156 = torch.nn.functional.leaky_relu(x_155, 0.01, True) + x_155 = None + x_157 = x_156 + x_150 + x_156 = x_150 = None + x_158 = torch.conv2d( + x_157, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_159 = torch.nn.functional.batch_norm( + x_158, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_158 = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_ = (None) + x_160 = torch.nn.functional.leaky_relu(x_159, 0.01, True) + x_159 = None + x_161 = torch.conv2d( + x_160, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_160 = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_ = (None) + x_162 = torch.nn.functional.batch_norm( + x_161, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_161 = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_ = (None) + x_163 = torch.nn.functional.leaky_relu(x_162, 0.01, True) + x_162 = None + x_164 = x_163 + x_157 + x_163 = x_157 = None + x_165 = torch.conv2d( + x_164, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_166 = torch.nn.functional.batch_norm( + x_165, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_165 = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_ = (None) + x_167 = torch.nn.functional.leaky_relu(x_166, 0.01, True) + x_166 = None + x_168 = torch.conv2d( + x_167, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_167 = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_ = (None) + x_169 = torch.nn.functional.batch_norm( + x_168, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_168 = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_ = (None) + x_170 = torch.nn.functional.leaky_relu(x_169, 0.01, True) + x_169 = None + x_171 = x_170 + x_164 + x_170 = x_164 = None + x_172 = torch.conv2d( + x_171, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_ = ( + None + ) + x_173 = torch.nn.functional.batch_norm( + x_172, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_172 = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_ = (None) + x_174 = torch.nn.functional.leaky_relu(x_173, 0.01, True) + x_173 = None + x_175 = torch.conv2d( + x_174, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_174 = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_ = (None) + x_176 = torch.nn.functional.batch_norm( + x_175, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_, + l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_175 = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_ = l_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_ = (None) + x_177 = torch.nn.functional.leaky_relu(x_176, 0.01, True) + x_176 = None + x_178 = x_177 + x_171 + x_177 = x_171 = None + x_179 = torch.nn.functional.adaptive_avg_pool2d(x_178, 1) + x_178 = None + x_180 = x_179.flatten(1, -1) + x_179 = None + x_181 = torch.nn.functional.dropout(x_180, 0.0, False, False) + x_180 = None + x_182 = torch._C._nn.linear( + x_181, + l_self_modules_head_modules_fc_parameters_weight_, + l_self_modules_head_modules_fc_parameters_bias_, + ) + x_181 = ( + l_self_modules_head_modules_fc_parameters_weight_ + ) = l_self_modules_head_modules_fc_parameters_bias_ = None + return (x_182,) diff --git a/samples/timm/darknetaa53.c2ns_in1k/weight_meta.py b/samples/timm/darknetaa53.c2ns_in1k/weight_meta.py new file mode 100644 index 0000000000..9ef5d7cca1 --- /dev/null +++ b/samples/timm/darknetaa53.c2ns_in1k/weight_meta.py @@ -0,0 +1,2638 @@ +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_conv_parameters_weight_" + shape = [32, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.078 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.224 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stem_modules_conv1_modules_bn_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stem_modules_conv1_modules_bn_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_conv_parameters_weight_" + shape = [64, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.175 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_conv_down_modules_1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [32, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.008 + std = 0.246 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [64, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.058 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_0_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_conv_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_conv_down_modules_1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [128, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_1_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_conv_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_conv_down_modules_1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.124 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_2_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_conv_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_conv_down_modules_1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_4_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_5_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_6_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv1_modules_bn_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_conv_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_3_modules_blocks_modules_7_modules_conv2_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_conv_parameters_weight_" + shape = [1024, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_conv_down_modules_1_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_conv_parameters_weight_" + shape = [1024, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_0_modules_conv2_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_conv_parameters_weight_" + shape = [1024, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_1_modules_conv2_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_conv_parameters_weight_" + shape = [1024, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_2_modules_conv2_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv1_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_conv_parameters_weight_" + shape = [1024, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.015 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_: + name = "L_self_modules_stages_modules_4_modules_blocks_modules_3_modules_conv2_modules_bn_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_weight_: + name = "L_self_modules_head_modules_fc_parameters_weight_" + shape = [1000, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.010 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_modules_fc_parameters_bias_: + name = "L_self_modules_head_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/deit3_base_patch16_224.fb_in1k/graph_hash.txt b/samples/timm/deit3_base_patch16_224.fb_in1k/graph_hash.txt new file mode 100644 index 0000000000..409c645a12 --- /dev/null +++ b/samples/timm/deit3_base_patch16_224.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +f6582d13c42b77e560d755b7c97489429c32219c0d78a2df64567083e29ff3e6 \ No newline at end of file diff --git a/samples/timm/deit3_base_patch16_224.fb_in1k/graph_net.json b/samples/timm/deit3_base_patch16_224.fb_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/deit3_base_patch16_224.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/deit3_base_patch16_224.fb_in1k/input_meta.py b/samples/timm/deit3_base_patch16_224.fb_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/deit3_base_patch16_224.fb_in1k/input_tensor_constraints.py b/samples/timm/deit3_base_patch16_224.fb_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/deit3_base_patch16_224.fb_in1k/model.py b/samples/timm/deit3_base_patch16_224.fb_in1k/model.py new file mode 100644 index 0000000000..4ecba32b80 --- /dev/null +++ b/samples/timm/deit3_base_patch16_224.fb_in1k/model.py @@ -0,0 +1,1868 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ + ) + l_self_modules_norm_parameters_weight_ = L_self_modules_norm_parameters_weight_ + l_self_modules_norm_parameters_bias_ = L_self_modules_norm_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + x_2 = x_1 + l_self_parameters_pos_embed_ + x_1 = l_self_parameters_pos_embed_ = None + x_3 = torch.cat([expand, x_2], dim=1) + expand = x_2 = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (768,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 197, 3, 12, 64) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 197, 768) + transpose_1 = None + x_8 = torch._C._nn.linear( + x_7, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + mul = x_9 * l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ + x_9 = l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ = None + x_10 = x_4 + mul + x_4 = mul = None + x_11 = torch.nn.functional.layer_norm( + x_10, + (768,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_11 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_13 = torch._C._nn.gelu(x_12, approximate="none") + x_12 = None + x_14 = torch.nn.functional.dropout(x_13, 0.0, False, False) + x_13 = None + x_15 = torch._C._nn.linear( + x_14, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_14 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + mul_1 = x_16 * l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ + x_16 = l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ = None + x_17 = x_10 + mul_1 + x_10 = mul_1 = None + x_18 = torch.nn.functional.layer_norm( + x_17, + (768,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_18, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_18 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 197, 3, 12, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_19 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_19.transpose(1, 2) + x_19 = None + x_20 = transpose_2.reshape(1, 197, 768) + transpose_2 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + mul_2 = x_22 * l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ + x_22 = l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ = None + x_23 = x_17 + mul_2 + x_17 = mul_2 = None + x_24 = torch.nn.functional.layer_norm( + x_23, + (768,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_25 = torch._C._nn.linear( + x_24, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_24 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_26 = torch._C._nn.gelu(x_25, approximate="none") + x_25 = None + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + mul_3 = x_29 * l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ + x_29 = l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ = None + x_30 = x_23 + mul_3 + x_23 = mul_3 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (768,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 197, 3, 12, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_32 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_32.transpose(1, 2) + x_32 = None + x_33 = transpose_3.reshape(1, 197, 768) + transpose_3 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_33 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + mul_4 = x_35 * l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ + x_35 = l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ = None + x_36 = x_30 + mul_4 + x_30 = mul_4 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (768,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_39 = torch._C._nn.gelu(x_38, approximate="none") + x_38 = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = torch._C._nn.linear( + x_40, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_40 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_42 = torch.nn.functional.dropout(x_41, 0.0, False, False) + x_41 = None + mul_5 = x_42 * l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ + x_42 = l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ = None + x_43 = x_36 + mul_5 + x_36 = mul_5 = None + x_44 = torch.nn.functional.layer_norm( + x_43, + (768,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_44 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 197, 3, 12, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_45 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_45.transpose(1, 2) + x_45 = None + x_46 = transpose_4.reshape(1, 197, 768) + transpose_4 = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + mul_6 = x_48 * l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ + x_48 = l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ = None + x_49 = x_43 + mul_6 + x_43 = mul_6 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (768,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_51 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_52 = torch._C._nn.gelu(x_51, approximate="none") + x_51 = None + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + mul_7 = x_55 * l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ + x_55 = l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ = None + x_56 = x_49 + mul_7 + x_49 = mul_7 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (768,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 197, 3, 12, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_5.reshape(1, 197, 768) + transpose_5 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + mul_8 = x_61 * l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ + x_61 = l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ = None + x_62 = x_56 + mul_8 + x_56 = mul_8 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (768,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + mul_9 = x_68 * l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ + x_68 = l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ = None + x_69 = x_62 + mul_9 + x_62 = mul_9 = None + x_70 = torch.nn.functional.layer_norm( + x_69, + (768,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_70 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 197, 3, 12, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_71 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_71.transpose(1, 2) + x_71 = None + x_72 = transpose_6.reshape(1, 197, 768) + transpose_6 = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + mul_10 = x_74 * l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ + x_74 = l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ = None + x_75 = x_69 + mul_10 + x_69 = mul_10 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (768,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_78 = torch._C._nn.gelu(x_77, approximate="none") + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch._C._nn.linear( + x_79, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_79 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + mul_11 = x_81 * l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ + x_81 = l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ = None + x_82 = x_75 + mul_11 + x_75 = mul_11 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (768,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 197, 3, 12, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_84 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_84.transpose(1, 2) + x_84 = None + x_85 = transpose_7.reshape(1, 197, 768) + transpose_7 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_85 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + mul_12 = x_87 * l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ + x_87 = l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ = None + x_88 = x_82 + mul_12 + x_82 = mul_12 = None + x_89 = torch.nn.functional.layer_norm( + x_88, + (768,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_89 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_91 = torch._C._nn.gelu(x_90, approximate="none") + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + mul_13 = x_94 * l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ + x_94 = l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ = None + x_95 = x_88 + mul_13 + x_88 = mul_13 = None + x_96 = torch.nn.functional.layer_norm( + x_95, + (768,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_96 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 197, 3, 12, 64) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_97 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_97.transpose(1, 2) + x_97 = None + x_98 = transpose_8.reshape(1, 197, 768) + transpose_8 = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + mul_14 = x_100 * l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ + x_100 = l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ = None + x_101 = x_95 + mul_14 + x_95 = mul_14 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (768,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch._C._nn.linear( + x_105, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_105 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + mul_15 = x_107 * l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ + x_107 = l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ = None + x_108 = x_101 + mul_15 + x_101 = mul_15 = None + x_109 = torch.nn.functional.layer_norm( + x_108, + (768,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_109, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_109 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 197, 3, 12, 64) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_9.reshape(1, 197, 768) + transpose_9 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + mul_16 = x_113 * l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ + x_113 = l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ = None + x_114 = x_108 + mul_16 + x_108 = mul_16 = None + x_115 = torch.nn.functional.layer_norm( + x_114, + (768,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_116 = torch._C._nn.linear( + x_115, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_115 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + mul_17 = x_120 * l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ + x_120 = l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ = None + x_121 = x_114 + mul_17 + x_114 = mul_17 = None + x_122 = torch.nn.functional.layer_norm( + x_121, + (768,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_122 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 197, 3, 12, 64) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_123 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_123.transpose(1, 2) + x_123 = None + x_124 = transpose_10.reshape(1, 197, 768) + transpose_10 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_124 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + mul_18 = x_126 * l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ + x_126 = l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ = None + x_127 = x_121 + mul_18 + x_121 = mul_18 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (768,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_130 = torch._C._nn.gelu(x_129, approximate="none") + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch._C._nn.linear( + x_131, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_131 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + mul_19 = x_133 * l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ + x_133 = l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ = None + x_134 = x_127 + mul_19 + x_127 = mul_19 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (768,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_135, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_135 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 197, 3, 12, 64) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_136 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_136.transpose(1, 2) + x_136 = None + x_137 = transpose_11.reshape(1, 197, 768) + transpose_11 = None + x_138 = torch._C._nn.linear( + x_137, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_137 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + mul_20 = x_139 * l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ + x_139 = l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ = None + x_140 = x_134 + mul_20 + x_134 = mul_20 = None + x_141 = torch.nn.functional.layer_norm( + x_140, + (768,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_141 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_143 = torch._C._nn.gelu(x_142, approximate="none") + x_142 = None + x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) + x_143 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_144 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + mul_21 = x_146 * l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ + x_146 = l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ = None + x_147 = x_140 + mul_21 + x_140 = mul_21 = None + x_148 = torch.nn.functional.layer_norm( + x_147, + (768,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_148, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_148 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 197, 3, 12, 64) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_149 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_149.transpose(1, 2) + x_149 = None + x_150 = transpose_12.reshape(1, 197, 768) + transpose_12 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + mul_22 = x_152 * l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ + x_152 = l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ = None + x_153 = x_147 + mul_22 + x_147 = mul_22 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (768,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_156 = torch._C._nn.gelu(x_155, approximate="none") + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + mul_23 = x_159 * l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ + x_159 = l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ = None + x_160 = x_153 + mul_23 + x_153 = mul_23 = None + x_161 = torch.nn.functional.layer_norm( + x_160, + (768,), + l_self_modules_norm_parameters_weight_, + l_self_modules_norm_parameters_bias_, + 1e-06, + ) + x_160 = ( + l_self_modules_norm_parameters_weight_ + ) = l_self_modules_norm_parameters_bias_ = None + x_162 = x_161[(slice(None, None, None), 0)] + x_161 = None + x_163 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_163 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_164,) diff --git a/samples/timm/deit3_base_patch16_224.fb_in1k/weight_meta.py b/samples/timm/deit3_base_patch16_224.fb_in1k/weight_meta.py new file mode 100644 index 0000000000..7844b32df5 --- /dev/null +++ b/samples/timm/deit3_base_patch16_224.fb_in1k/weight_meta.py @@ -0,0 +1,1796 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.228 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [768, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 196, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_weight_: + name = "L_self_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_bias_: + name = "L_self_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/deit3_huge_patch14_224.fb_in1k/graph_hash.txt b/samples/timm/deit3_huge_patch14_224.fb_in1k/graph_hash.txt new file mode 100644 index 0000000000..ca69e6da21 --- /dev/null +++ b/samples/timm/deit3_huge_patch14_224.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +707327cac3944d79a30f7b30bf5570beb58057638f3953b5b7d3a48b003a5750 \ No newline at end of file diff --git a/samples/timm/deit3_huge_patch14_224.fb_in1k/graph_net.json b/samples/timm/deit3_huge_patch14_224.fb_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/deit3_huge_patch14_224.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/deit3_huge_patch14_224.fb_in1k/input_meta.py b/samples/timm/deit3_huge_patch14_224.fb_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/deit3_huge_patch14_224.fb_in1k/input_tensor_constraints.py b/samples/timm/deit3_huge_patch14_224.fb_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/deit3_huge_patch14_224.fb_in1k/model.py b/samples/timm/deit3_huge_patch14_224.fb_in1k/model.py new file mode 100644 index 0000000000..0edee0fb1b --- /dev/null +++ b/samples/timm/deit3_huge_patch14_224.fb_in1k/model.py @@ -0,0 +1,4788 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_24_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_25_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_26_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_27_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_28_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_29_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_30_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_31_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_24_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_24_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_24_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_24_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_24_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_24_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_24_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_24_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_24_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_24_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_24_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_24_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_25_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_25_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_25_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_25_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_25_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_25_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_25_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_25_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_25_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_25_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_25_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_25_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_26_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_26_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_26_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_26_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_26_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_26_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_26_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_26_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_26_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_26_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_26_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_26_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_27_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_27_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_27_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_27_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_27_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_27_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_27_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_27_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_27_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_27_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_27_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_27_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_28_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_28_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_28_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_28_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_28_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_28_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_28_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_28_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_28_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_28_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_28_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_28_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_29_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_29_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_29_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_29_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_29_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_29_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_29_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_29_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_29_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_29_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_29_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_29_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_30_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_30_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_30_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_30_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_30_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_30_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_30_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_30_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_30_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_30_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_30_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_30_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_31_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_31_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_31_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_31_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_31_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_31_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_31_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_31_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_31_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_31_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_31_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_31_modules_ls2_parameters_gamma_ + ) + l_self_modules_norm_parameters_weight_ = L_self_modules_norm_parameters_weight_ + l_self_modules_norm_parameters_bias_ = L_self_modules_norm_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (14, 14), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + x_2 = x_1 + l_self_parameters_pos_embed_ + x_1 = l_self_parameters_pos_embed_ = None + x_3 = torch.cat([expand, x_2], dim=1) + expand = x_2 = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (1280,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 257, 3, 16, 80) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 257, 1280) + transpose_1 = None + x_8 = torch._C._nn.linear( + x_7, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + mul = x_9 * l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ + x_9 = l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ = None + x_10 = x_4 + mul + x_4 = mul = None + x_11 = torch.nn.functional.layer_norm( + x_10, + (1280,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_11 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_13 = torch._C._nn.gelu(x_12, approximate="none") + x_12 = None + x_14 = torch.nn.functional.dropout(x_13, 0.0, False, False) + x_13 = None + x_15 = torch._C._nn.linear( + x_14, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_14 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + mul_1 = x_16 * l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ + x_16 = l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ = None + x_17 = x_10 + mul_1 + x_10 = mul_1 = None + x_18 = torch.nn.functional.layer_norm( + x_17, + (1280,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_18, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_18 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 257, 3, 16, 80) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_19 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_19.transpose(1, 2) + x_19 = None + x_20 = transpose_2.reshape(1, 257, 1280) + transpose_2 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + mul_2 = x_22 * l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ + x_22 = l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ = None + x_23 = x_17 + mul_2 + x_17 = mul_2 = None + x_24 = torch.nn.functional.layer_norm( + x_23, + (1280,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_25 = torch._C._nn.linear( + x_24, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_24 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_26 = torch._C._nn.gelu(x_25, approximate="none") + x_25 = None + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + mul_3 = x_29 * l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ + x_29 = l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ = None + x_30 = x_23 + mul_3 + x_23 = mul_3 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (1280,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 257, 3, 16, 80) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_32 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_32.transpose(1, 2) + x_32 = None + x_33 = transpose_3.reshape(1, 257, 1280) + transpose_3 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_33 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + mul_4 = x_35 * l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ + x_35 = l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ = None + x_36 = x_30 + mul_4 + x_30 = mul_4 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (1280,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_39 = torch._C._nn.gelu(x_38, approximate="none") + x_38 = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = torch._C._nn.linear( + x_40, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_40 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_42 = torch.nn.functional.dropout(x_41, 0.0, False, False) + x_41 = None + mul_5 = x_42 * l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ + x_42 = l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ = None + x_43 = x_36 + mul_5 + x_36 = mul_5 = None + x_44 = torch.nn.functional.layer_norm( + x_43, + (1280,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_44 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 257, 3, 16, 80) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_45 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_45.transpose(1, 2) + x_45 = None + x_46 = transpose_4.reshape(1, 257, 1280) + transpose_4 = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + mul_6 = x_48 * l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ + x_48 = l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ = None + x_49 = x_43 + mul_6 + x_43 = mul_6 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (1280,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_51 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_52 = torch._C._nn.gelu(x_51, approximate="none") + x_51 = None + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + mul_7 = x_55 * l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ + x_55 = l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ = None + x_56 = x_49 + mul_7 + x_49 = mul_7 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (1280,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 257, 3, 16, 80) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_5.reshape(1, 257, 1280) + transpose_5 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + mul_8 = x_61 * l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ + x_61 = l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ = None + x_62 = x_56 + mul_8 + x_56 = mul_8 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (1280,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + mul_9 = x_68 * l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ + x_68 = l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ = None + x_69 = x_62 + mul_9 + x_62 = mul_9 = None + x_70 = torch.nn.functional.layer_norm( + x_69, + (1280,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_70 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 257, 3, 16, 80) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_71 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_71.transpose(1, 2) + x_71 = None + x_72 = transpose_6.reshape(1, 257, 1280) + transpose_6 = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + mul_10 = x_74 * l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ + x_74 = l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ = None + x_75 = x_69 + mul_10 + x_69 = mul_10 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (1280,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_78 = torch._C._nn.gelu(x_77, approximate="none") + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch._C._nn.linear( + x_79, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_79 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + mul_11 = x_81 * l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ + x_81 = l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ = None + x_82 = x_75 + mul_11 + x_75 = mul_11 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (1280,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 257, 3, 16, 80) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_84 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_84.transpose(1, 2) + x_84 = None + x_85 = transpose_7.reshape(1, 257, 1280) + transpose_7 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_85 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + mul_12 = x_87 * l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ + x_87 = l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ = None + x_88 = x_82 + mul_12 + x_82 = mul_12 = None + x_89 = torch.nn.functional.layer_norm( + x_88, + (1280,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_89 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_91 = torch._C._nn.gelu(x_90, approximate="none") + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + mul_13 = x_94 * l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ + x_94 = l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ = None + x_95 = x_88 + mul_13 + x_88 = mul_13 = None + x_96 = torch.nn.functional.layer_norm( + x_95, + (1280,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_96 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 257, 3, 16, 80) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_97 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_97.transpose(1, 2) + x_97 = None + x_98 = transpose_8.reshape(1, 257, 1280) + transpose_8 = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + mul_14 = x_100 * l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ + x_100 = l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ = None + x_101 = x_95 + mul_14 + x_95 = mul_14 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (1280,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch._C._nn.linear( + x_105, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_105 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + mul_15 = x_107 * l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ + x_107 = l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ = None + x_108 = x_101 + mul_15 + x_101 = mul_15 = None + x_109 = torch.nn.functional.layer_norm( + x_108, + (1280,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_109, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_109 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 257, 3, 16, 80) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_9.reshape(1, 257, 1280) + transpose_9 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + mul_16 = x_113 * l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ + x_113 = l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ = None + x_114 = x_108 + mul_16 + x_108 = mul_16 = None + x_115 = torch.nn.functional.layer_norm( + x_114, + (1280,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_116 = torch._C._nn.linear( + x_115, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_115 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + mul_17 = x_120 * l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ + x_120 = l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ = None + x_121 = x_114 + mul_17 + x_114 = mul_17 = None + x_122 = torch.nn.functional.layer_norm( + x_121, + (1280,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_122 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 257, 3, 16, 80) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_123 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_123.transpose(1, 2) + x_123 = None + x_124 = transpose_10.reshape(1, 257, 1280) + transpose_10 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_124 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + mul_18 = x_126 * l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ + x_126 = l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ = None + x_127 = x_121 + mul_18 + x_121 = mul_18 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (1280,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_130 = torch._C._nn.gelu(x_129, approximate="none") + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch._C._nn.linear( + x_131, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_131 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + mul_19 = x_133 * l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ + x_133 = l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ = None + x_134 = x_127 + mul_19 + x_127 = mul_19 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (1280,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_135, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_135 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 257, 3, 16, 80) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_136 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_136.transpose(1, 2) + x_136 = None + x_137 = transpose_11.reshape(1, 257, 1280) + transpose_11 = None + x_138 = torch._C._nn.linear( + x_137, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_137 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + mul_20 = x_139 * l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ + x_139 = l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ = None + x_140 = x_134 + mul_20 + x_134 = mul_20 = None + x_141 = torch.nn.functional.layer_norm( + x_140, + (1280,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_141 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_143 = torch._C._nn.gelu(x_142, approximate="none") + x_142 = None + x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) + x_143 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_144 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + mul_21 = x_146 * l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ + x_146 = l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ = None + x_147 = x_140 + mul_21 + x_140 = mul_21 = None + x_148 = torch.nn.functional.layer_norm( + x_147, + (1280,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_148, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_148 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 257, 3, 16, 80) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_149 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_149.transpose(1, 2) + x_149 = None + x_150 = transpose_12.reshape(1, 257, 1280) + transpose_12 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + mul_22 = x_152 * l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ + x_152 = l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ = None + x_153 = x_147 + mul_22 + x_147 = mul_22 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (1280,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_156 = torch._C._nn.gelu(x_155, approximate="none") + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + mul_23 = x_159 * l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ + x_159 = l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ = None + x_160 = x_153 + mul_23 + x_153 = mul_23 = None + x_161 = torch.nn.functional.layer_norm( + x_160, + (1280,), + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ + ) = None + linear_48 = torch._C._nn.linear( + x_161, + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_, + ) + x_161 = ( + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_24 = linear_48.reshape(1, 257, 3, 16, 80) + linear_48 = None + qkv_12 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_12 = unbind_12[0] + k_12 = unbind_12[1] + v_12 = unbind_12[2] + unbind_12 = None + x_162 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_13 = x_162.transpose(1, 2) + x_162 = None + x_163 = transpose_13.reshape(1, 257, 1280) + transpose_13 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_, + ) + x_163 = l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ + ) = None + x_165 = torch.nn.functional.dropout(x_164, 0.0, False, False) + x_164 = None + mul_24 = x_165 * l_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_ + x_165 = l_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_ = None + x_166 = x_160 + mul_24 + x_160 = mul_24 = None + x_167 = torch.nn.functional.layer_norm( + x_166, + (1280,), + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ + ) = None + x_168 = torch._C._nn.linear( + x_167, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_167 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_169 = torch._C._nn.gelu(x_168, approximate="none") + x_168 = None + x_170 = torch.nn.functional.dropout(x_169, 0.0, False, False) + x_169 = None + x_171 = torch._C._nn.linear( + x_170, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_170 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_172 = torch.nn.functional.dropout(x_171, 0.0, False, False) + x_171 = None + mul_25 = x_172 * l_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_ + x_172 = l_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_ = None + x_173 = x_166 + mul_25 + x_166 = mul_25 = None + x_174 = torch.nn.functional.layer_norm( + x_173, + (1280,), + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ + ) = None + linear_52 = torch._C._nn.linear( + x_174, + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_, + ) + x_174 = ( + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_26 = linear_52.reshape(1, 257, 3, 16, 80) + linear_52 = None + qkv_13 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_13 = unbind_13[0] + k_13 = unbind_13[1] + v_13 = unbind_13[2] + unbind_13 = None + x_175 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_14 = x_175.transpose(1, 2) + x_175 = None + x_176 = transpose_14.reshape(1, 257, 1280) + transpose_14 = None + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_, + ) + x_176 = l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ + ) = None + x_178 = torch.nn.functional.dropout(x_177, 0.0, False, False) + x_177 = None + mul_26 = x_178 * l_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_ + x_178 = l_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_ = None + x_179 = x_173 + mul_26 + x_173 = mul_26 = None + x_180 = torch.nn.functional.layer_norm( + x_179, + (1280,), + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ + ) = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_180 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_182 = torch._C._nn.gelu(x_181, approximate="none") + x_181 = None + x_183 = torch.nn.functional.dropout(x_182, 0.0, False, False) + x_182 = None + x_184 = torch._C._nn.linear( + x_183, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_183 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_185 = torch.nn.functional.dropout(x_184, 0.0, False, False) + x_184 = None + mul_27 = x_185 * l_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_ + x_185 = l_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_ = None + x_186 = x_179 + mul_27 + x_179 = mul_27 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (1280,), + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ + ) = None + linear_56 = torch._C._nn.linear( + x_187, + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_, + ) + x_187 = ( + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_28 = linear_56.reshape(1, 257, 3, 16, 80) + linear_56 = None + qkv_14 = reshape_28.permute(2, 0, 3, 1, 4) + reshape_28 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_14 = unbind_14[0] + k_14 = unbind_14[1] + v_14 = unbind_14[2] + unbind_14 = None + x_188 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_15 = x_188.transpose(1, 2) + x_188 = None + x_189 = transpose_15.reshape(1, 257, 1280) + transpose_15 = None + x_190 = torch._C._nn.linear( + x_189, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_, + ) + x_189 = l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ + ) = None + x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) + x_190 = None + mul_28 = x_191 * l_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_ + x_191 = l_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_ = None + x_192 = x_186 + mul_28 + x_186 = mul_28 = None + x_193 = torch.nn.functional.layer_norm( + x_192, + (1280,), + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ + ) = None + x_194 = torch._C._nn.linear( + x_193, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_193 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_195 = torch._C._nn.gelu(x_194, approximate="none") + x_194 = None + x_196 = torch.nn.functional.dropout(x_195, 0.0, False, False) + x_195 = None + x_197 = torch._C._nn.linear( + x_196, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_196 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_198 = torch.nn.functional.dropout(x_197, 0.0, False, False) + x_197 = None + mul_29 = x_198 * l_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_ + x_198 = l_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_ = None + x_199 = x_192 + mul_29 + x_192 = mul_29 = None + x_200 = torch.nn.functional.layer_norm( + x_199, + (1280,), + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ + ) = None + linear_60 = torch._C._nn.linear( + x_200, + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_, + ) + x_200 = ( + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_30 = linear_60.reshape(1, 257, 3, 16, 80) + linear_60 = None + qkv_15 = reshape_30.permute(2, 0, 3, 1, 4) + reshape_30 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_15 = unbind_15[0] + k_15 = unbind_15[1] + v_15 = unbind_15[2] + unbind_15 = None + x_201 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_16 = x_201.transpose(1, 2) + x_201 = None + x_202 = transpose_16.reshape(1, 257, 1280) + transpose_16 = None + x_203 = torch._C._nn.linear( + x_202, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_, + ) + x_202 = l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ + ) = None + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + mul_30 = x_204 * l_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_ + x_204 = l_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_ = None + x_205 = x_199 + mul_30 + x_199 = mul_30 = None + x_206 = torch.nn.functional.layer_norm( + x_205, + (1280,), + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ + ) = None + x_207 = torch._C._nn.linear( + x_206, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_206 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_208 = torch._C._nn.gelu(x_207, approximate="none") + x_207 = None + x_209 = torch.nn.functional.dropout(x_208, 0.0, False, False) + x_208 = None + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_209 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_211 = torch.nn.functional.dropout(x_210, 0.0, False, False) + x_210 = None + mul_31 = x_211 * l_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_ + x_211 = l_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_ = None + x_212 = x_205 + mul_31 + x_205 = mul_31 = None + x_213 = torch.nn.functional.layer_norm( + x_212, + (1280,), + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ + ) = None + linear_64 = torch._C._nn.linear( + x_213, + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_, + ) + x_213 = ( + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_32 = linear_64.reshape(1, 257, 3, 16, 80) + linear_64 = None + qkv_16 = reshape_32.permute(2, 0, 3, 1, 4) + reshape_32 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_16 = unbind_16[0] + k_16 = unbind_16[1] + v_16 = unbind_16[2] + unbind_16 = None + x_214 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, attn_mask=None, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_17 = x_214.transpose(1, 2) + x_214 = None + x_215 = transpose_17.reshape(1, 257, 1280) + transpose_17 = None + x_216 = torch._C._nn.linear( + x_215, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_, + ) + x_215 = l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ + ) = None + x_217 = torch.nn.functional.dropout(x_216, 0.0, False, False) + x_216 = None + mul_32 = x_217 * l_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_ + x_217 = l_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_ = None + x_218 = x_212 + mul_32 + x_212 = mul_32 = None + x_219 = torch.nn.functional.layer_norm( + x_218, + (1280,), + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ + ) = None + x_220 = torch._C._nn.linear( + x_219, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_219 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_221 = torch._C._nn.gelu(x_220, approximate="none") + x_220 = None + x_222 = torch.nn.functional.dropout(x_221, 0.0, False, False) + x_221 = None + x_223 = torch._C._nn.linear( + x_222, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_222 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_224 = torch.nn.functional.dropout(x_223, 0.0, False, False) + x_223 = None + mul_33 = x_224 * l_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_ + x_224 = l_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_ = None + x_225 = x_218 + mul_33 + x_218 = mul_33 = None + x_226 = torch.nn.functional.layer_norm( + x_225, + (1280,), + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ + ) = None + linear_68 = torch._C._nn.linear( + x_226, + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_, + ) + x_226 = ( + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_34 = linear_68.reshape(1, 257, 3, 16, 80) + linear_68 = None + qkv_17 = reshape_34.permute(2, 0, 3, 1, 4) + reshape_34 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_17 = unbind_17[0] + k_17 = unbind_17[1] + v_17 = unbind_17[2] + unbind_17 = None + x_227 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, attn_mask=None, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_18 = x_227.transpose(1, 2) + x_227 = None + x_228 = transpose_18.reshape(1, 257, 1280) + transpose_18 = None + x_229 = torch._C._nn.linear( + x_228, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_, + ) + x_228 = l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ + ) = None + x_230 = torch.nn.functional.dropout(x_229, 0.0, False, False) + x_229 = None + mul_34 = x_230 * l_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_ + x_230 = l_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_ = None + x_231 = x_225 + mul_34 + x_225 = mul_34 = None + x_232 = torch.nn.functional.layer_norm( + x_231, + (1280,), + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ + ) = None + x_233 = torch._C._nn.linear( + x_232, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_232 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_234 = torch._C._nn.gelu(x_233, approximate="none") + x_233 = None + x_235 = torch.nn.functional.dropout(x_234, 0.0, False, False) + x_234 = None + x_236 = torch._C._nn.linear( + x_235, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_235 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_237 = torch.nn.functional.dropout(x_236, 0.0, False, False) + x_236 = None + mul_35 = x_237 * l_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_ + x_237 = l_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_ = None + x_238 = x_231 + mul_35 + x_231 = mul_35 = None + x_239 = torch.nn.functional.layer_norm( + x_238, + (1280,), + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ + ) = None + linear_72 = torch._C._nn.linear( + x_239, + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_, + ) + x_239 = ( + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_36 = linear_72.reshape(1, 257, 3, 16, 80) + linear_72 = None + qkv_18 = reshape_36.permute(2, 0, 3, 1, 4) + reshape_36 = None + unbind_18 = qkv_18.unbind(0) + qkv_18 = None + q_18 = unbind_18[0] + k_18 = unbind_18[1] + v_18 = unbind_18[2] + unbind_18 = None + x_240 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_19 = x_240.transpose(1, 2) + x_240 = None + x_241 = transpose_19.reshape(1, 257, 1280) + transpose_19 = None + x_242 = torch._C._nn.linear( + x_241, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_, + ) + x_241 = l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ + ) = None + x_243 = torch.nn.functional.dropout(x_242, 0.0, False, False) + x_242 = None + mul_36 = x_243 * l_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_ + x_243 = l_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_ = None + x_244 = x_238 + mul_36 + x_238 = mul_36 = None + x_245 = torch.nn.functional.layer_norm( + x_244, + (1280,), + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ + ) = None + x_246 = torch._C._nn.linear( + x_245, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_245 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_247 = torch._C._nn.gelu(x_246, approximate="none") + x_246 = None + x_248 = torch.nn.functional.dropout(x_247, 0.0, False, False) + x_247 = None + x_249 = torch._C._nn.linear( + x_248, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_248 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_250 = torch.nn.functional.dropout(x_249, 0.0, False, False) + x_249 = None + mul_37 = x_250 * l_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_ + x_250 = l_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_ = None + x_251 = x_244 + mul_37 + x_244 = mul_37 = None + x_252 = torch.nn.functional.layer_norm( + x_251, + (1280,), + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ + ) = None + linear_76 = torch._C._nn.linear( + x_252, + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_, + ) + x_252 = ( + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_38 = linear_76.reshape(1, 257, 3, 16, 80) + linear_76 = None + qkv_19 = reshape_38.permute(2, 0, 3, 1, 4) + reshape_38 = None + unbind_19 = qkv_19.unbind(0) + qkv_19 = None + q_19 = unbind_19[0] + k_19 = unbind_19[1] + v_19 = unbind_19[2] + unbind_19 = None + x_253 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, attn_mask=None, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_20 = x_253.transpose(1, 2) + x_253 = None + x_254 = transpose_20.reshape(1, 257, 1280) + transpose_20 = None + x_255 = torch._C._nn.linear( + x_254, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_, + ) + x_254 = l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ + ) = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + mul_38 = x_256 * l_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_ + x_256 = l_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_ = None + x_257 = x_251 + mul_38 + x_251 = mul_38 = None + x_258 = torch.nn.functional.layer_norm( + x_257, + (1280,), + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ + ) = None + x_259 = torch._C._nn.linear( + x_258, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_258 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_260 = torch._C._nn.gelu(x_259, approximate="none") + x_259 = None + x_261 = torch.nn.functional.dropout(x_260, 0.0, False, False) + x_260 = None + x_262 = torch._C._nn.linear( + x_261, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_261 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_263 = torch.nn.functional.dropout(x_262, 0.0, False, False) + x_262 = None + mul_39 = x_263 * l_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_ + x_263 = l_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_ = None + x_264 = x_257 + mul_39 + x_257 = mul_39 = None + x_265 = torch.nn.functional.layer_norm( + x_264, + (1280,), + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ + ) = None + linear_80 = torch._C._nn.linear( + x_265, + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_, + ) + x_265 = ( + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_40 = linear_80.reshape(1, 257, 3, 16, 80) + linear_80 = None + qkv_20 = reshape_40.permute(2, 0, 3, 1, 4) + reshape_40 = None + unbind_20 = qkv_20.unbind(0) + qkv_20 = None + q_20 = unbind_20[0] + k_20 = unbind_20[1] + v_20 = unbind_20[2] + unbind_20 = None + x_266 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, attn_mask=None, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_21 = x_266.transpose(1, 2) + x_266 = None + x_267 = transpose_21.reshape(1, 257, 1280) + transpose_21 = None + x_268 = torch._C._nn.linear( + x_267, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_, + ) + x_267 = l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ + ) = None + x_269 = torch.nn.functional.dropout(x_268, 0.0, False, False) + x_268 = None + mul_40 = x_269 * l_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_ + x_269 = l_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_ = None + x_270 = x_264 + mul_40 + x_264 = mul_40 = None + x_271 = torch.nn.functional.layer_norm( + x_270, + (1280,), + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ + ) = None + x_272 = torch._C._nn.linear( + x_271, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_271 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_273 = torch._C._nn.gelu(x_272, approximate="none") + x_272 = None + x_274 = torch.nn.functional.dropout(x_273, 0.0, False, False) + x_273 = None + x_275 = torch._C._nn.linear( + x_274, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_274 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_276 = torch.nn.functional.dropout(x_275, 0.0, False, False) + x_275 = None + mul_41 = x_276 * l_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_ + x_276 = l_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_ = None + x_277 = x_270 + mul_41 + x_270 = mul_41 = None + x_278 = torch.nn.functional.layer_norm( + x_277, + (1280,), + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ + ) = None + linear_84 = torch._C._nn.linear( + x_278, + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_, + ) + x_278 = ( + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_42 = linear_84.reshape(1, 257, 3, 16, 80) + linear_84 = None + qkv_21 = reshape_42.permute(2, 0, 3, 1, 4) + reshape_42 = None + unbind_21 = qkv_21.unbind(0) + qkv_21 = None + q_21 = unbind_21[0] + k_21 = unbind_21[1] + v_21 = unbind_21[2] + unbind_21 = None + x_279 = torch._C._nn.scaled_dot_product_attention( + q_21, k_21, v_21, attn_mask=None, dropout_p=0.0 + ) + q_21 = k_21 = v_21 = None + transpose_22 = x_279.transpose(1, 2) + x_279 = None + x_280 = transpose_22.reshape(1, 257, 1280) + transpose_22 = None + x_281 = torch._C._nn.linear( + x_280, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_, + ) + x_280 = l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ + ) = None + x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) + x_281 = None + mul_42 = x_282 * l_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_ + x_282 = l_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_ = None + x_283 = x_277 + mul_42 + x_277 = mul_42 = None + x_284 = torch.nn.functional.layer_norm( + x_283, + (1280,), + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ + ) = None + x_285 = torch._C._nn.linear( + x_284, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_284 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_286 = torch._C._nn.gelu(x_285, approximate="none") + x_285 = None + x_287 = torch.nn.functional.dropout(x_286, 0.0, False, False) + x_286 = None + x_288 = torch._C._nn.linear( + x_287, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_287 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + mul_43 = x_289 * l_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_ + x_289 = l_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_ = None + x_290 = x_283 + mul_43 + x_283 = mul_43 = None + x_291 = torch.nn.functional.layer_norm( + x_290, + (1280,), + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ + ) = None + linear_88 = torch._C._nn.linear( + x_291, + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_, + ) + x_291 = ( + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_44 = linear_88.reshape(1, 257, 3, 16, 80) + linear_88 = None + qkv_22 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_22 = qkv_22.unbind(0) + qkv_22 = None + q_22 = unbind_22[0] + k_22 = unbind_22[1] + v_22 = unbind_22[2] + unbind_22 = None + x_292 = torch._C._nn.scaled_dot_product_attention( + q_22, k_22, v_22, attn_mask=None, dropout_p=0.0 + ) + q_22 = k_22 = v_22 = None + transpose_23 = x_292.transpose(1, 2) + x_292 = None + x_293 = transpose_23.reshape(1, 257, 1280) + transpose_23 = None + x_294 = torch._C._nn.linear( + x_293, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_, + ) + x_293 = l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ + ) = None + x_295 = torch.nn.functional.dropout(x_294, 0.0, False, False) + x_294 = None + mul_44 = x_295 * l_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_ + x_295 = l_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_ = None + x_296 = x_290 + mul_44 + x_290 = mul_44 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (1280,), + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ + ) = None + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_299 = torch._C._nn.gelu(x_298, approximate="none") + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_301 = torch._C._nn.linear( + x_300, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_300 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_302 = torch.nn.functional.dropout(x_301, 0.0, False, False) + x_301 = None + mul_45 = x_302 * l_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_ + x_302 = l_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_ = None + x_303 = x_296 + mul_45 + x_296 = mul_45 = None + x_304 = torch.nn.functional.layer_norm( + x_303, + (1280,), + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ + ) = None + linear_92 = torch._C._nn.linear( + x_304, + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_, + ) + x_304 = ( + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_46 = linear_92.reshape(1, 257, 3, 16, 80) + linear_92 = None + qkv_23 = reshape_46.permute(2, 0, 3, 1, 4) + reshape_46 = None + unbind_23 = qkv_23.unbind(0) + qkv_23 = None + q_23 = unbind_23[0] + k_23 = unbind_23[1] + v_23 = unbind_23[2] + unbind_23 = None + x_305 = torch._C._nn.scaled_dot_product_attention( + q_23, k_23, v_23, attn_mask=None, dropout_p=0.0 + ) + q_23 = k_23 = v_23 = None + transpose_24 = x_305.transpose(1, 2) + x_305 = None + x_306 = transpose_24.reshape(1, 257, 1280) + transpose_24 = None + x_307 = torch._C._nn.linear( + x_306, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_, + ) + x_306 = l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ + ) = None + x_308 = torch.nn.functional.dropout(x_307, 0.0, False, False) + x_307 = None + mul_46 = x_308 * l_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_ + x_308 = l_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_ = None + x_309 = x_303 + mul_46 + x_303 = mul_46 = None + x_310 = torch.nn.functional.layer_norm( + x_309, + (1280,), + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ + ) = None + x_311 = torch._C._nn.linear( + x_310, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_310 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_312 = torch._C._nn.gelu(x_311, approximate="none") + x_311 = None + x_313 = torch.nn.functional.dropout(x_312, 0.0, False, False) + x_312 = None + x_314 = torch._C._nn.linear( + x_313, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_313 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) + x_314 = None + mul_47 = x_315 * l_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_ + x_315 = l_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_ = None + x_316 = x_309 + mul_47 + x_309 = mul_47 = None + x_317 = torch.nn.functional.layer_norm( + x_316, + (1280,), + l_self_modules_blocks_modules_24_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_24_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_24_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_24_modules_norm1_parameters_bias_ + ) = None + linear_96 = torch._C._nn.linear( + x_317, + l_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_bias_, + ) + x_317 = ( + l_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_48 = linear_96.reshape(1, 257, 3, 16, 80) + linear_96 = None + qkv_24 = reshape_48.permute(2, 0, 3, 1, 4) + reshape_48 = None + unbind_24 = qkv_24.unbind(0) + qkv_24 = None + q_24 = unbind_24[0] + k_24 = unbind_24[1] + v_24 = unbind_24[2] + unbind_24 = None + x_318 = torch._C._nn.scaled_dot_product_attention( + q_24, k_24, v_24, attn_mask=None, dropout_p=0.0 + ) + q_24 = k_24 = v_24 = None + transpose_25 = x_318.transpose(1, 2) + x_318 = None + x_319 = transpose_25.reshape(1, 257, 1280) + transpose_25 = None + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_bias_, + ) + x_319 = l_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_bias_ + ) = None + x_321 = torch.nn.functional.dropout(x_320, 0.0, False, False) + x_320 = None + mul_48 = x_321 * l_self_modules_blocks_modules_24_modules_ls1_parameters_gamma_ + x_321 = l_self_modules_blocks_modules_24_modules_ls1_parameters_gamma_ = None + x_322 = x_316 + mul_48 + x_316 = mul_48 = None + x_323 = torch.nn.functional.layer_norm( + x_322, + (1280,), + l_self_modules_blocks_modules_24_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_24_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_24_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_24_modules_norm2_parameters_bias_ + ) = None + x_324 = torch._C._nn.linear( + x_323, + l_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_, + ) + x_323 = ( + l_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_325 = torch._C._nn.gelu(x_324, approximate="none") + x_324 = None + x_326 = torch.nn.functional.dropout(x_325, 0.0, False, False) + x_325 = None + x_327 = torch._C._nn.linear( + x_326, + l_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_, + ) + x_326 = ( + l_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_328 = torch.nn.functional.dropout(x_327, 0.0, False, False) + x_327 = None + mul_49 = x_328 * l_self_modules_blocks_modules_24_modules_ls2_parameters_gamma_ + x_328 = l_self_modules_blocks_modules_24_modules_ls2_parameters_gamma_ = None + x_329 = x_322 + mul_49 + x_322 = mul_49 = None + x_330 = torch.nn.functional.layer_norm( + x_329, + (1280,), + l_self_modules_blocks_modules_25_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_25_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_25_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_25_modules_norm1_parameters_bias_ + ) = None + linear_100 = torch._C._nn.linear( + x_330, + l_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_bias_, + ) + x_330 = ( + l_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_50 = linear_100.reshape(1, 257, 3, 16, 80) + linear_100 = None + qkv_25 = reshape_50.permute(2, 0, 3, 1, 4) + reshape_50 = None + unbind_25 = qkv_25.unbind(0) + qkv_25 = None + q_25 = unbind_25[0] + k_25 = unbind_25[1] + v_25 = unbind_25[2] + unbind_25 = None + x_331 = torch._C._nn.scaled_dot_product_attention( + q_25, k_25, v_25, attn_mask=None, dropout_p=0.0 + ) + q_25 = k_25 = v_25 = None + transpose_26 = x_331.transpose(1, 2) + x_331 = None + x_332 = transpose_26.reshape(1, 257, 1280) + transpose_26 = None + x_333 = torch._C._nn.linear( + x_332, + l_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_bias_, + ) + x_332 = l_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_bias_ + ) = None + x_334 = torch.nn.functional.dropout(x_333, 0.0, False, False) + x_333 = None + mul_50 = x_334 * l_self_modules_blocks_modules_25_modules_ls1_parameters_gamma_ + x_334 = l_self_modules_blocks_modules_25_modules_ls1_parameters_gamma_ = None + x_335 = x_329 + mul_50 + x_329 = mul_50 = None + x_336 = torch.nn.functional.layer_norm( + x_335, + (1280,), + l_self_modules_blocks_modules_25_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_25_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_25_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_25_modules_norm2_parameters_bias_ + ) = None + x_337 = torch._C._nn.linear( + x_336, + l_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_, + ) + x_336 = ( + l_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_338 = torch._C._nn.gelu(x_337, approximate="none") + x_337 = None + x_339 = torch.nn.functional.dropout(x_338, 0.0, False, False) + x_338 = None + x_340 = torch._C._nn.linear( + x_339, + l_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_, + ) + x_339 = ( + l_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_341 = torch.nn.functional.dropout(x_340, 0.0, False, False) + x_340 = None + mul_51 = x_341 * l_self_modules_blocks_modules_25_modules_ls2_parameters_gamma_ + x_341 = l_self_modules_blocks_modules_25_modules_ls2_parameters_gamma_ = None + x_342 = x_335 + mul_51 + x_335 = mul_51 = None + x_343 = torch.nn.functional.layer_norm( + x_342, + (1280,), + l_self_modules_blocks_modules_26_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_26_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_26_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_26_modules_norm1_parameters_bias_ + ) = None + linear_104 = torch._C._nn.linear( + x_343, + l_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_bias_, + ) + x_343 = ( + l_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_52 = linear_104.reshape(1, 257, 3, 16, 80) + linear_104 = None + qkv_26 = reshape_52.permute(2, 0, 3, 1, 4) + reshape_52 = None + unbind_26 = qkv_26.unbind(0) + qkv_26 = None + q_26 = unbind_26[0] + k_26 = unbind_26[1] + v_26 = unbind_26[2] + unbind_26 = None + x_344 = torch._C._nn.scaled_dot_product_attention( + q_26, k_26, v_26, attn_mask=None, dropout_p=0.0 + ) + q_26 = k_26 = v_26 = None + transpose_27 = x_344.transpose(1, 2) + x_344 = None + x_345 = transpose_27.reshape(1, 257, 1280) + transpose_27 = None + x_346 = torch._C._nn.linear( + x_345, + l_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_bias_, + ) + x_345 = l_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_bias_ + ) = None + x_347 = torch.nn.functional.dropout(x_346, 0.0, False, False) + x_346 = None + mul_52 = x_347 * l_self_modules_blocks_modules_26_modules_ls1_parameters_gamma_ + x_347 = l_self_modules_blocks_modules_26_modules_ls1_parameters_gamma_ = None + x_348 = x_342 + mul_52 + x_342 = mul_52 = None + x_349 = torch.nn.functional.layer_norm( + x_348, + (1280,), + l_self_modules_blocks_modules_26_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_26_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_26_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_26_modules_norm2_parameters_bias_ + ) = None + x_350 = torch._C._nn.linear( + x_349, + l_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_, + ) + x_349 = ( + l_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_351 = torch._C._nn.gelu(x_350, approximate="none") + x_350 = None + x_352 = torch.nn.functional.dropout(x_351, 0.0, False, False) + x_351 = None + x_353 = torch._C._nn.linear( + x_352, + l_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_, + ) + x_352 = ( + l_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_354 = torch.nn.functional.dropout(x_353, 0.0, False, False) + x_353 = None + mul_53 = x_354 * l_self_modules_blocks_modules_26_modules_ls2_parameters_gamma_ + x_354 = l_self_modules_blocks_modules_26_modules_ls2_parameters_gamma_ = None + x_355 = x_348 + mul_53 + x_348 = mul_53 = None + x_356 = torch.nn.functional.layer_norm( + x_355, + (1280,), + l_self_modules_blocks_modules_27_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_27_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_27_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_27_modules_norm1_parameters_bias_ + ) = None + linear_108 = torch._C._nn.linear( + x_356, + l_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_bias_, + ) + x_356 = ( + l_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_54 = linear_108.reshape(1, 257, 3, 16, 80) + linear_108 = None + qkv_27 = reshape_54.permute(2, 0, 3, 1, 4) + reshape_54 = None + unbind_27 = qkv_27.unbind(0) + qkv_27 = None + q_27 = unbind_27[0] + k_27 = unbind_27[1] + v_27 = unbind_27[2] + unbind_27 = None + x_357 = torch._C._nn.scaled_dot_product_attention( + q_27, k_27, v_27, attn_mask=None, dropout_p=0.0 + ) + q_27 = k_27 = v_27 = None + transpose_28 = x_357.transpose(1, 2) + x_357 = None + x_358 = transpose_28.reshape(1, 257, 1280) + transpose_28 = None + x_359 = torch._C._nn.linear( + x_358, + l_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_bias_, + ) + x_358 = l_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_bias_ + ) = None + x_360 = torch.nn.functional.dropout(x_359, 0.0, False, False) + x_359 = None + mul_54 = x_360 * l_self_modules_blocks_modules_27_modules_ls1_parameters_gamma_ + x_360 = l_self_modules_blocks_modules_27_modules_ls1_parameters_gamma_ = None + x_361 = x_355 + mul_54 + x_355 = mul_54 = None + x_362 = torch.nn.functional.layer_norm( + x_361, + (1280,), + l_self_modules_blocks_modules_27_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_27_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_27_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_27_modules_norm2_parameters_bias_ + ) = None + x_363 = torch._C._nn.linear( + x_362, + l_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_, + ) + x_362 = ( + l_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_364 = torch._C._nn.gelu(x_363, approximate="none") + x_363 = None + x_365 = torch.nn.functional.dropout(x_364, 0.0, False, False) + x_364 = None + x_366 = torch._C._nn.linear( + x_365, + l_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_, + ) + x_365 = ( + l_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_367 = torch.nn.functional.dropout(x_366, 0.0, False, False) + x_366 = None + mul_55 = x_367 * l_self_modules_blocks_modules_27_modules_ls2_parameters_gamma_ + x_367 = l_self_modules_blocks_modules_27_modules_ls2_parameters_gamma_ = None + x_368 = x_361 + mul_55 + x_361 = mul_55 = None + x_369 = torch.nn.functional.layer_norm( + x_368, + (1280,), + l_self_modules_blocks_modules_28_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_28_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_28_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_28_modules_norm1_parameters_bias_ + ) = None + linear_112 = torch._C._nn.linear( + x_369, + l_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_bias_, + ) + x_369 = ( + l_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_56 = linear_112.reshape(1, 257, 3, 16, 80) + linear_112 = None + qkv_28 = reshape_56.permute(2, 0, 3, 1, 4) + reshape_56 = None + unbind_28 = qkv_28.unbind(0) + qkv_28 = None + q_28 = unbind_28[0] + k_28 = unbind_28[1] + v_28 = unbind_28[2] + unbind_28 = None + x_370 = torch._C._nn.scaled_dot_product_attention( + q_28, k_28, v_28, attn_mask=None, dropout_p=0.0 + ) + q_28 = k_28 = v_28 = None + transpose_29 = x_370.transpose(1, 2) + x_370 = None + x_371 = transpose_29.reshape(1, 257, 1280) + transpose_29 = None + x_372 = torch._C._nn.linear( + x_371, + l_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_bias_, + ) + x_371 = l_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_bias_ + ) = None + x_373 = torch.nn.functional.dropout(x_372, 0.0, False, False) + x_372 = None + mul_56 = x_373 * l_self_modules_blocks_modules_28_modules_ls1_parameters_gamma_ + x_373 = l_self_modules_blocks_modules_28_modules_ls1_parameters_gamma_ = None + x_374 = x_368 + mul_56 + x_368 = mul_56 = None + x_375 = torch.nn.functional.layer_norm( + x_374, + (1280,), + l_self_modules_blocks_modules_28_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_28_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_28_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_28_modules_norm2_parameters_bias_ + ) = None + x_376 = torch._C._nn.linear( + x_375, + l_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_, + ) + x_375 = ( + l_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_377 = torch._C._nn.gelu(x_376, approximate="none") + x_376 = None + x_378 = torch.nn.functional.dropout(x_377, 0.0, False, False) + x_377 = None + x_379 = torch._C._nn.linear( + x_378, + l_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_, + ) + x_378 = ( + l_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_380 = torch.nn.functional.dropout(x_379, 0.0, False, False) + x_379 = None + mul_57 = x_380 * l_self_modules_blocks_modules_28_modules_ls2_parameters_gamma_ + x_380 = l_self_modules_blocks_modules_28_modules_ls2_parameters_gamma_ = None + x_381 = x_374 + mul_57 + x_374 = mul_57 = None + x_382 = torch.nn.functional.layer_norm( + x_381, + (1280,), + l_self_modules_blocks_modules_29_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_29_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_29_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_29_modules_norm1_parameters_bias_ + ) = None + linear_116 = torch._C._nn.linear( + x_382, + l_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_bias_, + ) + x_382 = ( + l_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_58 = linear_116.reshape(1, 257, 3, 16, 80) + linear_116 = None + qkv_29 = reshape_58.permute(2, 0, 3, 1, 4) + reshape_58 = None + unbind_29 = qkv_29.unbind(0) + qkv_29 = None + q_29 = unbind_29[0] + k_29 = unbind_29[1] + v_29 = unbind_29[2] + unbind_29 = None + x_383 = torch._C._nn.scaled_dot_product_attention( + q_29, k_29, v_29, attn_mask=None, dropout_p=0.0 + ) + q_29 = k_29 = v_29 = None + transpose_30 = x_383.transpose(1, 2) + x_383 = None + x_384 = transpose_30.reshape(1, 257, 1280) + transpose_30 = None + x_385 = torch._C._nn.linear( + x_384, + l_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_bias_, + ) + x_384 = l_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_bias_ + ) = None + x_386 = torch.nn.functional.dropout(x_385, 0.0, False, False) + x_385 = None + mul_58 = x_386 * l_self_modules_blocks_modules_29_modules_ls1_parameters_gamma_ + x_386 = l_self_modules_blocks_modules_29_modules_ls1_parameters_gamma_ = None + x_387 = x_381 + mul_58 + x_381 = mul_58 = None + x_388 = torch.nn.functional.layer_norm( + x_387, + (1280,), + l_self_modules_blocks_modules_29_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_29_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_29_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_29_modules_norm2_parameters_bias_ + ) = None + x_389 = torch._C._nn.linear( + x_388, + l_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_, + ) + x_388 = ( + l_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_390 = torch._C._nn.gelu(x_389, approximate="none") + x_389 = None + x_391 = torch.nn.functional.dropout(x_390, 0.0, False, False) + x_390 = None + x_392 = torch._C._nn.linear( + x_391, + l_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_, + ) + x_391 = ( + l_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_393 = torch.nn.functional.dropout(x_392, 0.0, False, False) + x_392 = None + mul_59 = x_393 * l_self_modules_blocks_modules_29_modules_ls2_parameters_gamma_ + x_393 = l_self_modules_blocks_modules_29_modules_ls2_parameters_gamma_ = None + x_394 = x_387 + mul_59 + x_387 = mul_59 = None + x_395 = torch.nn.functional.layer_norm( + x_394, + (1280,), + l_self_modules_blocks_modules_30_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_30_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_30_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_30_modules_norm1_parameters_bias_ + ) = None + linear_120 = torch._C._nn.linear( + x_395, + l_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_bias_, + ) + x_395 = ( + l_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_60 = linear_120.reshape(1, 257, 3, 16, 80) + linear_120 = None + qkv_30 = reshape_60.permute(2, 0, 3, 1, 4) + reshape_60 = None + unbind_30 = qkv_30.unbind(0) + qkv_30 = None + q_30 = unbind_30[0] + k_30 = unbind_30[1] + v_30 = unbind_30[2] + unbind_30 = None + x_396 = torch._C._nn.scaled_dot_product_attention( + q_30, k_30, v_30, attn_mask=None, dropout_p=0.0 + ) + q_30 = k_30 = v_30 = None + transpose_31 = x_396.transpose(1, 2) + x_396 = None + x_397 = transpose_31.reshape(1, 257, 1280) + transpose_31 = None + x_398 = torch._C._nn.linear( + x_397, + l_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_bias_, + ) + x_397 = l_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_bias_ + ) = None + x_399 = torch.nn.functional.dropout(x_398, 0.0, False, False) + x_398 = None + mul_60 = x_399 * l_self_modules_blocks_modules_30_modules_ls1_parameters_gamma_ + x_399 = l_self_modules_blocks_modules_30_modules_ls1_parameters_gamma_ = None + x_400 = x_394 + mul_60 + x_394 = mul_60 = None + x_401 = torch.nn.functional.layer_norm( + x_400, + (1280,), + l_self_modules_blocks_modules_30_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_30_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_30_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_30_modules_norm2_parameters_bias_ + ) = None + x_402 = torch._C._nn.linear( + x_401, + l_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_bias_, + ) + x_401 = ( + l_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_403 = torch._C._nn.gelu(x_402, approximate="none") + x_402 = None + x_404 = torch.nn.functional.dropout(x_403, 0.0, False, False) + x_403 = None + x_405 = torch._C._nn.linear( + x_404, + l_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_bias_, + ) + x_404 = ( + l_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_406 = torch.nn.functional.dropout(x_405, 0.0, False, False) + x_405 = None + mul_61 = x_406 * l_self_modules_blocks_modules_30_modules_ls2_parameters_gamma_ + x_406 = l_self_modules_blocks_modules_30_modules_ls2_parameters_gamma_ = None + x_407 = x_400 + mul_61 + x_400 = mul_61 = None + x_408 = torch.nn.functional.layer_norm( + x_407, + (1280,), + l_self_modules_blocks_modules_31_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_31_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_31_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_31_modules_norm1_parameters_bias_ + ) = None + linear_124 = torch._C._nn.linear( + x_408, + l_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_bias_, + ) + x_408 = ( + l_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_62 = linear_124.reshape(1, 257, 3, 16, 80) + linear_124 = None + qkv_31 = reshape_62.permute(2, 0, 3, 1, 4) + reshape_62 = None + unbind_31 = qkv_31.unbind(0) + qkv_31 = None + q_31 = unbind_31[0] + k_31 = unbind_31[1] + v_31 = unbind_31[2] + unbind_31 = None + x_409 = torch._C._nn.scaled_dot_product_attention( + q_31, k_31, v_31, attn_mask=None, dropout_p=0.0 + ) + q_31 = k_31 = v_31 = None + transpose_32 = x_409.transpose(1, 2) + x_409 = None + x_410 = transpose_32.reshape(1, 257, 1280) + transpose_32 = None + x_411 = torch._C._nn.linear( + x_410, + l_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_bias_, + ) + x_410 = l_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_bias_ + ) = None + x_412 = torch.nn.functional.dropout(x_411, 0.0, False, False) + x_411 = None + mul_62 = x_412 * l_self_modules_blocks_modules_31_modules_ls1_parameters_gamma_ + x_412 = l_self_modules_blocks_modules_31_modules_ls1_parameters_gamma_ = None + x_413 = x_407 + mul_62 + x_407 = mul_62 = None + x_414 = torch.nn.functional.layer_norm( + x_413, + (1280,), + l_self_modules_blocks_modules_31_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_31_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_31_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_31_modules_norm2_parameters_bias_ + ) = None + x_415 = torch._C._nn.linear( + x_414, + l_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_bias_, + ) + x_414 = ( + l_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_416 = torch._C._nn.gelu(x_415, approximate="none") + x_415 = None + x_417 = torch.nn.functional.dropout(x_416, 0.0, False, False) + x_416 = None + x_418 = torch._C._nn.linear( + x_417, + l_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_bias_, + ) + x_417 = ( + l_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_419 = torch.nn.functional.dropout(x_418, 0.0, False, False) + x_418 = None + mul_63 = x_419 * l_self_modules_blocks_modules_31_modules_ls2_parameters_gamma_ + x_419 = l_self_modules_blocks_modules_31_modules_ls2_parameters_gamma_ = None + x_420 = x_413 + mul_63 + x_413 = mul_63 = None + x_421 = torch.nn.functional.layer_norm( + x_420, + (1280,), + l_self_modules_norm_parameters_weight_, + l_self_modules_norm_parameters_bias_, + 1e-06, + ) + x_420 = ( + l_self_modules_norm_parameters_weight_ + ) = l_self_modules_norm_parameters_bias_ = None + x_422 = x_421[(slice(None, None, None), 0)] + x_421 = None + x_423 = torch.nn.functional.dropout(x_422, 0.0, False, False) + x_422 = None + x_424 = torch._C._nn.linear( + x_423, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_423 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_424,) diff --git a/samples/timm/deit3_huge_patch14_224.fb_in1k/weight_meta.py b/samples/timm/deit3_huge_patch14_224.fb_in1k/weight_meta.py new file mode 100644 index 0000000000..35cfbe6790 --- /dev/null +++ b/samples/timm/deit3_huge_patch14_224.fb_in1k/weight_meta.py @@ -0,0 +1,4676 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.226 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [1280, 3, 14, 14] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.024 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 256, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_24_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_24_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_24_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_24_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_24_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_24_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_24_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_24_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_24_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_24_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_24_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_25_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_25_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_25_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_25_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_25_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_25_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_25_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_25_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_25_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_25_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_25_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_26_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_26_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_26_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_26_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_26_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_26_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_26_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_26_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_26_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_26_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_26_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_27_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_27_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_27_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_27_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_27_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_27_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_27_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_27_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_27_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_27_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_27_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_28_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_28_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_28_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_28_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_28_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_28_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_28_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_28_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_28_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_28_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_28_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_29_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_29_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_29_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_29_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_29_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_29_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_29_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_29_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_29_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_29_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_29_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_30_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_30_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_30_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_30_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_30_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_30_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_30_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_30_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_30_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_30_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_30_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_31_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_31_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3840, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_31_modules_attn_modules_qkv_parameters_bias_" + shape = [3840] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1280, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_31_modules_attn_modules_proj_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_31_modules_ls1_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_31_modules_norm2_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_31_modules_norm2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_weight_" + shape = [5120, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_31_modules_mlp_modules_fc1_parameters_bias_" + shape = [5120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_weight_" + shape = [1280, 5120] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_31_modules_mlp_modules_fc2_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_31_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_31_modules_ls2_parameters_gamma_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_weight_: + name = "L_self_modules_norm_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_bias_: + name = "L_self_modules_norm_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 1280] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/deit3_large_patch16_224.fb_in1k/graph_hash.txt b/samples/timm/deit3_large_patch16_224.fb_in1k/graph_hash.txt new file mode 100644 index 0000000000..a9553b407e --- /dev/null +++ b/samples/timm/deit3_large_patch16_224.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +7adc40d79aba5a23828e77fe6b175c6b16b27f015e71f10b3c9e91352272f0ec \ No newline at end of file diff --git a/samples/timm/deit3_large_patch16_224.fb_in1k/graph_net.json b/samples/timm/deit3_large_patch16_224.fb_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/deit3_large_patch16_224.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/deit3_large_patch16_224.fb_in1k/input_meta.py b/samples/timm/deit3_large_patch16_224.fb_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/deit3_large_patch16_224.fb_in1k/input_tensor_constraints.py b/samples/timm/deit3_large_patch16_224.fb_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/deit3_large_patch16_224.fb_in1k/model.py b/samples/timm/deit3_large_patch16_224.fb_in1k/model.py new file mode 100644 index 0000000000..8be144f1bf --- /dev/null +++ b/samples/timm/deit3_large_patch16_224.fb_in1k/model.py @@ -0,0 +1,3620 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_ + ) + l_self_modules_norm_parameters_weight_ = L_self_modules_norm_parameters_weight_ + l_self_modules_norm_parameters_bias_ = L_self_modules_norm_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + x_2 = x_1 + l_self_parameters_pos_embed_ + x_1 = l_self_parameters_pos_embed_ = None + x_3 = torch.cat([expand, x_2], dim=1) + expand = x_2 = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (1024,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 197, 3, 16, 64) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 197, 1024) + transpose_1 = None + x_8 = torch._C._nn.linear( + x_7, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + mul = x_9 * l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ + x_9 = l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ = None + x_10 = x_4 + mul + x_4 = mul = None + x_11 = torch.nn.functional.layer_norm( + x_10, + (1024,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_11 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_13 = torch._C._nn.gelu(x_12, approximate="none") + x_12 = None + x_14 = torch.nn.functional.dropout(x_13, 0.0, False, False) + x_13 = None + x_15 = torch._C._nn.linear( + x_14, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_14 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + mul_1 = x_16 * l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ + x_16 = l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ = None + x_17 = x_10 + mul_1 + x_10 = mul_1 = None + x_18 = torch.nn.functional.layer_norm( + x_17, + (1024,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_18, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_18 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 197, 3, 16, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_19 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_19.transpose(1, 2) + x_19 = None + x_20 = transpose_2.reshape(1, 197, 1024) + transpose_2 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + mul_2 = x_22 * l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ + x_22 = l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ = None + x_23 = x_17 + mul_2 + x_17 = mul_2 = None + x_24 = torch.nn.functional.layer_norm( + x_23, + (1024,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_25 = torch._C._nn.linear( + x_24, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_24 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_26 = torch._C._nn.gelu(x_25, approximate="none") + x_25 = None + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + mul_3 = x_29 * l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ + x_29 = l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ = None + x_30 = x_23 + mul_3 + x_23 = mul_3 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (1024,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 197, 3, 16, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_32 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_32.transpose(1, 2) + x_32 = None + x_33 = transpose_3.reshape(1, 197, 1024) + transpose_3 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_33 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + mul_4 = x_35 * l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ + x_35 = l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ = None + x_36 = x_30 + mul_4 + x_30 = mul_4 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (1024,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_39 = torch._C._nn.gelu(x_38, approximate="none") + x_38 = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = torch._C._nn.linear( + x_40, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_40 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_42 = torch.nn.functional.dropout(x_41, 0.0, False, False) + x_41 = None + mul_5 = x_42 * l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ + x_42 = l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ = None + x_43 = x_36 + mul_5 + x_36 = mul_5 = None + x_44 = torch.nn.functional.layer_norm( + x_43, + (1024,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_44 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 197, 3, 16, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_45 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_45.transpose(1, 2) + x_45 = None + x_46 = transpose_4.reshape(1, 197, 1024) + transpose_4 = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + mul_6 = x_48 * l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ + x_48 = l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ = None + x_49 = x_43 + mul_6 + x_43 = mul_6 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (1024,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_51 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_52 = torch._C._nn.gelu(x_51, approximate="none") + x_51 = None + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + mul_7 = x_55 * l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ + x_55 = l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ = None + x_56 = x_49 + mul_7 + x_49 = mul_7 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (1024,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 197, 3, 16, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_5.reshape(1, 197, 1024) + transpose_5 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + mul_8 = x_61 * l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ + x_61 = l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ = None + x_62 = x_56 + mul_8 + x_56 = mul_8 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (1024,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + mul_9 = x_68 * l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ + x_68 = l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ = None + x_69 = x_62 + mul_9 + x_62 = mul_9 = None + x_70 = torch.nn.functional.layer_norm( + x_69, + (1024,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_70 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 197, 3, 16, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_71 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_71.transpose(1, 2) + x_71 = None + x_72 = transpose_6.reshape(1, 197, 1024) + transpose_6 = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + mul_10 = x_74 * l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ + x_74 = l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ = None + x_75 = x_69 + mul_10 + x_69 = mul_10 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (1024,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_78 = torch._C._nn.gelu(x_77, approximate="none") + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch._C._nn.linear( + x_79, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_79 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + mul_11 = x_81 * l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ + x_81 = l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ = None + x_82 = x_75 + mul_11 + x_75 = mul_11 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (1024,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 197, 3, 16, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_84 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_84.transpose(1, 2) + x_84 = None + x_85 = transpose_7.reshape(1, 197, 1024) + transpose_7 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_85 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + mul_12 = x_87 * l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ + x_87 = l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ = None + x_88 = x_82 + mul_12 + x_82 = mul_12 = None + x_89 = torch.nn.functional.layer_norm( + x_88, + (1024,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_89 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_91 = torch._C._nn.gelu(x_90, approximate="none") + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + mul_13 = x_94 * l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ + x_94 = l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ = None + x_95 = x_88 + mul_13 + x_88 = mul_13 = None + x_96 = torch.nn.functional.layer_norm( + x_95, + (1024,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_96 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 197, 3, 16, 64) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_97 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_97.transpose(1, 2) + x_97 = None + x_98 = transpose_8.reshape(1, 197, 1024) + transpose_8 = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + mul_14 = x_100 * l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ + x_100 = l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ = None + x_101 = x_95 + mul_14 + x_95 = mul_14 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (1024,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch._C._nn.linear( + x_105, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_105 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + mul_15 = x_107 * l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ + x_107 = l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ = None + x_108 = x_101 + mul_15 + x_101 = mul_15 = None + x_109 = torch.nn.functional.layer_norm( + x_108, + (1024,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_109, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_109 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 197, 3, 16, 64) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_9.reshape(1, 197, 1024) + transpose_9 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + mul_16 = x_113 * l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ + x_113 = l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ = None + x_114 = x_108 + mul_16 + x_108 = mul_16 = None + x_115 = torch.nn.functional.layer_norm( + x_114, + (1024,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_116 = torch._C._nn.linear( + x_115, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_115 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + mul_17 = x_120 * l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ + x_120 = l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ = None + x_121 = x_114 + mul_17 + x_114 = mul_17 = None + x_122 = torch.nn.functional.layer_norm( + x_121, + (1024,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_122 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 197, 3, 16, 64) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_123 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_123.transpose(1, 2) + x_123 = None + x_124 = transpose_10.reshape(1, 197, 1024) + transpose_10 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_124 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + mul_18 = x_126 * l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ + x_126 = l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ = None + x_127 = x_121 + mul_18 + x_121 = mul_18 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (1024,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_130 = torch._C._nn.gelu(x_129, approximate="none") + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch._C._nn.linear( + x_131, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_131 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + mul_19 = x_133 * l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ + x_133 = l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ = None + x_134 = x_127 + mul_19 + x_127 = mul_19 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (1024,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_135, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_135 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 197, 3, 16, 64) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_136 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_136.transpose(1, 2) + x_136 = None + x_137 = transpose_11.reshape(1, 197, 1024) + transpose_11 = None + x_138 = torch._C._nn.linear( + x_137, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_137 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + mul_20 = x_139 * l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ + x_139 = l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ = None + x_140 = x_134 + mul_20 + x_134 = mul_20 = None + x_141 = torch.nn.functional.layer_norm( + x_140, + (1024,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_141 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_143 = torch._C._nn.gelu(x_142, approximate="none") + x_142 = None + x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) + x_143 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_144 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + mul_21 = x_146 * l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ + x_146 = l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ = None + x_147 = x_140 + mul_21 + x_140 = mul_21 = None + x_148 = torch.nn.functional.layer_norm( + x_147, + (1024,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_148, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_148 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 197, 3, 16, 64) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_149 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_149.transpose(1, 2) + x_149 = None + x_150 = transpose_12.reshape(1, 197, 1024) + transpose_12 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + mul_22 = x_152 * l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ + x_152 = l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ = None + x_153 = x_147 + mul_22 + x_147 = mul_22 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (1024,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_156 = torch._C._nn.gelu(x_155, approximate="none") + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + mul_23 = x_159 * l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ + x_159 = l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ = None + x_160 = x_153 + mul_23 + x_153 = mul_23 = None + x_161 = torch.nn.functional.layer_norm( + x_160, + (1024,), + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_12_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_norm1_parameters_bias_ + ) = None + linear_48 = torch._C._nn.linear( + x_161, + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_, + ) + x_161 = ( + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_24 = linear_48.reshape(1, 197, 3, 16, 64) + linear_48 = None + qkv_12 = reshape_24.permute(2, 0, 3, 1, 4) + reshape_24 = None + unbind_12 = qkv_12.unbind(0) + qkv_12 = None + q_12 = unbind_12[0] + k_12 = unbind_12[1] + v_12 = unbind_12[2] + unbind_12 = None + x_162 = torch._C._nn.scaled_dot_product_attention( + q_12, k_12, v_12, attn_mask=None, dropout_p=0.0 + ) + q_12 = k_12 = v_12 = None + transpose_13 = x_162.transpose(1, 2) + x_162 = None + x_163 = transpose_13.reshape(1, 197, 1024) + transpose_13 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_, + ) + x_163 = l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_ + ) = None + x_165 = torch.nn.functional.dropout(x_164, 0.0, False, False) + x_164 = None + mul_24 = x_165 * l_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_ + x_165 = l_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_ = None + x_166 = x_160 + mul_24 + x_160 = mul_24 = None + x_167 = torch.nn.functional.layer_norm( + x_166, + (1024,), + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_12_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_12_modules_norm2_parameters_bias_ + ) = None + x_168 = torch._C._nn.linear( + x_167, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_, + ) + x_167 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_169 = torch._C._nn.gelu(x_168, approximate="none") + x_168 = None + x_170 = torch.nn.functional.dropout(x_169, 0.0, False, False) + x_169 = None + x_171 = torch._C._nn.linear( + x_170, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_, + ) + x_170 = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_172 = torch.nn.functional.dropout(x_171, 0.0, False, False) + x_171 = None + mul_25 = x_172 * l_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_ + x_172 = l_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_ = None + x_173 = x_166 + mul_25 + x_166 = mul_25 = None + x_174 = torch.nn.functional.layer_norm( + x_173, + (1024,), + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_13_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_norm1_parameters_bias_ + ) = None + linear_52 = torch._C._nn.linear( + x_174, + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_, + ) + x_174 = ( + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_26 = linear_52.reshape(1, 197, 3, 16, 64) + linear_52 = None + qkv_13 = reshape_26.permute(2, 0, 3, 1, 4) + reshape_26 = None + unbind_13 = qkv_13.unbind(0) + qkv_13 = None + q_13 = unbind_13[0] + k_13 = unbind_13[1] + v_13 = unbind_13[2] + unbind_13 = None + x_175 = torch._C._nn.scaled_dot_product_attention( + q_13, k_13, v_13, attn_mask=None, dropout_p=0.0 + ) + q_13 = k_13 = v_13 = None + transpose_14 = x_175.transpose(1, 2) + x_175 = None + x_176 = transpose_14.reshape(1, 197, 1024) + transpose_14 = None + x_177 = torch._C._nn.linear( + x_176, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_, + ) + x_176 = l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_ + ) = None + x_178 = torch.nn.functional.dropout(x_177, 0.0, False, False) + x_177 = None + mul_26 = x_178 * l_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_ + x_178 = l_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_ = None + x_179 = x_173 + mul_26 + x_173 = mul_26 = None + x_180 = torch.nn.functional.layer_norm( + x_179, + (1024,), + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_13_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_13_modules_norm2_parameters_bias_ + ) = None + x_181 = torch._C._nn.linear( + x_180, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_, + ) + x_180 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_182 = torch._C._nn.gelu(x_181, approximate="none") + x_181 = None + x_183 = torch.nn.functional.dropout(x_182, 0.0, False, False) + x_182 = None + x_184 = torch._C._nn.linear( + x_183, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_, + ) + x_183 = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_185 = torch.nn.functional.dropout(x_184, 0.0, False, False) + x_184 = None + mul_27 = x_185 * l_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_ + x_185 = l_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_ = None + x_186 = x_179 + mul_27 + x_179 = mul_27 = None + x_187 = torch.nn.functional.layer_norm( + x_186, + (1024,), + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_14_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_norm1_parameters_bias_ + ) = None + linear_56 = torch._C._nn.linear( + x_187, + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_, + ) + x_187 = ( + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_28 = linear_56.reshape(1, 197, 3, 16, 64) + linear_56 = None + qkv_14 = reshape_28.permute(2, 0, 3, 1, 4) + reshape_28 = None + unbind_14 = qkv_14.unbind(0) + qkv_14 = None + q_14 = unbind_14[0] + k_14 = unbind_14[1] + v_14 = unbind_14[2] + unbind_14 = None + x_188 = torch._C._nn.scaled_dot_product_attention( + q_14, k_14, v_14, attn_mask=None, dropout_p=0.0 + ) + q_14 = k_14 = v_14 = None + transpose_15 = x_188.transpose(1, 2) + x_188 = None + x_189 = transpose_15.reshape(1, 197, 1024) + transpose_15 = None + x_190 = torch._C._nn.linear( + x_189, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_, + ) + x_189 = l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_ + ) = None + x_191 = torch.nn.functional.dropout(x_190, 0.0, False, False) + x_190 = None + mul_28 = x_191 * l_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_ + x_191 = l_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_ = None + x_192 = x_186 + mul_28 + x_186 = mul_28 = None + x_193 = torch.nn.functional.layer_norm( + x_192, + (1024,), + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_14_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_14_modules_norm2_parameters_bias_ + ) = None + x_194 = torch._C._nn.linear( + x_193, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_, + ) + x_193 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_195 = torch._C._nn.gelu(x_194, approximate="none") + x_194 = None + x_196 = torch.nn.functional.dropout(x_195, 0.0, False, False) + x_195 = None + x_197 = torch._C._nn.linear( + x_196, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_, + ) + x_196 = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_198 = torch.nn.functional.dropout(x_197, 0.0, False, False) + x_197 = None + mul_29 = x_198 * l_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_ + x_198 = l_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_ = None + x_199 = x_192 + mul_29 + x_192 = mul_29 = None + x_200 = torch.nn.functional.layer_norm( + x_199, + (1024,), + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_15_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_norm1_parameters_bias_ + ) = None + linear_60 = torch._C._nn.linear( + x_200, + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_, + ) + x_200 = ( + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_30 = linear_60.reshape(1, 197, 3, 16, 64) + linear_60 = None + qkv_15 = reshape_30.permute(2, 0, 3, 1, 4) + reshape_30 = None + unbind_15 = qkv_15.unbind(0) + qkv_15 = None + q_15 = unbind_15[0] + k_15 = unbind_15[1] + v_15 = unbind_15[2] + unbind_15 = None + x_201 = torch._C._nn.scaled_dot_product_attention( + q_15, k_15, v_15, attn_mask=None, dropout_p=0.0 + ) + q_15 = k_15 = v_15 = None + transpose_16 = x_201.transpose(1, 2) + x_201 = None + x_202 = transpose_16.reshape(1, 197, 1024) + transpose_16 = None + x_203 = torch._C._nn.linear( + x_202, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_, + ) + x_202 = l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_ + ) = None + x_204 = torch.nn.functional.dropout(x_203, 0.0, False, False) + x_203 = None + mul_30 = x_204 * l_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_ + x_204 = l_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_ = None + x_205 = x_199 + mul_30 + x_199 = mul_30 = None + x_206 = torch.nn.functional.layer_norm( + x_205, + (1024,), + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_15_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_15_modules_norm2_parameters_bias_ + ) = None + x_207 = torch._C._nn.linear( + x_206, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_, + ) + x_206 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_208 = torch._C._nn.gelu(x_207, approximate="none") + x_207 = None + x_209 = torch.nn.functional.dropout(x_208, 0.0, False, False) + x_208 = None + x_210 = torch._C._nn.linear( + x_209, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_, + ) + x_209 = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_211 = torch.nn.functional.dropout(x_210, 0.0, False, False) + x_210 = None + mul_31 = x_211 * l_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_ + x_211 = l_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_ = None + x_212 = x_205 + mul_31 + x_205 = mul_31 = None + x_213 = torch.nn.functional.layer_norm( + x_212, + (1024,), + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_16_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_norm1_parameters_bias_ + ) = None + linear_64 = torch._C._nn.linear( + x_213, + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_, + ) + x_213 = ( + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_32 = linear_64.reshape(1, 197, 3, 16, 64) + linear_64 = None + qkv_16 = reshape_32.permute(2, 0, 3, 1, 4) + reshape_32 = None + unbind_16 = qkv_16.unbind(0) + qkv_16 = None + q_16 = unbind_16[0] + k_16 = unbind_16[1] + v_16 = unbind_16[2] + unbind_16 = None + x_214 = torch._C._nn.scaled_dot_product_attention( + q_16, k_16, v_16, attn_mask=None, dropout_p=0.0 + ) + q_16 = k_16 = v_16 = None + transpose_17 = x_214.transpose(1, 2) + x_214 = None + x_215 = transpose_17.reshape(1, 197, 1024) + transpose_17 = None + x_216 = torch._C._nn.linear( + x_215, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_, + ) + x_215 = l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_ + ) = None + x_217 = torch.nn.functional.dropout(x_216, 0.0, False, False) + x_216 = None + mul_32 = x_217 * l_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_ + x_217 = l_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_ = None + x_218 = x_212 + mul_32 + x_212 = mul_32 = None + x_219 = torch.nn.functional.layer_norm( + x_218, + (1024,), + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_16_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_16_modules_norm2_parameters_bias_ + ) = None + x_220 = torch._C._nn.linear( + x_219, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_, + ) + x_219 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_221 = torch._C._nn.gelu(x_220, approximate="none") + x_220 = None + x_222 = torch.nn.functional.dropout(x_221, 0.0, False, False) + x_221 = None + x_223 = torch._C._nn.linear( + x_222, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_, + ) + x_222 = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_224 = torch.nn.functional.dropout(x_223, 0.0, False, False) + x_223 = None + mul_33 = x_224 * l_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_ + x_224 = l_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_ = None + x_225 = x_218 + mul_33 + x_218 = mul_33 = None + x_226 = torch.nn.functional.layer_norm( + x_225, + (1024,), + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_17_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_norm1_parameters_bias_ + ) = None + linear_68 = torch._C._nn.linear( + x_226, + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_, + ) + x_226 = ( + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_34 = linear_68.reshape(1, 197, 3, 16, 64) + linear_68 = None + qkv_17 = reshape_34.permute(2, 0, 3, 1, 4) + reshape_34 = None + unbind_17 = qkv_17.unbind(0) + qkv_17 = None + q_17 = unbind_17[0] + k_17 = unbind_17[1] + v_17 = unbind_17[2] + unbind_17 = None + x_227 = torch._C._nn.scaled_dot_product_attention( + q_17, k_17, v_17, attn_mask=None, dropout_p=0.0 + ) + q_17 = k_17 = v_17 = None + transpose_18 = x_227.transpose(1, 2) + x_227 = None + x_228 = transpose_18.reshape(1, 197, 1024) + transpose_18 = None + x_229 = torch._C._nn.linear( + x_228, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_, + ) + x_228 = l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_ + ) = None + x_230 = torch.nn.functional.dropout(x_229, 0.0, False, False) + x_229 = None + mul_34 = x_230 * l_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_ + x_230 = l_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_ = None + x_231 = x_225 + mul_34 + x_225 = mul_34 = None + x_232 = torch.nn.functional.layer_norm( + x_231, + (1024,), + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_17_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_17_modules_norm2_parameters_bias_ + ) = None + x_233 = torch._C._nn.linear( + x_232, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_, + ) + x_232 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_234 = torch._C._nn.gelu(x_233, approximate="none") + x_233 = None + x_235 = torch.nn.functional.dropout(x_234, 0.0, False, False) + x_234 = None + x_236 = torch._C._nn.linear( + x_235, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_, + ) + x_235 = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_237 = torch.nn.functional.dropout(x_236, 0.0, False, False) + x_236 = None + mul_35 = x_237 * l_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_ + x_237 = l_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_ = None + x_238 = x_231 + mul_35 + x_231 = mul_35 = None + x_239 = torch.nn.functional.layer_norm( + x_238, + (1024,), + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_18_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_norm1_parameters_bias_ + ) = None + linear_72 = torch._C._nn.linear( + x_239, + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_, + ) + x_239 = ( + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_36 = linear_72.reshape(1, 197, 3, 16, 64) + linear_72 = None + qkv_18 = reshape_36.permute(2, 0, 3, 1, 4) + reshape_36 = None + unbind_18 = qkv_18.unbind(0) + qkv_18 = None + q_18 = unbind_18[0] + k_18 = unbind_18[1] + v_18 = unbind_18[2] + unbind_18 = None + x_240 = torch._C._nn.scaled_dot_product_attention( + q_18, k_18, v_18, attn_mask=None, dropout_p=0.0 + ) + q_18 = k_18 = v_18 = None + transpose_19 = x_240.transpose(1, 2) + x_240 = None + x_241 = transpose_19.reshape(1, 197, 1024) + transpose_19 = None + x_242 = torch._C._nn.linear( + x_241, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_, + ) + x_241 = l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_ + ) = None + x_243 = torch.nn.functional.dropout(x_242, 0.0, False, False) + x_242 = None + mul_36 = x_243 * l_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_ + x_243 = l_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_ = None + x_244 = x_238 + mul_36 + x_238 = mul_36 = None + x_245 = torch.nn.functional.layer_norm( + x_244, + (1024,), + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_18_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_18_modules_norm2_parameters_bias_ + ) = None + x_246 = torch._C._nn.linear( + x_245, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_, + ) + x_245 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_247 = torch._C._nn.gelu(x_246, approximate="none") + x_246 = None + x_248 = torch.nn.functional.dropout(x_247, 0.0, False, False) + x_247 = None + x_249 = torch._C._nn.linear( + x_248, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_, + ) + x_248 = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_250 = torch.nn.functional.dropout(x_249, 0.0, False, False) + x_249 = None + mul_37 = x_250 * l_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_ + x_250 = l_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_ = None + x_251 = x_244 + mul_37 + x_244 = mul_37 = None + x_252 = torch.nn.functional.layer_norm( + x_251, + (1024,), + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_19_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_norm1_parameters_bias_ + ) = None + linear_76 = torch._C._nn.linear( + x_252, + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_, + ) + x_252 = ( + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_38 = linear_76.reshape(1, 197, 3, 16, 64) + linear_76 = None + qkv_19 = reshape_38.permute(2, 0, 3, 1, 4) + reshape_38 = None + unbind_19 = qkv_19.unbind(0) + qkv_19 = None + q_19 = unbind_19[0] + k_19 = unbind_19[1] + v_19 = unbind_19[2] + unbind_19 = None + x_253 = torch._C._nn.scaled_dot_product_attention( + q_19, k_19, v_19, attn_mask=None, dropout_p=0.0 + ) + q_19 = k_19 = v_19 = None + transpose_20 = x_253.transpose(1, 2) + x_253 = None + x_254 = transpose_20.reshape(1, 197, 1024) + transpose_20 = None + x_255 = torch._C._nn.linear( + x_254, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_, + ) + x_254 = l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_ + ) = None + x_256 = torch.nn.functional.dropout(x_255, 0.0, False, False) + x_255 = None + mul_38 = x_256 * l_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_ + x_256 = l_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_ = None + x_257 = x_251 + mul_38 + x_251 = mul_38 = None + x_258 = torch.nn.functional.layer_norm( + x_257, + (1024,), + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_19_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_19_modules_norm2_parameters_bias_ + ) = None + x_259 = torch._C._nn.linear( + x_258, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_, + ) + x_258 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_260 = torch._C._nn.gelu(x_259, approximate="none") + x_259 = None + x_261 = torch.nn.functional.dropout(x_260, 0.0, False, False) + x_260 = None + x_262 = torch._C._nn.linear( + x_261, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_, + ) + x_261 = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_263 = torch.nn.functional.dropout(x_262, 0.0, False, False) + x_262 = None + mul_39 = x_263 * l_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_ + x_263 = l_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_ = None + x_264 = x_257 + mul_39 + x_257 = mul_39 = None + x_265 = torch.nn.functional.layer_norm( + x_264, + (1024,), + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_20_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_norm1_parameters_bias_ + ) = None + linear_80 = torch._C._nn.linear( + x_265, + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_, + ) + x_265 = ( + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_40 = linear_80.reshape(1, 197, 3, 16, 64) + linear_80 = None + qkv_20 = reshape_40.permute(2, 0, 3, 1, 4) + reshape_40 = None + unbind_20 = qkv_20.unbind(0) + qkv_20 = None + q_20 = unbind_20[0] + k_20 = unbind_20[1] + v_20 = unbind_20[2] + unbind_20 = None + x_266 = torch._C._nn.scaled_dot_product_attention( + q_20, k_20, v_20, attn_mask=None, dropout_p=0.0 + ) + q_20 = k_20 = v_20 = None + transpose_21 = x_266.transpose(1, 2) + x_266 = None + x_267 = transpose_21.reshape(1, 197, 1024) + transpose_21 = None + x_268 = torch._C._nn.linear( + x_267, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_, + ) + x_267 = l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_ + ) = None + x_269 = torch.nn.functional.dropout(x_268, 0.0, False, False) + x_268 = None + mul_40 = x_269 * l_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_ + x_269 = l_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_ = None + x_270 = x_264 + mul_40 + x_264 = mul_40 = None + x_271 = torch.nn.functional.layer_norm( + x_270, + (1024,), + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_20_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_20_modules_norm2_parameters_bias_ + ) = None + x_272 = torch._C._nn.linear( + x_271, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_, + ) + x_271 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_273 = torch._C._nn.gelu(x_272, approximate="none") + x_272 = None + x_274 = torch.nn.functional.dropout(x_273, 0.0, False, False) + x_273 = None + x_275 = torch._C._nn.linear( + x_274, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_, + ) + x_274 = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_276 = torch.nn.functional.dropout(x_275, 0.0, False, False) + x_275 = None + mul_41 = x_276 * l_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_ + x_276 = l_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_ = None + x_277 = x_270 + mul_41 + x_270 = mul_41 = None + x_278 = torch.nn.functional.layer_norm( + x_277, + (1024,), + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_21_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_norm1_parameters_bias_ + ) = None + linear_84 = torch._C._nn.linear( + x_278, + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_, + ) + x_278 = ( + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_42 = linear_84.reshape(1, 197, 3, 16, 64) + linear_84 = None + qkv_21 = reshape_42.permute(2, 0, 3, 1, 4) + reshape_42 = None + unbind_21 = qkv_21.unbind(0) + qkv_21 = None + q_21 = unbind_21[0] + k_21 = unbind_21[1] + v_21 = unbind_21[2] + unbind_21 = None + x_279 = torch._C._nn.scaled_dot_product_attention( + q_21, k_21, v_21, attn_mask=None, dropout_p=0.0 + ) + q_21 = k_21 = v_21 = None + transpose_22 = x_279.transpose(1, 2) + x_279 = None + x_280 = transpose_22.reshape(1, 197, 1024) + transpose_22 = None + x_281 = torch._C._nn.linear( + x_280, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_, + ) + x_280 = l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_ + ) = None + x_282 = torch.nn.functional.dropout(x_281, 0.0, False, False) + x_281 = None + mul_42 = x_282 * l_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_ + x_282 = l_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_ = None + x_283 = x_277 + mul_42 + x_277 = mul_42 = None + x_284 = torch.nn.functional.layer_norm( + x_283, + (1024,), + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_21_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_21_modules_norm2_parameters_bias_ + ) = None + x_285 = torch._C._nn.linear( + x_284, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_, + ) + x_284 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_286 = torch._C._nn.gelu(x_285, approximate="none") + x_285 = None + x_287 = torch.nn.functional.dropout(x_286, 0.0, False, False) + x_286 = None + x_288 = torch._C._nn.linear( + x_287, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_, + ) + x_287 = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_289 = torch.nn.functional.dropout(x_288, 0.0, False, False) + x_288 = None + mul_43 = x_289 * l_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_ + x_289 = l_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_ = None + x_290 = x_283 + mul_43 + x_283 = mul_43 = None + x_291 = torch.nn.functional.layer_norm( + x_290, + (1024,), + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_22_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_norm1_parameters_bias_ + ) = None + linear_88 = torch._C._nn.linear( + x_291, + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_, + ) + x_291 = ( + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_44 = linear_88.reshape(1, 197, 3, 16, 64) + linear_88 = None + qkv_22 = reshape_44.permute(2, 0, 3, 1, 4) + reshape_44 = None + unbind_22 = qkv_22.unbind(0) + qkv_22 = None + q_22 = unbind_22[0] + k_22 = unbind_22[1] + v_22 = unbind_22[2] + unbind_22 = None + x_292 = torch._C._nn.scaled_dot_product_attention( + q_22, k_22, v_22, attn_mask=None, dropout_p=0.0 + ) + q_22 = k_22 = v_22 = None + transpose_23 = x_292.transpose(1, 2) + x_292 = None + x_293 = transpose_23.reshape(1, 197, 1024) + transpose_23 = None + x_294 = torch._C._nn.linear( + x_293, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_, + ) + x_293 = l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_ + ) = None + x_295 = torch.nn.functional.dropout(x_294, 0.0, False, False) + x_294 = None + mul_44 = x_295 * l_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_ + x_295 = l_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_ = None + x_296 = x_290 + mul_44 + x_290 = mul_44 = None + x_297 = torch.nn.functional.layer_norm( + x_296, + (1024,), + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_22_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_22_modules_norm2_parameters_bias_ + ) = None + x_298 = torch._C._nn.linear( + x_297, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_, + ) + x_297 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_299 = torch._C._nn.gelu(x_298, approximate="none") + x_298 = None + x_300 = torch.nn.functional.dropout(x_299, 0.0, False, False) + x_299 = None + x_301 = torch._C._nn.linear( + x_300, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_, + ) + x_300 = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_302 = torch.nn.functional.dropout(x_301, 0.0, False, False) + x_301 = None + mul_45 = x_302 * l_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_ + x_302 = l_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_ = None + x_303 = x_296 + mul_45 + x_296 = mul_45 = None + x_304 = torch.nn.functional.layer_norm( + x_303, + (1024,), + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_23_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_norm1_parameters_bias_ + ) = None + linear_92 = torch._C._nn.linear( + x_304, + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_, + ) + x_304 = ( + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_46 = linear_92.reshape(1, 197, 3, 16, 64) + linear_92 = None + qkv_23 = reshape_46.permute(2, 0, 3, 1, 4) + reshape_46 = None + unbind_23 = qkv_23.unbind(0) + qkv_23 = None + q_23 = unbind_23[0] + k_23 = unbind_23[1] + v_23 = unbind_23[2] + unbind_23 = None + x_305 = torch._C._nn.scaled_dot_product_attention( + q_23, k_23, v_23, attn_mask=None, dropout_p=0.0 + ) + q_23 = k_23 = v_23 = None + transpose_24 = x_305.transpose(1, 2) + x_305 = None + x_306 = transpose_24.reshape(1, 197, 1024) + transpose_24 = None + x_307 = torch._C._nn.linear( + x_306, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_, + ) + x_306 = l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_ + ) = None + x_308 = torch.nn.functional.dropout(x_307, 0.0, False, False) + x_307 = None + mul_46 = x_308 * l_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_ + x_308 = l_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_ = None + x_309 = x_303 + mul_46 + x_303 = mul_46 = None + x_310 = torch.nn.functional.layer_norm( + x_309, + (1024,), + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_23_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_23_modules_norm2_parameters_bias_ + ) = None + x_311 = torch._C._nn.linear( + x_310, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_, + ) + x_310 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_312 = torch._C._nn.gelu(x_311, approximate="none") + x_311 = None + x_313 = torch.nn.functional.dropout(x_312, 0.0, False, False) + x_312 = None + x_314 = torch._C._nn.linear( + x_313, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_, + ) + x_313 = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_315 = torch.nn.functional.dropout(x_314, 0.0, False, False) + x_314 = None + mul_47 = x_315 * l_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_ + x_315 = l_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_ = None + x_316 = x_309 + mul_47 + x_309 = mul_47 = None + x_317 = torch.nn.functional.layer_norm( + x_316, + (1024,), + l_self_modules_norm_parameters_weight_, + l_self_modules_norm_parameters_bias_, + 1e-06, + ) + x_316 = ( + l_self_modules_norm_parameters_weight_ + ) = l_self_modules_norm_parameters_bias_ = None + x_318 = x_317[(slice(None, None, None), 0)] + x_317 = None + x_319 = torch.nn.functional.dropout(x_318, 0.0, False, False) + x_318 = None + x_320 = torch._C._nn.linear( + x_319, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_319 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_320,) diff --git a/samples/timm/deit3_large_patch16_224.fb_in1k/weight_meta.py b/samples/timm/deit3_large_patch16_224.fb_in1k/weight_meta.py new file mode 100644 index 0000000000..5582416cec --- /dev/null +++ b/samples/timm/deit3_large_patch16_224.fb_in1k/weight_meta.py @@ -0,0 +1,3524 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.230 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [1024, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 196, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_12_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_12_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_12_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_13_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_13_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_13_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_14_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_14_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_14_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_15_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_15_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_15_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_16_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_16_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_16_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_17_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_17_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_17_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_18_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_18_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_18_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_19_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_19_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_19_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_20_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_20_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_20_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_21_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_21_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_21_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_22_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_22_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_22_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [3072, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_qkv_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_weight_" + ) + shape = [1024, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_attn_modules_proj_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_23_modules_ls1_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_norm2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_weight_" + shape = [4096, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc1_parameters_bias_" + shape = [4096] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_weight_" + shape = [1024, 4096] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_23_modules_mlp_modules_fc2_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_23_modules_ls2_parameters_gamma_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_weight_: + name = "L_self_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_bias_: + name = "L_self_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/deit3_medium_patch16_224.fb_in1k/graph_hash.txt b/samples/timm/deit3_medium_patch16_224.fb_in1k/graph_hash.txt new file mode 100644 index 0000000000..916e637e1b --- /dev/null +++ b/samples/timm/deit3_medium_patch16_224.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +c3ddd27825c23ebd4fdf6c5ca7a97ccd364dcb24ed54194e97395690eedf7055 \ No newline at end of file diff --git a/samples/timm/deit3_medium_patch16_224.fb_in1k/graph_net.json b/samples/timm/deit3_medium_patch16_224.fb_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/deit3_medium_patch16_224.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/deit3_medium_patch16_224.fb_in1k/input_meta.py b/samples/timm/deit3_medium_patch16_224.fb_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/deit3_medium_patch16_224.fb_in1k/input_tensor_constraints.py b/samples/timm/deit3_medium_patch16_224.fb_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/deit3_medium_patch16_224.fb_in1k/model.py b/samples/timm/deit3_medium_patch16_224.fb_in1k/model.py new file mode 100644 index 0000000000..0d76c56591 --- /dev/null +++ b/samples/timm/deit3_medium_patch16_224.fb_in1k/model.py @@ -0,0 +1,1868 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ + ) + l_self_modules_norm_parameters_weight_ = L_self_modules_norm_parameters_weight_ + l_self_modules_norm_parameters_bias_ = L_self_modules_norm_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + x_2 = x_1 + l_self_parameters_pos_embed_ + x_1 = l_self_parameters_pos_embed_ = None + x_3 = torch.cat([expand, x_2], dim=1) + expand = x_2 = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (512,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 197, 3, 8, 64) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 197, 512) + transpose_1 = None + x_8 = torch._C._nn.linear( + x_7, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + mul = x_9 * l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ + x_9 = l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ = None + x_10 = x_4 + mul + x_4 = mul = None + x_11 = torch.nn.functional.layer_norm( + x_10, + (512,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_11 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_13 = torch._C._nn.gelu(x_12, approximate="none") + x_12 = None + x_14 = torch.nn.functional.dropout(x_13, 0.0, False, False) + x_13 = None + x_15 = torch._C._nn.linear( + x_14, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_14 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + mul_1 = x_16 * l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ + x_16 = l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ = None + x_17 = x_10 + mul_1 + x_10 = mul_1 = None + x_18 = torch.nn.functional.layer_norm( + x_17, + (512,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_18, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_18 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 197, 3, 8, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_19 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_19.transpose(1, 2) + x_19 = None + x_20 = transpose_2.reshape(1, 197, 512) + transpose_2 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + mul_2 = x_22 * l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ + x_22 = l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ = None + x_23 = x_17 + mul_2 + x_17 = mul_2 = None + x_24 = torch.nn.functional.layer_norm( + x_23, + (512,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_25 = torch._C._nn.linear( + x_24, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_24 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_26 = torch._C._nn.gelu(x_25, approximate="none") + x_25 = None + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + mul_3 = x_29 * l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ + x_29 = l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ = None + x_30 = x_23 + mul_3 + x_23 = mul_3 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (512,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 197, 3, 8, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_32 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_32.transpose(1, 2) + x_32 = None + x_33 = transpose_3.reshape(1, 197, 512) + transpose_3 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_33 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + mul_4 = x_35 * l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ + x_35 = l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ = None + x_36 = x_30 + mul_4 + x_30 = mul_4 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (512,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_39 = torch._C._nn.gelu(x_38, approximate="none") + x_38 = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = torch._C._nn.linear( + x_40, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_40 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_42 = torch.nn.functional.dropout(x_41, 0.0, False, False) + x_41 = None + mul_5 = x_42 * l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ + x_42 = l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ = None + x_43 = x_36 + mul_5 + x_36 = mul_5 = None + x_44 = torch.nn.functional.layer_norm( + x_43, + (512,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_44 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 197, 3, 8, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_45 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_45.transpose(1, 2) + x_45 = None + x_46 = transpose_4.reshape(1, 197, 512) + transpose_4 = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + mul_6 = x_48 * l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ + x_48 = l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ = None + x_49 = x_43 + mul_6 + x_43 = mul_6 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (512,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_51 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_52 = torch._C._nn.gelu(x_51, approximate="none") + x_51 = None + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + mul_7 = x_55 * l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ + x_55 = l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ = None + x_56 = x_49 + mul_7 + x_49 = mul_7 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (512,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 197, 3, 8, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_5.reshape(1, 197, 512) + transpose_5 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + mul_8 = x_61 * l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ + x_61 = l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ = None + x_62 = x_56 + mul_8 + x_56 = mul_8 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (512,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + mul_9 = x_68 * l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ + x_68 = l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ = None + x_69 = x_62 + mul_9 + x_62 = mul_9 = None + x_70 = torch.nn.functional.layer_norm( + x_69, + (512,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_70 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 197, 3, 8, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_71 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_71.transpose(1, 2) + x_71 = None + x_72 = transpose_6.reshape(1, 197, 512) + transpose_6 = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + mul_10 = x_74 * l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ + x_74 = l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ = None + x_75 = x_69 + mul_10 + x_69 = mul_10 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (512,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_78 = torch._C._nn.gelu(x_77, approximate="none") + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch._C._nn.linear( + x_79, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_79 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + mul_11 = x_81 * l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ + x_81 = l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ = None + x_82 = x_75 + mul_11 + x_75 = mul_11 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (512,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 197, 3, 8, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_84 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_84.transpose(1, 2) + x_84 = None + x_85 = transpose_7.reshape(1, 197, 512) + transpose_7 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_85 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + mul_12 = x_87 * l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ + x_87 = l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ = None + x_88 = x_82 + mul_12 + x_82 = mul_12 = None + x_89 = torch.nn.functional.layer_norm( + x_88, + (512,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_89 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_91 = torch._C._nn.gelu(x_90, approximate="none") + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + mul_13 = x_94 * l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ + x_94 = l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ = None + x_95 = x_88 + mul_13 + x_88 = mul_13 = None + x_96 = torch.nn.functional.layer_norm( + x_95, + (512,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_96 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 197, 3, 8, 64) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_97 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_97.transpose(1, 2) + x_97 = None + x_98 = transpose_8.reshape(1, 197, 512) + transpose_8 = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + mul_14 = x_100 * l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ + x_100 = l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ = None + x_101 = x_95 + mul_14 + x_95 = mul_14 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (512,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch._C._nn.linear( + x_105, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_105 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + mul_15 = x_107 * l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ + x_107 = l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ = None + x_108 = x_101 + mul_15 + x_101 = mul_15 = None + x_109 = torch.nn.functional.layer_norm( + x_108, + (512,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_109, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_109 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 197, 3, 8, 64) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_9.reshape(1, 197, 512) + transpose_9 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + mul_16 = x_113 * l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ + x_113 = l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ = None + x_114 = x_108 + mul_16 + x_108 = mul_16 = None + x_115 = torch.nn.functional.layer_norm( + x_114, + (512,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_116 = torch._C._nn.linear( + x_115, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_115 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + mul_17 = x_120 * l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ + x_120 = l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ = None + x_121 = x_114 + mul_17 + x_114 = mul_17 = None + x_122 = torch.nn.functional.layer_norm( + x_121, + (512,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_122 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 197, 3, 8, 64) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_123 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_123.transpose(1, 2) + x_123 = None + x_124 = transpose_10.reshape(1, 197, 512) + transpose_10 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_124 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + mul_18 = x_126 * l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ + x_126 = l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ = None + x_127 = x_121 + mul_18 + x_121 = mul_18 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (512,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_130 = torch._C._nn.gelu(x_129, approximate="none") + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch._C._nn.linear( + x_131, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_131 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + mul_19 = x_133 * l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ + x_133 = l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ = None + x_134 = x_127 + mul_19 + x_127 = mul_19 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (512,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_135, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_135 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 197, 3, 8, 64) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_136 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_136.transpose(1, 2) + x_136 = None + x_137 = transpose_11.reshape(1, 197, 512) + transpose_11 = None + x_138 = torch._C._nn.linear( + x_137, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_137 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + mul_20 = x_139 * l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ + x_139 = l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ = None + x_140 = x_134 + mul_20 + x_134 = mul_20 = None + x_141 = torch.nn.functional.layer_norm( + x_140, + (512,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_141 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_143 = torch._C._nn.gelu(x_142, approximate="none") + x_142 = None + x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) + x_143 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_144 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + mul_21 = x_146 * l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ + x_146 = l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ = None + x_147 = x_140 + mul_21 + x_140 = mul_21 = None + x_148 = torch.nn.functional.layer_norm( + x_147, + (512,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_148, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_148 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 197, 3, 8, 64) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_149 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_149.transpose(1, 2) + x_149 = None + x_150 = transpose_12.reshape(1, 197, 512) + transpose_12 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + mul_22 = x_152 * l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ + x_152 = l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ = None + x_153 = x_147 + mul_22 + x_147 = mul_22 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (512,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_156 = torch._C._nn.gelu(x_155, approximate="none") + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + mul_23 = x_159 * l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ + x_159 = l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ = None + x_160 = x_153 + mul_23 + x_153 = mul_23 = None + x_161 = torch.nn.functional.layer_norm( + x_160, + (512,), + l_self_modules_norm_parameters_weight_, + l_self_modules_norm_parameters_bias_, + 1e-06, + ) + x_160 = ( + l_self_modules_norm_parameters_weight_ + ) = l_self_modules_norm_parameters_bias_ = None + x_162 = x_161[(slice(None, None, None), 0)] + x_161 = None + x_163 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_163 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_164,) diff --git a/samples/timm/deit3_medium_patch16_224.fb_in1k/weight_meta.py b/samples/timm/deit3_medium_patch16_224.fb_in1k/weight_meta.py new file mode 100644 index 0000000000..d43437c234 --- /dev/null +++ b/samples/timm/deit3_medium_patch16_224.fb_in1k/weight_meta.py @@ -0,0 +1,1796 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.226 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [512, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 196, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1536, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [512, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [2048, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [2048] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [512, 2048] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_weight_: + name = "L_self_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_bias_: + name = "L_self_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 512] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/deit3_small_patch16_224.fb_in1k/graph_hash.txt b/samples/timm/deit3_small_patch16_224.fb_in1k/graph_hash.txt new file mode 100644 index 0000000000..36014e9579 --- /dev/null +++ b/samples/timm/deit3_small_patch16_224.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +528fbf8dda6333381aba8de97f9d882159ac642553a4ac9d788244a60b2425d9 \ No newline at end of file diff --git a/samples/timm/deit3_small_patch16_224.fb_in1k/graph_net.json b/samples/timm/deit3_small_patch16_224.fb_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/deit3_small_patch16_224.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/deit3_small_patch16_224.fb_in1k/input_meta.py b/samples/timm/deit3_small_patch16_224.fb_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/deit3_small_patch16_224.fb_in1k/input_tensor_constraints.py b/samples/timm/deit3_small_patch16_224.fb_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/deit3_small_patch16_224.fb_in1k/model.py b/samples/timm/deit3_small_patch16_224.fb_in1k/model.py new file mode 100644 index 0000000000..ea4b9f7d23 --- /dev/null +++ b/samples/timm/deit3_small_patch16_224.fb_in1k/model.py @@ -0,0 +1,1868 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ = ( + L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ = ( + L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ + ) + l_self_modules_norm_parameters_weight_ = L_self_modules_norm_parameters_weight_ + l_self_modules_norm_parameters_bias_ = L_self_modules_norm_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + x_2 = x_1 + l_self_parameters_pos_embed_ + x_1 = l_self_parameters_pos_embed_ = None + x_3 = torch.cat([expand, x_2], dim=1) + expand = x_2 = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (384,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 197, 3, 6, 64) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 197, 384) + transpose_1 = None + x_8 = torch._C._nn.linear( + x_7, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + mul = x_9 * l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ + x_9 = l_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_ = None + x_10 = x_4 + mul + x_4 = mul = None + x_11 = torch.nn.functional.layer_norm( + x_10, + (384,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_11 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_13 = torch._C._nn.gelu(x_12, approximate="none") + x_12 = None + x_14 = torch.nn.functional.dropout(x_13, 0.0, False, False) + x_13 = None + x_15 = torch._C._nn.linear( + x_14, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_14 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + mul_1 = x_16 * l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ + x_16 = l_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_ = None + x_17 = x_10 + mul_1 + x_10 = mul_1 = None + x_18 = torch.nn.functional.layer_norm( + x_17, + (384,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_18, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_18 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 197, 3, 6, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_19 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_19.transpose(1, 2) + x_19 = None + x_20 = transpose_2.reshape(1, 197, 384) + transpose_2 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + mul_2 = x_22 * l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ + x_22 = l_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_ = None + x_23 = x_17 + mul_2 + x_17 = mul_2 = None + x_24 = torch.nn.functional.layer_norm( + x_23, + (384,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_25 = torch._C._nn.linear( + x_24, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_24 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_26 = torch._C._nn.gelu(x_25, approximate="none") + x_25 = None + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + mul_3 = x_29 * l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ + x_29 = l_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_ = None + x_30 = x_23 + mul_3 + x_23 = mul_3 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (384,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 197, 3, 6, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_32 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_32.transpose(1, 2) + x_32 = None + x_33 = transpose_3.reshape(1, 197, 384) + transpose_3 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_33 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + mul_4 = x_35 * l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ + x_35 = l_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_ = None + x_36 = x_30 + mul_4 + x_30 = mul_4 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (384,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_39 = torch._C._nn.gelu(x_38, approximate="none") + x_38 = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = torch._C._nn.linear( + x_40, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_40 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_42 = torch.nn.functional.dropout(x_41, 0.0, False, False) + x_41 = None + mul_5 = x_42 * l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ + x_42 = l_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_ = None + x_43 = x_36 + mul_5 + x_36 = mul_5 = None + x_44 = torch.nn.functional.layer_norm( + x_43, + (384,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_44 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 197, 3, 6, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_45 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_45.transpose(1, 2) + x_45 = None + x_46 = transpose_4.reshape(1, 197, 384) + transpose_4 = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + mul_6 = x_48 * l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ + x_48 = l_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_ = None + x_49 = x_43 + mul_6 + x_43 = mul_6 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (384,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_51 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_52 = torch._C._nn.gelu(x_51, approximate="none") + x_51 = None + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + mul_7 = x_55 * l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ + x_55 = l_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_ = None + x_56 = x_49 + mul_7 + x_49 = mul_7 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (384,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 197, 3, 6, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_5.reshape(1, 197, 384) + transpose_5 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + mul_8 = x_61 * l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ + x_61 = l_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_ = None + x_62 = x_56 + mul_8 + x_56 = mul_8 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (384,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + mul_9 = x_68 * l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ + x_68 = l_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_ = None + x_69 = x_62 + mul_9 + x_62 = mul_9 = None + x_70 = torch.nn.functional.layer_norm( + x_69, + (384,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_70 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 197, 3, 6, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_71 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_71.transpose(1, 2) + x_71 = None + x_72 = transpose_6.reshape(1, 197, 384) + transpose_6 = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + mul_10 = x_74 * l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ + x_74 = l_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_ = None + x_75 = x_69 + mul_10 + x_69 = mul_10 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (384,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_78 = torch._C._nn.gelu(x_77, approximate="none") + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch._C._nn.linear( + x_79, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_79 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + mul_11 = x_81 * l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ + x_81 = l_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_ = None + x_82 = x_75 + mul_11 + x_75 = mul_11 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (384,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 197, 3, 6, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_84 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_84.transpose(1, 2) + x_84 = None + x_85 = transpose_7.reshape(1, 197, 384) + transpose_7 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_85 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + mul_12 = x_87 * l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ + x_87 = l_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_ = None + x_88 = x_82 + mul_12 + x_82 = mul_12 = None + x_89 = torch.nn.functional.layer_norm( + x_88, + (384,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_89 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_91 = torch._C._nn.gelu(x_90, approximate="none") + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + mul_13 = x_94 * l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ + x_94 = l_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_ = None + x_95 = x_88 + mul_13 + x_88 = mul_13 = None + x_96 = torch.nn.functional.layer_norm( + x_95, + (384,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_96 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 197, 3, 6, 64) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_97 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_97.transpose(1, 2) + x_97 = None + x_98 = transpose_8.reshape(1, 197, 384) + transpose_8 = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + mul_14 = x_100 * l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ + x_100 = l_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_ = None + x_101 = x_95 + mul_14 + x_95 = mul_14 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (384,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch._C._nn.linear( + x_105, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_105 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + mul_15 = x_107 * l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ + x_107 = l_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_ = None + x_108 = x_101 + mul_15 + x_101 = mul_15 = None + x_109 = torch.nn.functional.layer_norm( + x_108, + (384,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_109, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_109 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 197, 3, 6, 64) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_9.reshape(1, 197, 384) + transpose_9 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + mul_16 = x_113 * l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ + x_113 = l_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_ = None + x_114 = x_108 + mul_16 + x_108 = mul_16 = None + x_115 = torch.nn.functional.layer_norm( + x_114, + (384,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_116 = torch._C._nn.linear( + x_115, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_115 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + mul_17 = x_120 * l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ + x_120 = l_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_ = None + x_121 = x_114 + mul_17 + x_114 = mul_17 = None + x_122 = torch.nn.functional.layer_norm( + x_121, + (384,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_122 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 197, 3, 6, 64) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_123 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_123.transpose(1, 2) + x_123 = None + x_124 = transpose_10.reshape(1, 197, 384) + transpose_10 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_124 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + mul_18 = x_126 * l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ + x_126 = l_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_ = None + x_127 = x_121 + mul_18 + x_121 = mul_18 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (384,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_130 = torch._C._nn.gelu(x_129, approximate="none") + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch._C._nn.linear( + x_131, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_131 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + mul_19 = x_133 * l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ + x_133 = l_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_ = None + x_134 = x_127 + mul_19 + x_127 = mul_19 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (384,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_135, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_135 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 197, 3, 6, 64) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_136 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_136.transpose(1, 2) + x_136 = None + x_137 = transpose_11.reshape(1, 197, 384) + transpose_11 = None + x_138 = torch._C._nn.linear( + x_137, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_137 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + mul_20 = x_139 * l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ + x_139 = l_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_ = None + x_140 = x_134 + mul_20 + x_134 = mul_20 = None + x_141 = torch.nn.functional.layer_norm( + x_140, + (384,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_141 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_143 = torch._C._nn.gelu(x_142, approximate="none") + x_142 = None + x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) + x_143 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_144 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + mul_21 = x_146 * l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ + x_146 = l_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_ = None + x_147 = x_140 + mul_21 + x_140 = mul_21 = None + x_148 = torch.nn.functional.layer_norm( + x_147, + (384,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_148, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_148 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 197, 3, 6, 64) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_149 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_149.transpose(1, 2) + x_149 = None + x_150 = transpose_12.reshape(1, 197, 384) + transpose_12 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + mul_22 = x_152 * l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ + x_152 = l_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_ = None + x_153 = x_147 + mul_22 + x_147 = mul_22 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (384,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_156 = torch._C._nn.gelu(x_155, approximate="none") + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + mul_23 = x_159 * l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ + x_159 = l_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_ = None + x_160 = x_153 + mul_23 + x_153 = mul_23 = None + x_161 = torch.nn.functional.layer_norm( + x_160, + (384,), + l_self_modules_norm_parameters_weight_, + l_self_modules_norm_parameters_bias_, + 1e-06, + ) + x_160 = ( + l_self_modules_norm_parameters_weight_ + ) = l_self_modules_norm_parameters_bias_ = None + x_162 = x_161[(slice(None, None, None), 0)] + x_161 = None + x_163 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_163 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_164,) diff --git a/samples/timm/deit3_small_patch16_224.fb_in1k/weight_meta.py b/samples/timm/deit3_small_patch16_224.fb_in1k/weight_meta.py new file mode 100644 index 0000000000..7cab472b3c --- /dev/null +++ b/samples/timm/deit3_small_patch16_224.fb_in1k/weight_meta.py @@ -0,0 +1,1796 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.225 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [384, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 196, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_0_modules_ls1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_0_modules_ls2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_1_modules_ls1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_1_modules_ls2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_2_modules_ls1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_2_modules_ls2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_3_modules_ls1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_3_modules_ls2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_4_modules_ls1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_4_modules_ls2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_5_modules_ls1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_5_modules_ls2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_6_modules_ls1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_6_modules_ls2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_7_modules_ls1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_7_modules_ls2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_8_modules_ls1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_8_modules_ls2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_9_modules_ls1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_9_modules_ls2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_10_modules_ls1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_10_modules_ls2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_: + name = "L_self_modules_blocks_modules_11_modules_ls1_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_: + name = "L_self_modules_blocks_modules_11_modules_ls2_parameters_gamma_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_weight_: + name = "L_self_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_bias_: + name = "L_self_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/deit_base_distilled_patch16_224.fb_in1k/graph_hash.txt b/samples/timm/deit_base_distilled_patch16_224.fb_in1k/graph_hash.txt new file mode 100644 index 0000000000..a70a97d2db --- /dev/null +++ b/samples/timm/deit_base_distilled_patch16_224.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +c83bcf76f42e39c7be55f30e59ef7cef3a5854c9abd5fd91ef151a609a558094 \ No newline at end of file diff --git a/samples/timm/deit_base_distilled_patch16_224.fb_in1k/graph_net.json b/samples/timm/deit_base_distilled_patch16_224.fb_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/deit_base_distilled_patch16_224.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/deit_base_distilled_patch16_224.fb_in1k/input_meta.py b/samples/timm/deit_base_distilled_patch16_224.fb_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/deit_base_distilled_patch16_224.fb_in1k/input_tensor_constraints.py b/samples/timm/deit_base_distilled_patch16_224.fb_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/deit_base_distilled_patch16_224.fb_in1k/model.py b/samples/timm/deit_base_distilled_patch16_224.fb_in1k/model.py new file mode 100644 index 0000000000..878dfe9b36 --- /dev/null +++ b/samples/timm/deit_base_distilled_patch16_224.fb_in1k/model.py @@ -0,0 +1,1747 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_parameters_dist_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_dist_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_dist_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_parameters_dist_token_ = L_self_parameters_dist_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_norm_parameters_weight_ = L_self_modules_norm_parameters_weight_ + l_self_modules_norm_parameters_bias_ = L_self_modules_norm_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + l_self_modules_head_dist_parameters_weight_ = ( + L_self_modules_head_dist_parameters_weight_ + ) + l_self_modules_head_dist_parameters_bias_ = ( + L_self_modules_head_dist_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + expand_1 = l_self_parameters_dist_token_.expand(1, -1, -1) + l_self_parameters_dist_token_ = None + x_2 = torch.cat((expand, expand_1, x_1), dim=1) + expand = expand_1 = x_1 = None + x_3 = x_2 + l_self_parameters_pos_embed_ + x_2 = l_self_parameters_pos_embed_ = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (768,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 198, 3, 12, 64) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 198, 768) + transpose_1 = None + x_8 = torch._C._nn.linear( + x_7, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = x_4 + x_9 + x_4 = x_9 = None + x_11 = torch.nn.functional.layer_norm( + x_10, + (768,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_11 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_13 = torch._C._nn.gelu(x_12, approximate="none") + x_12 = None + x_14 = torch.nn.functional.dropout(x_13, 0.0, False, False) + x_13 = None + x_15 = torch._C._nn.linear( + x_14, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_14 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + x_17 = x_10 + x_16 + x_10 = x_16 = None + x_18 = torch.nn.functional.layer_norm( + x_17, + (768,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_18, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_18 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 198, 3, 12, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_19 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_19.transpose(1, 2) + x_19 = None + x_20 = transpose_2.reshape(1, 198, 768) + transpose_2 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_17 + x_22 + x_17 = x_22 = None + x_24 = torch.nn.functional.layer_norm( + x_23, + (768,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_25 = torch._C._nn.linear( + x_24, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_24 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_26 = torch._C._nn.gelu(x_25, approximate="none") + x_25 = None + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + x_30 = x_23 + x_29 + x_23 = x_29 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (768,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 198, 3, 12, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_32 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_32.transpose(1, 2) + x_32 = None + x_33 = transpose_3.reshape(1, 198, 768) + transpose_3 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_33 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + x_36 = x_30 + x_35 + x_30 = x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (768,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_39 = torch._C._nn.gelu(x_38, approximate="none") + x_38 = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = torch._C._nn.linear( + x_40, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_40 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_42 = torch.nn.functional.dropout(x_41, 0.0, False, False) + x_41 = None + x_43 = x_36 + x_42 + x_36 = x_42 = None + x_44 = torch.nn.functional.layer_norm( + x_43, + (768,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_44 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 198, 3, 12, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_45 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_45.transpose(1, 2) + x_45 = None + x_46 = transpose_4.reshape(1, 198, 768) + transpose_4 = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_43 + x_48 + x_43 = x_48 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (768,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_51 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_52 = torch._C._nn.gelu(x_51, approximate="none") + x_51 = None + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = x_49 + x_55 + x_49 = x_55 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (768,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 198, 3, 12, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_5.reshape(1, 198, 768) + transpose_5 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + x_62 = x_56 + x_61 + x_56 = x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (768,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_62 + x_68 + x_62 = x_68 = None + x_70 = torch.nn.functional.layer_norm( + x_69, + (768,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_70 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 198, 3, 12, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_71 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_71.transpose(1, 2) + x_71 = None + x_72 = transpose_6.reshape(1, 198, 768) + transpose_6 = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + x_75 = x_69 + x_74 + x_69 = x_74 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (768,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_78 = torch._C._nn.gelu(x_77, approximate="none") + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch._C._nn.linear( + x_79, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_79 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = x_75 + x_81 + x_75 = x_81 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (768,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 198, 3, 12, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_84 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_84.transpose(1, 2) + x_84 = None + x_85 = transpose_7.reshape(1, 198, 768) + transpose_7 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_85 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + x_88 = x_82 + x_87 + x_82 = x_87 = None + x_89 = torch.nn.functional.layer_norm( + x_88, + (768,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_89 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_91 = torch._C._nn.gelu(x_90, approximate="none") + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_88 + x_94 + x_88 = x_94 = None + x_96 = torch.nn.functional.layer_norm( + x_95, + (768,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_96 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 198, 3, 12, 64) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_97 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_97.transpose(1, 2) + x_97 = None + x_98 = transpose_8.reshape(1, 198, 768) + transpose_8 = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + x_101 = x_95 + x_100 + x_95 = x_100 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (768,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch._C._nn.linear( + x_105, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_105 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + x_108 = x_101 + x_107 + x_101 = x_107 = None + x_109 = torch.nn.functional.layer_norm( + x_108, + (768,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_109, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_109 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 198, 3, 12, 64) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_9.reshape(1, 198, 768) + transpose_9 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_108 + x_113 + x_108 = x_113 = None + x_115 = torch.nn.functional.layer_norm( + x_114, + (768,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_116 = torch._C._nn.linear( + x_115, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_115 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + x_121 = x_114 + x_120 + x_114 = x_120 = None + x_122 = torch.nn.functional.layer_norm( + x_121, + (768,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_122 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 198, 3, 12, 64) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_123 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_123.transpose(1, 2) + x_123 = None + x_124 = transpose_10.reshape(1, 198, 768) + transpose_10 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_124 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_121 + x_126 + x_121 = x_126 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (768,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_130 = torch._C._nn.gelu(x_129, approximate="none") + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch._C._nn.linear( + x_131, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_131 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + x_134 = x_127 + x_133 + x_127 = x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (768,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_135, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_135 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 198, 3, 12, 64) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_136 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_136.transpose(1, 2) + x_136 = None + x_137 = transpose_11.reshape(1, 198, 768) + transpose_11 = None + x_138 = torch._C._nn.linear( + x_137, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_137 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = x_134 + x_139 + x_134 = x_139 = None + x_141 = torch.nn.functional.layer_norm( + x_140, + (768,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_141 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_143 = torch._C._nn.gelu(x_142, approximate="none") + x_142 = None + x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) + x_143 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_144 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = x_140 + x_146 + x_140 = x_146 = None + x_148 = torch.nn.functional.layer_norm( + x_147, + (768,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_148, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_148 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 198, 3, 12, 64) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_149 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_149.transpose(1, 2) + x_149 = None + x_150 = transpose_12.reshape(1, 198, 768) + transpose_12 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + x_153 = x_147 + x_152 + x_147 = x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (768,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_156 = torch._C._nn.gelu(x_155, approximate="none") + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_153 + x_159 + x_153 = x_159 = None + x_161 = torch.nn.functional.layer_norm( + x_160, + (768,), + l_self_modules_norm_parameters_weight_, + l_self_modules_norm_parameters_bias_, + 1e-06, + ) + x_160 = ( + l_self_modules_norm_parameters_weight_ + ) = l_self_modules_norm_parameters_bias_ = None + x_162 = x_161[(slice(None, None, None), 0)] + x_dist = x_161[(slice(None, None, None), 1)] + x_161 = None + x_163 = torch._C._nn.linear( + x_162, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_162 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + x_dist_1 = torch._C._nn.linear( + x_dist, + l_self_modules_head_dist_parameters_weight_, + l_self_modules_head_dist_parameters_bias_, + ) + x_dist = ( + l_self_modules_head_dist_parameters_weight_ + ) = l_self_modules_head_dist_parameters_bias_ = None + add_25 = x_163 + x_dist_1 + x_163 = x_dist_1 = None + x_164 = add_25 / 2 + add_25 = None + return (x_164,) diff --git a/samples/timm/deit_base_distilled_patch16_224.fb_in1k/weight_meta.py b/samples/timm/deit_base_distilled_patch16_224.fb_in1k/weight_meta.py new file mode 100644 index 0000000000..1ff0dd1d22 --- /dev/null +++ b/samples/timm/deit_base_distilled_patch16_224.fb_in1k/weight_meta.py @@ -0,0 +1,1586 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.224 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [768, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 198, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_dist_token_: + name = "L_self_parameters_dist_token_" + shape = [1, 1, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_weight_: + name = "L_self_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_bias_: + name = "L_self_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_dist_parameters_weight_: + name = "L_self_modules_head_dist_parameters_weight_" + shape = [1000, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_dist_parameters_bias_: + name = "L_self_modules_head_dist_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/deit_base_patch16_224.fb_in1k/graph_hash.txt b/samples/timm/deit_base_patch16_224.fb_in1k/graph_hash.txt new file mode 100644 index 0000000000..40fb26dcda --- /dev/null +++ b/samples/timm/deit_base_patch16_224.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +1d0ef300e50462f683e3b88ea3674a392cb6ecc17bc7dd39f8e2c65367e829a5 \ No newline at end of file diff --git a/samples/timm/deit_base_patch16_224.fb_in1k/graph_net.json b/samples/timm/deit_base_patch16_224.fb_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/deit_base_patch16_224.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/deit_base_patch16_224.fb_in1k/input_meta.py b/samples/timm/deit_base_patch16_224.fb_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/deit_base_patch16_224.fb_in1k/input_tensor_constraints.py b/samples/timm/deit_base_patch16_224.fb_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/deit_base_patch16_224.fb_in1k/model.py b/samples/timm/deit_base_patch16_224.fb_in1k/model.py new file mode 100644 index 0000000000..49c6c7e90d --- /dev/null +++ b/samples/timm/deit_base_patch16_224.fb_in1k/model.py @@ -0,0 +1,1724 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_norm_parameters_weight_ = L_self_modules_norm_parameters_weight_ + l_self_modules_norm_parameters_bias_ = L_self_modules_norm_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + x_2 = torch.cat([expand, x_1], dim=1) + expand = x_1 = None + x_3 = x_2 + l_self_parameters_pos_embed_ + x_2 = l_self_parameters_pos_embed_ = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (768,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 197, 3, 12, 64) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 197, 768) + transpose_1 = None + x_8 = torch._C._nn.linear( + x_7, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = x_4 + x_9 + x_4 = x_9 = None + x_11 = torch.nn.functional.layer_norm( + x_10, + (768,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_11 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_13 = torch._C._nn.gelu(x_12, approximate="none") + x_12 = None + x_14 = torch.nn.functional.dropout(x_13, 0.0, False, False) + x_13 = None + x_15 = torch._C._nn.linear( + x_14, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_14 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + x_17 = x_10 + x_16 + x_10 = x_16 = None + x_18 = torch.nn.functional.layer_norm( + x_17, + (768,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_18, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_18 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 197, 3, 12, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_19 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_19.transpose(1, 2) + x_19 = None + x_20 = transpose_2.reshape(1, 197, 768) + transpose_2 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_17 + x_22 + x_17 = x_22 = None + x_24 = torch.nn.functional.layer_norm( + x_23, + (768,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_25 = torch._C._nn.linear( + x_24, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_24 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_26 = torch._C._nn.gelu(x_25, approximate="none") + x_25 = None + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + x_30 = x_23 + x_29 + x_23 = x_29 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (768,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 197, 3, 12, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_32 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_32.transpose(1, 2) + x_32 = None + x_33 = transpose_3.reshape(1, 197, 768) + transpose_3 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_33 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + x_36 = x_30 + x_35 + x_30 = x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (768,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_39 = torch._C._nn.gelu(x_38, approximate="none") + x_38 = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = torch._C._nn.linear( + x_40, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_40 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_42 = torch.nn.functional.dropout(x_41, 0.0, False, False) + x_41 = None + x_43 = x_36 + x_42 + x_36 = x_42 = None + x_44 = torch.nn.functional.layer_norm( + x_43, + (768,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_44 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 197, 3, 12, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_45 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_45.transpose(1, 2) + x_45 = None + x_46 = transpose_4.reshape(1, 197, 768) + transpose_4 = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_43 + x_48 + x_43 = x_48 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (768,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_51 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_52 = torch._C._nn.gelu(x_51, approximate="none") + x_51 = None + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = x_49 + x_55 + x_49 = x_55 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (768,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 197, 3, 12, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_5.reshape(1, 197, 768) + transpose_5 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + x_62 = x_56 + x_61 + x_56 = x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (768,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_62 + x_68 + x_62 = x_68 = None + x_70 = torch.nn.functional.layer_norm( + x_69, + (768,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_70 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 197, 3, 12, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_71 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_71.transpose(1, 2) + x_71 = None + x_72 = transpose_6.reshape(1, 197, 768) + transpose_6 = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + x_75 = x_69 + x_74 + x_69 = x_74 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (768,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_78 = torch._C._nn.gelu(x_77, approximate="none") + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch._C._nn.linear( + x_79, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_79 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = x_75 + x_81 + x_75 = x_81 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (768,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 197, 3, 12, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_84 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_84.transpose(1, 2) + x_84 = None + x_85 = transpose_7.reshape(1, 197, 768) + transpose_7 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_85 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + x_88 = x_82 + x_87 + x_82 = x_87 = None + x_89 = torch.nn.functional.layer_norm( + x_88, + (768,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_89 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_91 = torch._C._nn.gelu(x_90, approximate="none") + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_88 + x_94 + x_88 = x_94 = None + x_96 = torch.nn.functional.layer_norm( + x_95, + (768,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_96 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 197, 3, 12, 64) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_97 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_97.transpose(1, 2) + x_97 = None + x_98 = transpose_8.reshape(1, 197, 768) + transpose_8 = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + x_101 = x_95 + x_100 + x_95 = x_100 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (768,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch._C._nn.linear( + x_105, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_105 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + x_108 = x_101 + x_107 + x_101 = x_107 = None + x_109 = torch.nn.functional.layer_norm( + x_108, + (768,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_109, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_109 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 197, 3, 12, 64) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_9.reshape(1, 197, 768) + transpose_9 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_108 + x_113 + x_108 = x_113 = None + x_115 = torch.nn.functional.layer_norm( + x_114, + (768,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_116 = torch._C._nn.linear( + x_115, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_115 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + x_121 = x_114 + x_120 + x_114 = x_120 = None + x_122 = torch.nn.functional.layer_norm( + x_121, + (768,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_122 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 197, 3, 12, 64) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_123 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_123.transpose(1, 2) + x_123 = None + x_124 = transpose_10.reshape(1, 197, 768) + transpose_10 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_124 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_121 + x_126 + x_121 = x_126 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (768,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_130 = torch._C._nn.gelu(x_129, approximate="none") + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch._C._nn.linear( + x_131, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_131 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + x_134 = x_127 + x_133 + x_127 = x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (768,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_135, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_135 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 197, 3, 12, 64) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_136 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_136.transpose(1, 2) + x_136 = None + x_137 = transpose_11.reshape(1, 197, 768) + transpose_11 = None + x_138 = torch._C._nn.linear( + x_137, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_137 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = x_134 + x_139 + x_134 = x_139 = None + x_141 = torch.nn.functional.layer_norm( + x_140, + (768,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_141 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_143 = torch._C._nn.gelu(x_142, approximate="none") + x_142 = None + x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) + x_143 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_144 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = x_140 + x_146 + x_140 = x_146 = None + x_148 = torch.nn.functional.layer_norm( + x_147, + (768,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_148, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_148 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 197, 3, 12, 64) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_149 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_149.transpose(1, 2) + x_149 = None + x_150 = transpose_12.reshape(1, 197, 768) + transpose_12 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + x_153 = x_147 + x_152 + x_147 = x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (768,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_156 = torch._C._nn.gelu(x_155, approximate="none") + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_153 + x_159 + x_153 = x_159 = None + x_161 = torch.nn.functional.layer_norm( + x_160, + (768,), + l_self_modules_norm_parameters_weight_, + l_self_modules_norm_parameters_bias_, + 1e-06, + ) + x_160 = ( + l_self_modules_norm_parameters_weight_ + ) = l_self_modules_norm_parameters_bias_ = None + x_162 = x_161[(slice(None, None, None), 0)] + x_161 = None + x_163 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_163 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_164,) diff --git a/samples/timm/deit_base_patch16_224.fb_in1k/weight_meta.py b/samples/timm/deit_base_patch16_224.fb_in1k/weight_meta.py new file mode 100644 index 0000000000..e6b299a89d --- /dev/null +++ b/samples/timm/deit_base_patch16_224.fb_in1k/weight_meta.py @@ -0,0 +1,1556 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.230 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [768, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 197, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [2304, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [2304] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [768, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [3072, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [3072] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [768, 3072] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_weight_: + name = "L_self_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_bias_: + name = "L_self_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/deit_small_distilled_patch16_224.fb_in1k/graph_hash.txt b/samples/timm/deit_small_distilled_patch16_224.fb_in1k/graph_hash.txt new file mode 100644 index 0000000000..f602a0660c --- /dev/null +++ b/samples/timm/deit_small_distilled_patch16_224.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +a6fd91eca0badc635ce20aead25e47f6b7aaeef35de34742b760572723b41494 \ No newline at end of file diff --git a/samples/timm/deit_small_distilled_patch16_224.fb_in1k/graph_net.json b/samples/timm/deit_small_distilled_patch16_224.fb_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/deit_small_distilled_patch16_224.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/deit_small_distilled_patch16_224.fb_in1k/input_meta.py b/samples/timm/deit_small_distilled_patch16_224.fb_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/deit_small_distilled_patch16_224.fb_in1k/input_tensor_constraints.py b/samples/timm/deit_small_distilled_patch16_224.fb_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/deit_small_distilled_patch16_224.fb_in1k/model.py b/samples/timm/deit_small_distilled_patch16_224.fb_in1k/model.py new file mode 100644 index 0000000000..9d7fb709b7 --- /dev/null +++ b/samples/timm/deit_small_distilled_patch16_224.fb_in1k/model.py @@ -0,0 +1,1747 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_parameters_dist_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_dist_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_dist_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_parameters_dist_token_ = L_self_parameters_dist_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_norm_parameters_weight_ = L_self_modules_norm_parameters_weight_ + l_self_modules_norm_parameters_bias_ = L_self_modules_norm_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + l_self_modules_head_dist_parameters_weight_ = ( + L_self_modules_head_dist_parameters_weight_ + ) + l_self_modules_head_dist_parameters_bias_ = ( + L_self_modules_head_dist_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + expand_1 = l_self_parameters_dist_token_.expand(1, -1, -1) + l_self_parameters_dist_token_ = None + x_2 = torch.cat((expand, expand_1, x_1), dim=1) + expand = expand_1 = x_1 = None + x_3 = x_2 + l_self_parameters_pos_embed_ + x_2 = l_self_parameters_pos_embed_ = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (384,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 198, 3, 6, 64) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 198, 384) + transpose_1 = None + x_8 = torch._C._nn.linear( + x_7, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = x_4 + x_9 + x_4 = x_9 = None + x_11 = torch.nn.functional.layer_norm( + x_10, + (384,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_11 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_13 = torch._C._nn.gelu(x_12, approximate="none") + x_12 = None + x_14 = torch.nn.functional.dropout(x_13, 0.0, False, False) + x_13 = None + x_15 = torch._C._nn.linear( + x_14, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_14 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + x_17 = x_10 + x_16 + x_10 = x_16 = None + x_18 = torch.nn.functional.layer_norm( + x_17, + (384,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_18, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_18 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 198, 3, 6, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_19 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_19.transpose(1, 2) + x_19 = None + x_20 = transpose_2.reshape(1, 198, 384) + transpose_2 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_17 + x_22 + x_17 = x_22 = None + x_24 = torch.nn.functional.layer_norm( + x_23, + (384,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_25 = torch._C._nn.linear( + x_24, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_24 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_26 = torch._C._nn.gelu(x_25, approximate="none") + x_25 = None + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + x_30 = x_23 + x_29 + x_23 = x_29 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (384,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 198, 3, 6, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_32 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_32.transpose(1, 2) + x_32 = None + x_33 = transpose_3.reshape(1, 198, 384) + transpose_3 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_33 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + x_36 = x_30 + x_35 + x_30 = x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (384,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_39 = torch._C._nn.gelu(x_38, approximate="none") + x_38 = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = torch._C._nn.linear( + x_40, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_40 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_42 = torch.nn.functional.dropout(x_41, 0.0, False, False) + x_41 = None + x_43 = x_36 + x_42 + x_36 = x_42 = None + x_44 = torch.nn.functional.layer_norm( + x_43, + (384,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_44 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 198, 3, 6, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_45 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_45.transpose(1, 2) + x_45 = None + x_46 = transpose_4.reshape(1, 198, 384) + transpose_4 = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_43 + x_48 + x_43 = x_48 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (384,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_51 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_52 = torch._C._nn.gelu(x_51, approximate="none") + x_51 = None + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = x_49 + x_55 + x_49 = x_55 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (384,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 198, 3, 6, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_5.reshape(1, 198, 384) + transpose_5 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + x_62 = x_56 + x_61 + x_56 = x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (384,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_62 + x_68 + x_62 = x_68 = None + x_70 = torch.nn.functional.layer_norm( + x_69, + (384,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_70 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 198, 3, 6, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_71 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_71.transpose(1, 2) + x_71 = None + x_72 = transpose_6.reshape(1, 198, 384) + transpose_6 = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + x_75 = x_69 + x_74 + x_69 = x_74 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (384,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_78 = torch._C._nn.gelu(x_77, approximate="none") + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch._C._nn.linear( + x_79, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_79 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = x_75 + x_81 + x_75 = x_81 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (384,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 198, 3, 6, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_84 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_84.transpose(1, 2) + x_84 = None + x_85 = transpose_7.reshape(1, 198, 384) + transpose_7 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_85 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + x_88 = x_82 + x_87 + x_82 = x_87 = None + x_89 = torch.nn.functional.layer_norm( + x_88, + (384,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_89 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_91 = torch._C._nn.gelu(x_90, approximate="none") + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_88 + x_94 + x_88 = x_94 = None + x_96 = torch.nn.functional.layer_norm( + x_95, + (384,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_96 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 198, 3, 6, 64) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_97 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_97.transpose(1, 2) + x_97 = None + x_98 = transpose_8.reshape(1, 198, 384) + transpose_8 = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + x_101 = x_95 + x_100 + x_95 = x_100 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (384,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch._C._nn.linear( + x_105, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_105 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + x_108 = x_101 + x_107 + x_101 = x_107 = None + x_109 = torch.nn.functional.layer_norm( + x_108, + (384,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_109, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_109 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 198, 3, 6, 64) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_9.reshape(1, 198, 384) + transpose_9 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_108 + x_113 + x_108 = x_113 = None + x_115 = torch.nn.functional.layer_norm( + x_114, + (384,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_116 = torch._C._nn.linear( + x_115, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_115 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + x_121 = x_114 + x_120 + x_114 = x_120 = None + x_122 = torch.nn.functional.layer_norm( + x_121, + (384,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_122 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 198, 3, 6, 64) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_123 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_123.transpose(1, 2) + x_123 = None + x_124 = transpose_10.reshape(1, 198, 384) + transpose_10 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_124 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_121 + x_126 + x_121 = x_126 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (384,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_130 = torch._C._nn.gelu(x_129, approximate="none") + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch._C._nn.linear( + x_131, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_131 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + x_134 = x_127 + x_133 + x_127 = x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (384,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_135, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_135 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 198, 3, 6, 64) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_136 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_136.transpose(1, 2) + x_136 = None + x_137 = transpose_11.reshape(1, 198, 384) + transpose_11 = None + x_138 = torch._C._nn.linear( + x_137, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_137 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = x_134 + x_139 + x_134 = x_139 = None + x_141 = torch.nn.functional.layer_norm( + x_140, + (384,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_141 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_143 = torch._C._nn.gelu(x_142, approximate="none") + x_142 = None + x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) + x_143 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_144 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = x_140 + x_146 + x_140 = x_146 = None + x_148 = torch.nn.functional.layer_norm( + x_147, + (384,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_148, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_148 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 198, 3, 6, 64) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_149 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_149.transpose(1, 2) + x_149 = None + x_150 = transpose_12.reshape(1, 198, 384) + transpose_12 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + x_153 = x_147 + x_152 + x_147 = x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (384,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_156 = torch._C._nn.gelu(x_155, approximate="none") + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_153 + x_159 + x_153 = x_159 = None + x_161 = torch.nn.functional.layer_norm( + x_160, + (384,), + l_self_modules_norm_parameters_weight_, + l_self_modules_norm_parameters_bias_, + 1e-06, + ) + x_160 = ( + l_self_modules_norm_parameters_weight_ + ) = l_self_modules_norm_parameters_bias_ = None + x_162 = x_161[(slice(None, None, None), 0)] + x_dist = x_161[(slice(None, None, None), 1)] + x_161 = None + x_163 = torch._C._nn.linear( + x_162, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_162 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + x_dist_1 = torch._C._nn.linear( + x_dist, + l_self_modules_head_dist_parameters_weight_, + l_self_modules_head_dist_parameters_bias_, + ) + x_dist = ( + l_self_modules_head_dist_parameters_weight_ + ) = l_self_modules_head_dist_parameters_bias_ = None + add_25 = x_163 + x_dist_1 + x_163 = x_dist_1 = None + x_164 = add_25 / 2 + add_25 = None + return (x_164,) diff --git a/samples/timm/deit_small_distilled_patch16_224.fb_in1k/weight_meta.py b/samples/timm/deit_small_distilled_patch16_224.fb_in1k/weight_meta.py new file mode 100644 index 0000000000..251165b153 --- /dev/null +++ b/samples/timm/deit_small_distilled_patch16_224.fb_in1k/weight_meta.py @@ -0,0 +1,1586 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.228 + std = 1.285 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [384, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 198, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_dist_token_: + name = "L_self_parameters_dist_token_" + shape = [1, 1, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_weight_: + name = "L_self_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_bias_: + name = "L_self_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_dist_parameters_weight_: + name = "L_self_modules_head_dist_parameters_weight_" + shape = [1000, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_dist_parameters_bias_: + name = "L_self_modules_head_dist_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/deit_small_patch16_224.fb_in1k/graph_hash.txt b/samples/timm/deit_small_patch16_224.fb_in1k/graph_hash.txt new file mode 100644 index 0000000000..4d06412e6f --- /dev/null +++ b/samples/timm/deit_small_patch16_224.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +dd39f4fd656faad35082e94c9b6536c58e6fa1fde59fbf16b0e8d53977453a6f \ No newline at end of file diff --git a/samples/timm/deit_small_patch16_224.fb_in1k/graph_net.json b/samples/timm/deit_small_patch16_224.fb_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/deit_small_patch16_224.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/deit_small_patch16_224.fb_in1k/input_meta.py b/samples/timm/deit_small_patch16_224.fb_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/deit_small_patch16_224.fb_in1k/input_tensor_constraints.py b/samples/timm/deit_small_patch16_224.fb_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/deit_small_patch16_224.fb_in1k/model.py b/samples/timm/deit_small_patch16_224.fb_in1k/model.py new file mode 100644 index 0000000000..d40e5970e3 --- /dev/null +++ b/samples/timm/deit_small_patch16_224.fb_in1k/model.py @@ -0,0 +1,1724 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_norm_parameters_weight_ = L_self_modules_norm_parameters_weight_ + l_self_modules_norm_parameters_bias_ = L_self_modules_norm_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + x_2 = torch.cat([expand, x_1], dim=1) + expand = x_1 = None + x_3 = x_2 + l_self_parameters_pos_embed_ + x_2 = l_self_parameters_pos_embed_ = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (384,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 197, 3, 6, 64) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 197, 384) + transpose_1 = None + x_8 = torch._C._nn.linear( + x_7, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = x_4 + x_9 + x_4 = x_9 = None + x_11 = torch.nn.functional.layer_norm( + x_10, + (384,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_11 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_13 = torch._C._nn.gelu(x_12, approximate="none") + x_12 = None + x_14 = torch.nn.functional.dropout(x_13, 0.0, False, False) + x_13 = None + x_15 = torch._C._nn.linear( + x_14, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_14 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + x_17 = x_10 + x_16 + x_10 = x_16 = None + x_18 = torch.nn.functional.layer_norm( + x_17, + (384,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_18, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_18 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 197, 3, 6, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_19 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_19.transpose(1, 2) + x_19 = None + x_20 = transpose_2.reshape(1, 197, 384) + transpose_2 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_17 + x_22 + x_17 = x_22 = None + x_24 = torch.nn.functional.layer_norm( + x_23, + (384,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_25 = torch._C._nn.linear( + x_24, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_24 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_26 = torch._C._nn.gelu(x_25, approximate="none") + x_25 = None + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + x_30 = x_23 + x_29 + x_23 = x_29 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (384,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 197, 3, 6, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_32 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_32.transpose(1, 2) + x_32 = None + x_33 = transpose_3.reshape(1, 197, 384) + transpose_3 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_33 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + x_36 = x_30 + x_35 + x_30 = x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (384,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_39 = torch._C._nn.gelu(x_38, approximate="none") + x_38 = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = torch._C._nn.linear( + x_40, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_40 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_42 = torch.nn.functional.dropout(x_41, 0.0, False, False) + x_41 = None + x_43 = x_36 + x_42 + x_36 = x_42 = None + x_44 = torch.nn.functional.layer_norm( + x_43, + (384,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_44 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 197, 3, 6, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_45 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_45.transpose(1, 2) + x_45 = None + x_46 = transpose_4.reshape(1, 197, 384) + transpose_4 = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_43 + x_48 + x_43 = x_48 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (384,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_51 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_52 = torch._C._nn.gelu(x_51, approximate="none") + x_51 = None + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = x_49 + x_55 + x_49 = x_55 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (384,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 197, 3, 6, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_5.reshape(1, 197, 384) + transpose_5 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + x_62 = x_56 + x_61 + x_56 = x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (384,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_62 + x_68 + x_62 = x_68 = None + x_70 = torch.nn.functional.layer_norm( + x_69, + (384,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_70 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 197, 3, 6, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_71 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_71.transpose(1, 2) + x_71 = None + x_72 = transpose_6.reshape(1, 197, 384) + transpose_6 = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + x_75 = x_69 + x_74 + x_69 = x_74 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (384,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_78 = torch._C._nn.gelu(x_77, approximate="none") + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch._C._nn.linear( + x_79, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_79 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = x_75 + x_81 + x_75 = x_81 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (384,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 197, 3, 6, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_84 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_84.transpose(1, 2) + x_84 = None + x_85 = transpose_7.reshape(1, 197, 384) + transpose_7 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_85 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + x_88 = x_82 + x_87 + x_82 = x_87 = None + x_89 = torch.nn.functional.layer_norm( + x_88, + (384,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_89 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_91 = torch._C._nn.gelu(x_90, approximate="none") + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_88 + x_94 + x_88 = x_94 = None + x_96 = torch.nn.functional.layer_norm( + x_95, + (384,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_96 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 197, 3, 6, 64) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_97 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_97.transpose(1, 2) + x_97 = None + x_98 = transpose_8.reshape(1, 197, 384) + transpose_8 = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + x_101 = x_95 + x_100 + x_95 = x_100 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (384,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch._C._nn.linear( + x_105, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_105 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + x_108 = x_101 + x_107 + x_101 = x_107 = None + x_109 = torch.nn.functional.layer_norm( + x_108, + (384,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_109, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_109 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 197, 3, 6, 64) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_9.reshape(1, 197, 384) + transpose_9 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_108 + x_113 + x_108 = x_113 = None + x_115 = torch.nn.functional.layer_norm( + x_114, + (384,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_116 = torch._C._nn.linear( + x_115, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_115 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + x_121 = x_114 + x_120 + x_114 = x_120 = None + x_122 = torch.nn.functional.layer_norm( + x_121, + (384,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_122 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 197, 3, 6, 64) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_123 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_123.transpose(1, 2) + x_123 = None + x_124 = transpose_10.reshape(1, 197, 384) + transpose_10 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_124 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_121 + x_126 + x_121 = x_126 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (384,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_130 = torch._C._nn.gelu(x_129, approximate="none") + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch._C._nn.linear( + x_131, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_131 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + x_134 = x_127 + x_133 + x_127 = x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (384,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_135, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_135 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 197, 3, 6, 64) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_136 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_136.transpose(1, 2) + x_136 = None + x_137 = transpose_11.reshape(1, 197, 384) + transpose_11 = None + x_138 = torch._C._nn.linear( + x_137, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_137 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = x_134 + x_139 + x_134 = x_139 = None + x_141 = torch.nn.functional.layer_norm( + x_140, + (384,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_141 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_143 = torch._C._nn.gelu(x_142, approximate="none") + x_142 = None + x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) + x_143 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_144 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = x_140 + x_146 + x_140 = x_146 = None + x_148 = torch.nn.functional.layer_norm( + x_147, + (384,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_148, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_148 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 197, 3, 6, 64) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_149 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_149.transpose(1, 2) + x_149 = None + x_150 = transpose_12.reshape(1, 197, 384) + transpose_12 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + x_153 = x_147 + x_152 + x_147 = x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (384,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_156 = torch._C._nn.gelu(x_155, approximate="none") + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_153 + x_159 + x_153 = x_159 = None + x_161 = torch.nn.functional.layer_norm( + x_160, + (384,), + l_self_modules_norm_parameters_weight_, + l_self_modules_norm_parameters_bias_, + 1e-06, + ) + x_160 = ( + l_self_modules_norm_parameters_weight_ + ) = l_self_modules_norm_parameters_bias_ = None + x_162 = x_161[(slice(None, None, None), 0)] + x_161 = None + x_163 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_163 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_164,) diff --git a/samples/timm/deit_small_patch16_224.fb_in1k/weight_meta.py b/samples/timm/deit_small_patch16_224.fb_in1k/weight_meta.py new file mode 100644 index 0000000000..f6c0831e3b --- /dev/null +++ b/samples/timm/deit_small_patch16_224.fb_in1k/weight_meta.py @@ -0,0 +1,1556 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.229 + std = 1.288 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [384, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 197, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [1152, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [384, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [1536, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [384, 1536] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_weight_: + name = "L_self_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_bias_: + name = "L_self_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 384] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/graph_hash.txt b/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/graph_hash.txt new file mode 100644 index 0000000000..9c0d5dce41 --- /dev/null +++ b/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +8f0e9cfd2900cb44268638505c1031a525ea15e20dbf26a3feddfa71f0f303b5 \ No newline at end of file diff --git a/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/graph_net.json b/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/input_meta.py b/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/input_tensor_constraints.py b/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/model.py b/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/model.py new file mode 100644 index 0000000000..df876a6e69 --- /dev/null +++ b/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/model.py @@ -0,0 +1,1747 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_parameters_dist_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_dist_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_dist_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_parameters_dist_token_ = L_self_parameters_dist_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_norm_parameters_weight_ = L_self_modules_norm_parameters_weight_ + l_self_modules_norm_parameters_bias_ = L_self_modules_norm_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + l_self_modules_head_dist_parameters_weight_ = ( + L_self_modules_head_dist_parameters_weight_ + ) + l_self_modules_head_dist_parameters_bias_ = ( + L_self_modules_head_dist_parameters_bias_ + ) + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + expand_1 = l_self_parameters_dist_token_.expand(1, -1, -1) + l_self_parameters_dist_token_ = None + x_2 = torch.cat((expand, expand_1, x_1), dim=1) + expand = expand_1 = x_1 = None + x_3 = x_2 + l_self_parameters_pos_embed_ + x_2 = l_self_parameters_pos_embed_ = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (192,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 198, 3, 3, 64) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 198, 192) + transpose_1 = None + x_8 = torch._C._nn.linear( + x_7, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = x_4 + x_9 + x_4 = x_9 = None + x_11 = torch.nn.functional.layer_norm( + x_10, + (192,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_11 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_13 = torch._C._nn.gelu(x_12, approximate="none") + x_12 = None + x_14 = torch.nn.functional.dropout(x_13, 0.0, False, False) + x_13 = None + x_15 = torch._C._nn.linear( + x_14, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_14 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + x_17 = x_10 + x_16 + x_10 = x_16 = None + x_18 = torch.nn.functional.layer_norm( + x_17, + (192,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_18, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_18 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 198, 3, 3, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_19 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_19.transpose(1, 2) + x_19 = None + x_20 = transpose_2.reshape(1, 198, 192) + transpose_2 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_17 + x_22 + x_17 = x_22 = None + x_24 = torch.nn.functional.layer_norm( + x_23, + (192,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_25 = torch._C._nn.linear( + x_24, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_24 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_26 = torch._C._nn.gelu(x_25, approximate="none") + x_25 = None + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + x_30 = x_23 + x_29 + x_23 = x_29 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (192,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 198, 3, 3, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_32 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_32.transpose(1, 2) + x_32 = None + x_33 = transpose_3.reshape(1, 198, 192) + transpose_3 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_33 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + x_36 = x_30 + x_35 + x_30 = x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (192,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_39 = torch._C._nn.gelu(x_38, approximate="none") + x_38 = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = torch._C._nn.linear( + x_40, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_40 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_42 = torch.nn.functional.dropout(x_41, 0.0, False, False) + x_41 = None + x_43 = x_36 + x_42 + x_36 = x_42 = None + x_44 = torch.nn.functional.layer_norm( + x_43, + (192,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_44 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 198, 3, 3, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_45 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_45.transpose(1, 2) + x_45 = None + x_46 = transpose_4.reshape(1, 198, 192) + transpose_4 = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_43 + x_48 + x_43 = x_48 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (192,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_51 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_52 = torch._C._nn.gelu(x_51, approximate="none") + x_51 = None + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = x_49 + x_55 + x_49 = x_55 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (192,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 198, 3, 3, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_5.reshape(1, 198, 192) + transpose_5 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + x_62 = x_56 + x_61 + x_56 = x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (192,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_62 + x_68 + x_62 = x_68 = None + x_70 = torch.nn.functional.layer_norm( + x_69, + (192,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_70 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 198, 3, 3, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_71 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_71.transpose(1, 2) + x_71 = None + x_72 = transpose_6.reshape(1, 198, 192) + transpose_6 = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + x_75 = x_69 + x_74 + x_69 = x_74 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (192,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_78 = torch._C._nn.gelu(x_77, approximate="none") + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch._C._nn.linear( + x_79, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_79 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = x_75 + x_81 + x_75 = x_81 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (192,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 198, 3, 3, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_84 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_84.transpose(1, 2) + x_84 = None + x_85 = transpose_7.reshape(1, 198, 192) + transpose_7 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_85 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + x_88 = x_82 + x_87 + x_82 = x_87 = None + x_89 = torch.nn.functional.layer_norm( + x_88, + (192,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_89 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_91 = torch._C._nn.gelu(x_90, approximate="none") + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_88 + x_94 + x_88 = x_94 = None + x_96 = torch.nn.functional.layer_norm( + x_95, + (192,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_96 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 198, 3, 3, 64) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_97 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_97.transpose(1, 2) + x_97 = None + x_98 = transpose_8.reshape(1, 198, 192) + transpose_8 = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + x_101 = x_95 + x_100 + x_95 = x_100 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (192,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch._C._nn.linear( + x_105, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_105 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + x_108 = x_101 + x_107 + x_101 = x_107 = None + x_109 = torch.nn.functional.layer_norm( + x_108, + (192,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_109, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_109 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 198, 3, 3, 64) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_9.reshape(1, 198, 192) + transpose_9 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_108 + x_113 + x_108 = x_113 = None + x_115 = torch.nn.functional.layer_norm( + x_114, + (192,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_116 = torch._C._nn.linear( + x_115, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_115 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + x_121 = x_114 + x_120 + x_114 = x_120 = None + x_122 = torch.nn.functional.layer_norm( + x_121, + (192,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_122 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 198, 3, 3, 64) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_123 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_123.transpose(1, 2) + x_123 = None + x_124 = transpose_10.reshape(1, 198, 192) + transpose_10 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_124 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_121 + x_126 + x_121 = x_126 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (192,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_130 = torch._C._nn.gelu(x_129, approximate="none") + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch._C._nn.linear( + x_131, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_131 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + x_134 = x_127 + x_133 + x_127 = x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (192,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_135, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_135 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 198, 3, 3, 64) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_136 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_136.transpose(1, 2) + x_136 = None + x_137 = transpose_11.reshape(1, 198, 192) + transpose_11 = None + x_138 = torch._C._nn.linear( + x_137, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_137 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = x_134 + x_139 + x_134 = x_139 = None + x_141 = torch.nn.functional.layer_norm( + x_140, + (192,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_141 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_143 = torch._C._nn.gelu(x_142, approximate="none") + x_142 = None + x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) + x_143 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_144 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = x_140 + x_146 + x_140 = x_146 = None + x_148 = torch.nn.functional.layer_norm( + x_147, + (192,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_148, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_148 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 198, 3, 3, 64) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_149 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_149.transpose(1, 2) + x_149 = None + x_150 = transpose_12.reshape(1, 198, 192) + transpose_12 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + x_153 = x_147 + x_152 + x_147 = x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (192,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_156 = torch._C._nn.gelu(x_155, approximate="none") + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_153 + x_159 + x_153 = x_159 = None + x_161 = torch.nn.functional.layer_norm( + x_160, + (192,), + l_self_modules_norm_parameters_weight_, + l_self_modules_norm_parameters_bias_, + 1e-06, + ) + x_160 = ( + l_self_modules_norm_parameters_weight_ + ) = l_self_modules_norm_parameters_bias_ = None + x_162 = x_161[(slice(None, None, None), 0)] + x_dist = x_161[(slice(None, None, None), 1)] + x_161 = None + x_163 = torch._C._nn.linear( + x_162, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_162 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + x_dist_1 = torch._C._nn.linear( + x_dist, + l_self_modules_head_dist_parameters_weight_, + l_self_modules_head_dist_parameters_bias_, + ) + x_dist = ( + l_self_modules_head_dist_parameters_weight_ + ) = l_self_modules_head_dist_parameters_bias_ = None + add_25 = x_163 + x_dist_1 + x_163 = x_dist_1 = None + x_164 = add_25 / 2 + add_25 = None + return (x_164,) diff --git a/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/weight_meta.py b/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/weight_meta.py new file mode 100644 index 0000000000..93d831932e --- /dev/null +++ b/samples/timm/deit_tiny_distilled_patch16_224.fb_in1k/weight_meta.py @@ -0,0 +1,1586 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.226 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [192, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 198, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_dist_token_: + name = "L_self_parameters_dist_token_" + shape = [1, 1, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_weight_: + name = "L_self_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_bias_: + name = "L_self_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_dist_parameters_weight_: + name = "L_self_modules_head_dist_parameters_weight_" + shape = [1000, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_dist_parameters_bias_: + name = "L_self_modules_head_dist_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/deit_tiny_patch16_224.fb_in1k/graph_hash.txt b/samples/timm/deit_tiny_patch16_224.fb_in1k/graph_hash.txt new file mode 100644 index 0000000000..b269f07194 --- /dev/null +++ b/samples/timm/deit_tiny_patch16_224.fb_in1k/graph_hash.txt @@ -0,0 +1 @@ +280ac56eaaf4721e2023e2a905b9645fb0eea6be3fa203adaded06373c53b6b5 \ No newline at end of file diff --git a/samples/timm/deit_tiny_patch16_224.fb_in1k/graph_net.json b/samples/timm/deit_tiny_patch16_224.fb_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/deit_tiny_patch16_224.fb_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/deit_tiny_patch16_224.fb_in1k/input_meta.py b/samples/timm/deit_tiny_patch16_224.fb_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/deit_tiny_patch16_224.fb_in1k/input_tensor_constraints.py b/samples/timm/deit_tiny_patch16_224.fb_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/deit_tiny_patch16_224.fb_in1k/model.py b/samples/timm/deit_tiny_patch16_224.fb_in1k/model.py new file mode 100644 index 0000000000..9b99f4fff8 --- /dev/null +++ b/samples/timm/deit_tiny_patch16_224.fb_in1k/model.py @@ -0,0 +1,1724 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_x_: torch.Tensor, + L_self_modules_patch_embed_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_patch_embed_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_parameters_pos_embed_: torch.nn.parameter.Parameter, + L_self_parameters_cls_token_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_head_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_x_ = L_x_ + l_self_modules_patch_embed_modules_proj_parameters_weight_ = ( + L_self_modules_patch_embed_modules_proj_parameters_weight_ + ) + l_self_modules_patch_embed_modules_proj_parameters_bias_ = ( + L_self_modules_patch_embed_modules_proj_parameters_bias_ + ) + l_self_parameters_pos_embed_ = L_self_parameters_pos_embed_ + l_self_parameters_cls_token_ = L_self_parameters_cls_token_ + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ = ( + L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) + l_self_modules_norm_parameters_weight_ = L_self_modules_norm_parameters_weight_ + l_self_modules_norm_parameters_bias_ = L_self_modules_norm_parameters_bias_ + l_self_modules_head_parameters_weight_ = L_self_modules_head_parameters_weight_ + l_self_modules_head_parameters_bias_ = L_self_modules_head_parameters_bias_ + x = torch.conv2d( + l_x_, + l_self_modules_patch_embed_modules_proj_parameters_weight_, + l_self_modules_patch_embed_modules_proj_parameters_bias_, + (16, 16), + (0, 0), + (1, 1), + 1, + ) + l_x_ = ( + l_self_modules_patch_embed_modules_proj_parameters_weight_ + ) = l_self_modules_patch_embed_modules_proj_parameters_bias_ = None + flatten = x.flatten(2) + x = None + x_1 = flatten.transpose(1, 2) + flatten = None + expand = l_self_parameters_cls_token_.expand(1, -1, -1) + l_self_parameters_cls_token_ = None + x_2 = torch.cat([expand, x_1], dim=1) + expand = x_1 = None + x_3 = x_2 + l_self_parameters_pos_embed_ + x_2 = l_self_parameters_pos_embed_ = None + x_4 = torch.nn.functional.dropout(x_3, 0.0, False, False) + x_3 = None + x_5 = torch.nn.functional.layer_norm( + x_4, + (192,), + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm1_parameters_bias_ + ) = None + linear = torch._C._nn.linear( + x_5, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_, + ) + x_5 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape = linear.reshape(1, 197, 3, 3, 64) + linear = None + qkv = reshape.permute(2, 0, 3, 1, 4) + reshape = None + unbind = qkv.unbind(0) + qkv = None + q = unbind[0] + k = unbind[1] + v = unbind[2] + unbind = None + x_6 = torch._C._nn.scaled_dot_product_attention( + q, k, v, attn_mask=None, dropout_p=0.0 + ) + q = k = v = None + transpose_1 = x_6.transpose(1, 2) + x_6 = None + x_7 = transpose_1.reshape(1, 197, 192) + transpose_1 = None + x_8 = torch._C._nn.linear( + x_7, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_, + ) + x_7 = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_ + ) = None + x_9 = torch.nn.functional.dropout(x_8, 0.0, False, False) + x_8 = None + x_10 = x_4 + x_9 + x_4 = x_9 = None + x_11 = torch.nn.functional.layer_norm( + x_10, + (192,), + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_0_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_0_modules_norm2_parameters_bias_ + ) = None + x_12 = torch._C._nn.linear( + x_11, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_, + ) + x_11 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_13 = torch._C._nn.gelu(x_12, approximate="none") + x_12 = None + x_14 = torch.nn.functional.dropout(x_13, 0.0, False, False) + x_13 = None + x_15 = torch._C._nn.linear( + x_14, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_, + ) + x_14 = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_16 = torch.nn.functional.dropout(x_15, 0.0, False, False) + x_15 = None + x_17 = x_10 + x_16 + x_10 = x_16 = None + x_18 = torch.nn.functional.layer_norm( + x_17, + (192,), + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm1_parameters_bias_ + ) = None + linear_4 = torch._C._nn.linear( + x_18, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_, + ) + x_18 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_2 = linear_4.reshape(1, 197, 3, 3, 64) + linear_4 = None + qkv_1 = reshape_2.permute(2, 0, 3, 1, 4) + reshape_2 = None + unbind_1 = qkv_1.unbind(0) + qkv_1 = None + q_1 = unbind_1[0] + k_1 = unbind_1[1] + v_1 = unbind_1[2] + unbind_1 = None + x_19 = torch._C._nn.scaled_dot_product_attention( + q_1, k_1, v_1, attn_mask=None, dropout_p=0.0 + ) + q_1 = k_1 = v_1 = None + transpose_2 = x_19.transpose(1, 2) + x_19 = None + x_20 = transpose_2.reshape(1, 197, 192) + transpose_2 = None + x_21 = torch._C._nn.linear( + x_20, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_, + ) + x_20 = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_ + ) = None + x_22 = torch.nn.functional.dropout(x_21, 0.0, False, False) + x_21 = None + x_23 = x_17 + x_22 + x_17 = x_22 = None + x_24 = torch.nn.functional.layer_norm( + x_23, + (192,), + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_1_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_1_modules_norm2_parameters_bias_ + ) = None + x_25 = torch._C._nn.linear( + x_24, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_, + ) + x_24 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_26 = torch._C._nn.gelu(x_25, approximate="none") + x_25 = None + x_27 = torch.nn.functional.dropout(x_26, 0.0, False, False) + x_26 = None + x_28 = torch._C._nn.linear( + x_27, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_, + ) + x_27 = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_29 = torch.nn.functional.dropout(x_28, 0.0, False, False) + x_28 = None + x_30 = x_23 + x_29 + x_23 = x_29 = None + x_31 = torch.nn.functional.layer_norm( + x_30, + (192,), + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm1_parameters_bias_ + ) = None + linear_8 = torch._C._nn.linear( + x_31, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_, + ) + x_31 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_4 = linear_8.reshape(1, 197, 3, 3, 64) + linear_8 = None + qkv_2 = reshape_4.permute(2, 0, 3, 1, 4) + reshape_4 = None + unbind_2 = qkv_2.unbind(0) + qkv_2 = None + q_2 = unbind_2[0] + k_2 = unbind_2[1] + v_2 = unbind_2[2] + unbind_2 = None + x_32 = torch._C._nn.scaled_dot_product_attention( + q_2, k_2, v_2, attn_mask=None, dropout_p=0.0 + ) + q_2 = k_2 = v_2 = None + transpose_3 = x_32.transpose(1, 2) + x_32 = None + x_33 = transpose_3.reshape(1, 197, 192) + transpose_3 = None + x_34 = torch._C._nn.linear( + x_33, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_, + ) + x_33 = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_ + ) = None + x_35 = torch.nn.functional.dropout(x_34, 0.0, False, False) + x_34 = None + x_36 = x_30 + x_35 + x_30 = x_35 = None + x_37 = torch.nn.functional.layer_norm( + x_36, + (192,), + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_2_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_2_modules_norm2_parameters_bias_ + ) = None + x_38 = torch._C._nn.linear( + x_37, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_, + ) + x_37 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_39 = torch._C._nn.gelu(x_38, approximate="none") + x_38 = None + x_40 = torch.nn.functional.dropout(x_39, 0.0, False, False) + x_39 = None + x_41 = torch._C._nn.linear( + x_40, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_, + ) + x_40 = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_42 = torch.nn.functional.dropout(x_41, 0.0, False, False) + x_41 = None + x_43 = x_36 + x_42 + x_36 = x_42 = None + x_44 = torch.nn.functional.layer_norm( + x_43, + (192,), + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm1_parameters_bias_ + ) = None + linear_12 = torch._C._nn.linear( + x_44, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_, + ) + x_44 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_6 = linear_12.reshape(1, 197, 3, 3, 64) + linear_12 = None + qkv_3 = reshape_6.permute(2, 0, 3, 1, 4) + reshape_6 = None + unbind_3 = qkv_3.unbind(0) + qkv_3 = None + q_3 = unbind_3[0] + k_3 = unbind_3[1] + v_3 = unbind_3[2] + unbind_3 = None + x_45 = torch._C._nn.scaled_dot_product_attention( + q_3, k_3, v_3, attn_mask=None, dropout_p=0.0 + ) + q_3 = k_3 = v_3 = None + transpose_4 = x_45.transpose(1, 2) + x_45 = None + x_46 = transpose_4.reshape(1, 197, 192) + transpose_4 = None + x_47 = torch._C._nn.linear( + x_46, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_, + ) + x_46 = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_ + ) = None + x_48 = torch.nn.functional.dropout(x_47, 0.0, False, False) + x_47 = None + x_49 = x_43 + x_48 + x_43 = x_48 = None + x_50 = torch.nn.functional.layer_norm( + x_49, + (192,), + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_3_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_3_modules_norm2_parameters_bias_ + ) = None + x_51 = torch._C._nn.linear( + x_50, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_, + ) + x_50 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_52 = torch._C._nn.gelu(x_51, approximate="none") + x_51 = None + x_53 = torch.nn.functional.dropout(x_52, 0.0, False, False) + x_52 = None + x_54 = torch._C._nn.linear( + x_53, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_, + ) + x_53 = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_55 = torch.nn.functional.dropout(x_54, 0.0, False, False) + x_54 = None + x_56 = x_49 + x_55 + x_49 = x_55 = None + x_57 = torch.nn.functional.layer_norm( + x_56, + (192,), + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm1_parameters_bias_ + ) = None + linear_16 = torch._C._nn.linear( + x_57, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_, + ) + x_57 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_8 = linear_16.reshape(1, 197, 3, 3, 64) + linear_16 = None + qkv_4 = reshape_8.permute(2, 0, 3, 1, 4) + reshape_8 = None + unbind_4 = qkv_4.unbind(0) + qkv_4 = None + q_4 = unbind_4[0] + k_4 = unbind_4[1] + v_4 = unbind_4[2] + unbind_4 = None + x_58 = torch._C._nn.scaled_dot_product_attention( + q_4, k_4, v_4, attn_mask=None, dropout_p=0.0 + ) + q_4 = k_4 = v_4 = None + transpose_5 = x_58.transpose(1, 2) + x_58 = None + x_59 = transpose_5.reshape(1, 197, 192) + transpose_5 = None + x_60 = torch._C._nn.linear( + x_59, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_, + ) + x_59 = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_ + ) = None + x_61 = torch.nn.functional.dropout(x_60, 0.0, False, False) + x_60 = None + x_62 = x_56 + x_61 + x_56 = x_61 = None + x_63 = torch.nn.functional.layer_norm( + x_62, + (192,), + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_4_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_4_modules_norm2_parameters_bias_ + ) = None + x_64 = torch._C._nn.linear( + x_63, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_, + ) + x_63 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_65 = torch._C._nn.gelu(x_64, approximate="none") + x_64 = None + x_66 = torch.nn.functional.dropout(x_65, 0.0, False, False) + x_65 = None + x_67 = torch._C._nn.linear( + x_66, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_, + ) + x_66 = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_68 = torch.nn.functional.dropout(x_67, 0.0, False, False) + x_67 = None + x_69 = x_62 + x_68 + x_62 = x_68 = None + x_70 = torch.nn.functional.layer_norm( + x_69, + (192,), + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm1_parameters_bias_ + ) = None + linear_20 = torch._C._nn.linear( + x_70, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_, + ) + x_70 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_10 = linear_20.reshape(1, 197, 3, 3, 64) + linear_20 = None + qkv_5 = reshape_10.permute(2, 0, 3, 1, 4) + reshape_10 = None + unbind_5 = qkv_5.unbind(0) + qkv_5 = None + q_5 = unbind_5[0] + k_5 = unbind_5[1] + v_5 = unbind_5[2] + unbind_5 = None + x_71 = torch._C._nn.scaled_dot_product_attention( + q_5, k_5, v_5, attn_mask=None, dropout_p=0.0 + ) + q_5 = k_5 = v_5 = None + transpose_6 = x_71.transpose(1, 2) + x_71 = None + x_72 = transpose_6.reshape(1, 197, 192) + transpose_6 = None + x_73 = torch._C._nn.linear( + x_72, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_, + ) + x_72 = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_ + ) = None + x_74 = torch.nn.functional.dropout(x_73, 0.0, False, False) + x_73 = None + x_75 = x_69 + x_74 + x_69 = x_74 = None + x_76 = torch.nn.functional.layer_norm( + x_75, + (192,), + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_5_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_5_modules_norm2_parameters_bias_ + ) = None + x_77 = torch._C._nn.linear( + x_76, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_, + ) + x_76 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_78 = torch._C._nn.gelu(x_77, approximate="none") + x_77 = None + x_79 = torch.nn.functional.dropout(x_78, 0.0, False, False) + x_78 = None + x_80 = torch._C._nn.linear( + x_79, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_, + ) + x_79 = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_81 = torch.nn.functional.dropout(x_80, 0.0, False, False) + x_80 = None + x_82 = x_75 + x_81 + x_75 = x_81 = None + x_83 = torch.nn.functional.layer_norm( + x_82, + (192,), + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm1_parameters_bias_ + ) = None + linear_24 = torch._C._nn.linear( + x_83, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_, + ) + x_83 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_12 = linear_24.reshape(1, 197, 3, 3, 64) + linear_24 = None + qkv_6 = reshape_12.permute(2, 0, 3, 1, 4) + reshape_12 = None + unbind_6 = qkv_6.unbind(0) + qkv_6 = None + q_6 = unbind_6[0] + k_6 = unbind_6[1] + v_6 = unbind_6[2] + unbind_6 = None + x_84 = torch._C._nn.scaled_dot_product_attention( + q_6, k_6, v_6, attn_mask=None, dropout_p=0.0 + ) + q_6 = k_6 = v_6 = None + transpose_7 = x_84.transpose(1, 2) + x_84 = None + x_85 = transpose_7.reshape(1, 197, 192) + transpose_7 = None + x_86 = torch._C._nn.linear( + x_85, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_, + ) + x_85 = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_ + ) = None + x_87 = torch.nn.functional.dropout(x_86, 0.0, False, False) + x_86 = None + x_88 = x_82 + x_87 + x_82 = x_87 = None + x_89 = torch.nn.functional.layer_norm( + x_88, + (192,), + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_6_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_6_modules_norm2_parameters_bias_ + ) = None + x_90 = torch._C._nn.linear( + x_89, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_, + ) + x_89 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_91 = torch._C._nn.gelu(x_90, approximate="none") + x_90 = None + x_92 = torch.nn.functional.dropout(x_91, 0.0, False, False) + x_91 = None + x_93 = torch._C._nn.linear( + x_92, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_, + ) + x_92 = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_94 = torch.nn.functional.dropout(x_93, 0.0, False, False) + x_93 = None + x_95 = x_88 + x_94 + x_88 = x_94 = None + x_96 = torch.nn.functional.layer_norm( + x_95, + (192,), + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm1_parameters_bias_ + ) = None + linear_28 = torch._C._nn.linear( + x_96, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_, + ) + x_96 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_14 = linear_28.reshape(1, 197, 3, 3, 64) + linear_28 = None + qkv_7 = reshape_14.permute(2, 0, 3, 1, 4) + reshape_14 = None + unbind_7 = qkv_7.unbind(0) + qkv_7 = None + q_7 = unbind_7[0] + k_7 = unbind_7[1] + v_7 = unbind_7[2] + unbind_7 = None + x_97 = torch._C._nn.scaled_dot_product_attention( + q_7, k_7, v_7, attn_mask=None, dropout_p=0.0 + ) + q_7 = k_7 = v_7 = None + transpose_8 = x_97.transpose(1, 2) + x_97 = None + x_98 = transpose_8.reshape(1, 197, 192) + transpose_8 = None + x_99 = torch._C._nn.linear( + x_98, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_, + ) + x_98 = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_ + ) = None + x_100 = torch.nn.functional.dropout(x_99, 0.0, False, False) + x_99 = None + x_101 = x_95 + x_100 + x_95 = x_100 = None + x_102 = torch.nn.functional.layer_norm( + x_101, + (192,), + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_7_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_7_modules_norm2_parameters_bias_ + ) = None + x_103 = torch._C._nn.linear( + x_102, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_, + ) + x_102 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_104 = torch._C._nn.gelu(x_103, approximate="none") + x_103 = None + x_105 = torch.nn.functional.dropout(x_104, 0.0, False, False) + x_104 = None + x_106 = torch._C._nn.linear( + x_105, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_, + ) + x_105 = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_107 = torch.nn.functional.dropout(x_106, 0.0, False, False) + x_106 = None + x_108 = x_101 + x_107 + x_101 = x_107 = None + x_109 = torch.nn.functional.layer_norm( + x_108, + (192,), + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm1_parameters_bias_ + ) = None + linear_32 = torch._C._nn.linear( + x_109, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_, + ) + x_109 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_16 = linear_32.reshape(1, 197, 3, 3, 64) + linear_32 = None + qkv_8 = reshape_16.permute(2, 0, 3, 1, 4) + reshape_16 = None + unbind_8 = qkv_8.unbind(0) + qkv_8 = None + q_8 = unbind_8[0] + k_8 = unbind_8[1] + v_8 = unbind_8[2] + unbind_8 = None + x_110 = torch._C._nn.scaled_dot_product_attention( + q_8, k_8, v_8, attn_mask=None, dropout_p=0.0 + ) + q_8 = k_8 = v_8 = None + transpose_9 = x_110.transpose(1, 2) + x_110 = None + x_111 = transpose_9.reshape(1, 197, 192) + transpose_9 = None + x_112 = torch._C._nn.linear( + x_111, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_, + ) + x_111 = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_ + ) = None + x_113 = torch.nn.functional.dropout(x_112, 0.0, False, False) + x_112 = None + x_114 = x_108 + x_113 + x_108 = x_113 = None + x_115 = torch.nn.functional.layer_norm( + x_114, + (192,), + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_8_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_8_modules_norm2_parameters_bias_ + ) = None + x_116 = torch._C._nn.linear( + x_115, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_, + ) + x_115 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_117 = torch._C._nn.gelu(x_116, approximate="none") + x_116 = None + x_118 = torch.nn.functional.dropout(x_117, 0.0, False, False) + x_117 = None + x_119 = torch._C._nn.linear( + x_118, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_, + ) + x_118 = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_120 = torch.nn.functional.dropout(x_119, 0.0, False, False) + x_119 = None + x_121 = x_114 + x_120 + x_114 = x_120 = None + x_122 = torch.nn.functional.layer_norm( + x_121, + (192,), + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm1_parameters_bias_ + ) = None + linear_36 = torch._C._nn.linear( + x_122, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_, + ) + x_122 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_18 = linear_36.reshape(1, 197, 3, 3, 64) + linear_36 = None + qkv_9 = reshape_18.permute(2, 0, 3, 1, 4) + reshape_18 = None + unbind_9 = qkv_9.unbind(0) + qkv_9 = None + q_9 = unbind_9[0] + k_9 = unbind_9[1] + v_9 = unbind_9[2] + unbind_9 = None + x_123 = torch._C._nn.scaled_dot_product_attention( + q_9, k_9, v_9, attn_mask=None, dropout_p=0.0 + ) + q_9 = k_9 = v_9 = None + transpose_10 = x_123.transpose(1, 2) + x_123 = None + x_124 = transpose_10.reshape(1, 197, 192) + transpose_10 = None + x_125 = torch._C._nn.linear( + x_124, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_, + ) + x_124 = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_ + ) = None + x_126 = torch.nn.functional.dropout(x_125, 0.0, False, False) + x_125 = None + x_127 = x_121 + x_126 + x_121 = x_126 = None + x_128 = torch.nn.functional.layer_norm( + x_127, + (192,), + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_9_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_9_modules_norm2_parameters_bias_ + ) = None + x_129 = torch._C._nn.linear( + x_128, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_, + ) + x_128 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_130 = torch._C._nn.gelu(x_129, approximate="none") + x_129 = None + x_131 = torch.nn.functional.dropout(x_130, 0.0, False, False) + x_130 = None + x_132 = torch._C._nn.linear( + x_131, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_, + ) + x_131 = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_133 = torch.nn.functional.dropout(x_132, 0.0, False, False) + x_132 = None + x_134 = x_127 + x_133 + x_127 = x_133 = None + x_135 = torch.nn.functional.layer_norm( + x_134, + (192,), + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm1_parameters_bias_ + ) = None + linear_40 = torch._C._nn.linear( + x_135, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_, + ) + x_135 = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_20 = linear_40.reshape(1, 197, 3, 3, 64) + linear_40 = None + qkv_10 = reshape_20.permute(2, 0, 3, 1, 4) + reshape_20 = None + unbind_10 = qkv_10.unbind(0) + qkv_10 = None + q_10 = unbind_10[0] + k_10 = unbind_10[1] + v_10 = unbind_10[2] + unbind_10 = None + x_136 = torch._C._nn.scaled_dot_product_attention( + q_10, k_10, v_10, attn_mask=None, dropout_p=0.0 + ) + q_10 = k_10 = v_10 = None + transpose_11 = x_136.transpose(1, 2) + x_136 = None + x_137 = transpose_11.reshape(1, 197, 192) + transpose_11 = None + x_138 = torch._C._nn.linear( + x_137, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_, + ) + x_137 = l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_ + ) = None + x_139 = torch.nn.functional.dropout(x_138, 0.0, False, False) + x_138 = None + x_140 = x_134 + x_139 + x_134 = x_139 = None + x_141 = torch.nn.functional.layer_norm( + x_140, + (192,), + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_10_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_10_modules_norm2_parameters_bias_ + ) = None + x_142 = torch._C._nn.linear( + x_141, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_, + ) + x_141 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_143 = torch._C._nn.gelu(x_142, approximate="none") + x_142 = None + x_144 = torch.nn.functional.dropout(x_143, 0.0, False, False) + x_143 = None + x_145 = torch._C._nn.linear( + x_144, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_, + ) + x_144 = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_146 = torch.nn.functional.dropout(x_145, 0.0, False, False) + x_145 = None + x_147 = x_140 + x_146 + x_140 = x_146 = None + x_148 = torch.nn.functional.layer_norm( + x_147, + (192,), + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm1_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm1_parameters_bias_ + ) = None + linear_44 = torch._C._nn.linear( + x_148, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_, + ) + x_148 = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_ + ) = None + reshape_22 = linear_44.reshape(1, 197, 3, 3, 64) + linear_44 = None + qkv_11 = reshape_22.permute(2, 0, 3, 1, 4) + reshape_22 = None + unbind_11 = qkv_11.unbind(0) + qkv_11 = None + q_11 = unbind_11[0] + k_11 = unbind_11[1] + v_11 = unbind_11[2] + unbind_11 = None + x_149 = torch._C._nn.scaled_dot_product_attention( + q_11, k_11, v_11, attn_mask=None, dropout_p=0.0 + ) + q_11 = k_11 = v_11 = None + transpose_12 = x_149.transpose(1, 2) + x_149 = None + x_150 = transpose_12.reshape(1, 197, 192) + transpose_12 = None + x_151 = torch._C._nn.linear( + x_150, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_, + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_, + ) + x_150 = l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_ + ) = None + x_152 = torch.nn.functional.dropout(x_151, 0.0, False, False) + x_151 = None + x_153 = x_147 + x_152 + x_147 = x_152 = None + x_154 = torch.nn.functional.layer_norm( + x_153, + (192,), + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_, + 1e-06, + ) + l_self_modules_blocks_modules_11_modules_norm2_parameters_weight_ = ( + l_self_modules_blocks_modules_11_modules_norm2_parameters_bias_ + ) = None + x_155 = torch._C._nn.linear( + x_154, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_, + ) + x_154 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_ + ) = None + x_156 = torch._C._nn.gelu(x_155, approximate="none") + x_155 = None + x_157 = torch.nn.functional.dropout(x_156, 0.0, False, False) + x_156 = None + x_158 = torch._C._nn.linear( + x_157, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_, + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_, + ) + x_157 = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_ + ) = ( + l_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_ + ) = None + x_159 = torch.nn.functional.dropout(x_158, 0.0, False, False) + x_158 = None + x_160 = x_153 + x_159 + x_153 = x_159 = None + x_161 = torch.nn.functional.layer_norm( + x_160, + (192,), + l_self_modules_norm_parameters_weight_, + l_self_modules_norm_parameters_bias_, + 1e-06, + ) + x_160 = ( + l_self_modules_norm_parameters_weight_ + ) = l_self_modules_norm_parameters_bias_ = None + x_162 = x_161[(slice(None, None, None), 0)] + x_161 = None + x_163 = torch.nn.functional.dropout(x_162, 0.0, False, False) + x_162 = None + x_164 = torch._C._nn.linear( + x_163, + l_self_modules_head_parameters_weight_, + l_self_modules_head_parameters_bias_, + ) + x_163 = ( + l_self_modules_head_parameters_weight_ + ) = l_self_modules_head_parameters_bias_ = None + return (x_164,) diff --git a/samples/timm/deit_tiny_patch16_224.fb_in1k/weight_meta.py b/samples/timm/deit_tiny_patch16_224.fb_in1k/weight_meta.py new file mode 100644 index 0000000000..a76a4e305e --- /dev/null +++ b/samples/timm/deit_tiny_patch16_224.fb_in1k/weight_meta.py @@ -0,0 +1,1556 @@ +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.228 + std = 1.283 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_weight_: + name = "L_self_modules_patch_embed_modules_proj_parameters_weight_" + shape = [192, 3, 16, 16] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_patch_embed_modules_proj_parameters_bias_: + name = "L_self_modules_patch_embed_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_pos_embed_: + name = "L_self_parameters_pos_embed_" + shape = [1, 197, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_parameters_cls_token_: + name = "L_self_parameters_cls_token_" + shape = [1, 1, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_0_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_1_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_2_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_3_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_4_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_5_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_6_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_7_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_8_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_weight_" + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_9_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_10_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_weight_" + ) + shape = [576, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_qkv_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_: + name = ( + "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_weight_" + ) + shape = [192, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_attn_modules_proj_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_weight_" + shape = [768, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_weight_" + shape = [192, 768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_: + name = "L_self_modules_blocks_modules_11_modules_mlp_modules_fc2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_weight_: + name = "L_self_modules_norm_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_norm_parameters_bias_: + name = "L_self_modules_norm_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_weight_: + name = "L_self_modules_head_parameters_weight_" + shape = [1000, 192] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.020 + data = None + + +class Program_weight_tensor_meta_L_self_modules_head_parameters_bias_: + name = "L_self_modules_head_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/densenet121.ra_in1k/graph_hash.txt b/samples/timm/densenet121.ra_in1k/graph_hash.txt new file mode 100644 index 0000000000..4aefe5c6c2 --- /dev/null +++ b/samples/timm/densenet121.ra_in1k/graph_hash.txt @@ -0,0 +1 @@ +7d877cb38ecbe22220db7dc1f7b2d0cf9438bfb4f70998ecee1ae10941e4eeb2 \ No newline at end of file diff --git a/samples/timm/densenet121.ra_in1k/graph_net.json b/samples/timm/densenet121.ra_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/densenet121.ra_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/densenet121.ra_in1k/input_meta.py b/samples/timm/densenet121.ra_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/densenet121.ra_in1k/input_tensor_constraints.py b/samples/timm/densenet121.ra_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/densenet121.ra_in1k/model.py b/samples/timm/densenet121.ra_in1k/model.py new file mode 100644 index 0000000000..7aaea4dc33 --- /dev/null +++ b/samples/timm/densenet121.ra_in1k/model.py @@ -0,0 +1,4977 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_features_modules_conv0_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_features_modules_norm0_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_norm0_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_norm0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm5_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_norm5_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_norm5_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm5_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_classifier_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_classifier_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_features_modules_conv0_parameters_weight_ = ( + L_self_modules_features_modules_conv0_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_features_modules_norm0_buffers_running_mean_ = ( + L_self_modules_features_modules_norm0_buffers_running_mean_ + ) + l_self_modules_features_modules_norm0_buffers_running_var_ = ( + L_self_modules_features_modules_norm0_buffers_running_var_ + ) + l_self_modules_features_modules_norm0_parameters_weight_ = ( + L_self_modules_features_modules_norm0_parameters_weight_ + ) + l_self_modules_features_modules_norm0_parameters_bias_ = ( + L_self_modules_features_modules_norm0_parameters_bias_ + ) + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition1_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition1_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition1_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition2_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition2_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition2_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition3_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition3_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition3_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ + l_self_modules_features_modules_norm5_buffers_running_mean_ = ( + L_self_modules_features_modules_norm5_buffers_running_mean_ + ) + l_self_modules_features_modules_norm5_buffers_running_var_ = ( + L_self_modules_features_modules_norm5_buffers_running_var_ + ) + l_self_modules_features_modules_norm5_parameters_weight_ = ( + L_self_modules_features_modules_norm5_parameters_weight_ + ) + l_self_modules_features_modules_norm5_parameters_bias_ = ( + L_self_modules_features_modules_norm5_parameters_bias_ + ) + l_self_modules_classifier_parameters_weight_ = ( + L_self_modules_classifier_parameters_weight_ + ) + l_self_modules_classifier_parameters_bias_ = ( + L_self_modules_classifier_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_features_modules_conv0_parameters_weight_, + None, + (2, 2), + (3, 3), + (1, 1), + 1, + ) + l_x_ = l_self_modules_features_modules_conv0_parameters_weight_ = None + x = torch.nn.functional.batch_norm( + input_1, + l_self_modules_features_modules_norm0_buffers_running_mean_, + l_self_modules_features_modules_norm0_buffers_running_var_, + l_self_modules_features_modules_norm0_parameters_weight_, + l_self_modules_features_modules_norm0_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_1 = ( + l_self_modules_features_modules_norm0_buffers_running_mean_ + ) = ( + l_self_modules_features_modules_norm0_buffers_running_var_ + ) = ( + l_self_modules_features_modules_norm0_parameters_weight_ + ) = l_self_modules_features_modules_norm0_parameters_bias_ = None + x_1 = torch.nn.functional.relu(x, inplace=True) + x = None + input_2 = torch.nn.functional.max_pool2d( + x_1, 3, 2, 1, 1, ceil_mode=False, return_indices=False + ) + x_1 = None + concated_features = torch.cat([input_2], 1) + x_2 = torch.nn.functional.batch_norm( + concated_features, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_3 = torch.nn.functional.relu(x_2, inplace=True) + x_2 = None + bottleneck_output = torch.conv2d( + x_3, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_3 = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + bottleneck_output, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_5 = torch.nn.functional.relu(x_4, inplace=True) + x_4 = None + new_features = torch.conv2d( + x_5, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_5 = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_1 = torch.cat([input_2, new_features], 1) + x_6 = torch.nn.functional.batch_norm( + concated_features_1, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_1 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_7 = torch.nn.functional.relu(x_6, inplace=True) + x_6 = None + bottleneck_output_1 = torch.conv2d( + x_7, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_7 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_8 = torch.nn.functional.batch_norm( + bottleneck_output_1, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_1 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_9 = torch.nn.functional.relu(x_8, inplace=True) + x_8 = None + new_features_1 = torch.conv2d( + x_9, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_9 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_2 = torch.cat([input_2, new_features, new_features_1], 1) + x_10 = torch.nn.functional.batch_norm( + concated_features_2, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_2 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_11 = torch.nn.functional.relu(x_10, inplace=True) + x_10 = None + bottleneck_output_2 = torch.conv2d( + x_11, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_11 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_12 = torch.nn.functional.batch_norm( + bottleneck_output_2, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_2 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_13 = torch.nn.functional.relu(x_12, inplace=True) + x_12 = None + new_features_2 = torch.conv2d( + x_13, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_13 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_3 = torch.cat( + [input_2, new_features, new_features_1, new_features_2], 1 + ) + x_14 = torch.nn.functional.batch_norm( + concated_features_3, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_3 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_15 = torch.nn.functional.relu(x_14, inplace=True) + x_14 = None + bottleneck_output_3 = torch.conv2d( + x_15, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_15 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_16 = torch.nn.functional.batch_norm( + bottleneck_output_3, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_3 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_17 = torch.nn.functional.relu(x_16, inplace=True) + x_16 = None + new_features_3 = torch.conv2d( + x_17, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_17 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_4 = torch.cat( + [input_2, new_features, new_features_1, new_features_2, new_features_3], 1 + ) + x_18 = torch.nn.functional.batch_norm( + concated_features_4, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_4 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_19 = torch.nn.functional.relu(x_18, inplace=True) + x_18 = None + bottleneck_output_4 = torch.conv2d( + x_19, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_19 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_20 = torch.nn.functional.batch_norm( + bottleneck_output_4, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_4 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_21 = torch.nn.functional.relu(x_20, inplace=True) + x_20 = None + new_features_4 = torch.conv2d( + x_21, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_21 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_5 = torch.cat( + [ + input_2, + new_features, + new_features_1, + new_features_2, + new_features_3, + new_features_4, + ], + 1, + ) + x_22 = torch.nn.functional.batch_norm( + concated_features_5, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_5 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_23 = torch.nn.functional.relu(x_22, inplace=True) + x_22 = None + bottleneck_output_5 = torch.conv2d( + x_23, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_24 = torch.nn.functional.batch_norm( + bottleneck_output_5, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_5 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_25 = torch.nn.functional.relu(x_24, inplace=True) + x_24 = None + new_features_5 = torch.conv2d( + x_25, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_25 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + input_3 = torch.cat( + [ + input_2, + new_features, + new_features_1, + new_features_2, + new_features_3, + new_features_4, + new_features_5, + ], + 1, + ) + input_2 = ( + new_features + ) = ( + new_features_1 + ) = new_features_2 = new_features_3 = new_features_4 = new_features_5 = None + x_26 = torch.nn.functional.batch_norm( + input_3, + l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_3 = l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_ + ) = None + x_27 = torch.nn.functional.relu(x_26, inplace=True) + x_26 = None + input_4 = torch.conv2d( + x_27, + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_27 = ( + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_ + ) = None + input_5 = torch._C._nn.avg_pool2d(input_4, 2, 2, 0, False, True, None) + input_4 = None + concated_features_6 = torch.cat([input_5], 1) + x_28 = torch.nn.functional.batch_norm( + concated_features_6, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_6 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_29 = torch.nn.functional.relu(x_28, inplace=True) + x_28 = None + bottleneck_output_6 = torch.conv2d( + x_29, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_29 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_30 = torch.nn.functional.batch_norm( + bottleneck_output_6, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_6 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_31 = torch.nn.functional.relu(x_30, inplace=True) + x_30 = None + new_features_6 = torch.conv2d( + x_31, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_31 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_7 = torch.cat([input_5, new_features_6], 1) + x_32 = torch.nn.functional.batch_norm( + concated_features_7, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_7 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_33 = torch.nn.functional.relu(x_32, inplace=True) + x_32 = None + bottleneck_output_7 = torch.conv2d( + x_33, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_33 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_34 = torch.nn.functional.batch_norm( + bottleneck_output_7, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_7 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_35 = torch.nn.functional.relu(x_34, inplace=True) + x_34 = None + new_features_7 = torch.conv2d( + x_35, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_35 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_8 = torch.cat([input_5, new_features_6, new_features_7], 1) + x_36 = torch.nn.functional.batch_norm( + concated_features_8, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_8 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_37 = torch.nn.functional.relu(x_36, inplace=True) + x_36 = None + bottleneck_output_8 = torch.conv2d( + x_37, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_37 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_38 = torch.nn.functional.batch_norm( + bottleneck_output_8, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_8 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_39 = torch.nn.functional.relu(x_38, inplace=True) + x_38 = None + new_features_8 = torch.conv2d( + x_39, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_39 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_9 = torch.cat( + [input_5, new_features_6, new_features_7, new_features_8], 1 + ) + x_40 = torch.nn.functional.batch_norm( + concated_features_9, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_9 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_41 = torch.nn.functional.relu(x_40, inplace=True) + x_40 = None + bottleneck_output_9 = torch.conv2d( + x_41, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_41 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_42 = torch.nn.functional.batch_norm( + bottleneck_output_9, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_9 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_43 = torch.nn.functional.relu(x_42, inplace=True) + x_42 = None + new_features_9 = torch.conv2d( + x_43, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_43 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_10 = torch.cat( + [input_5, new_features_6, new_features_7, new_features_8, new_features_9], 1 + ) + x_44 = torch.nn.functional.batch_norm( + concated_features_10, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_10 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_45 = torch.nn.functional.relu(x_44, inplace=True) + x_44 = None + bottleneck_output_10 = torch.conv2d( + x_45, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_45 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_46 = torch.nn.functional.batch_norm( + bottleneck_output_10, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_10 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_47 = torch.nn.functional.relu(x_46, inplace=True) + x_46 = None + new_features_10 = torch.conv2d( + x_47, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_47 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_11 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + ], + 1, + ) + x_48 = torch.nn.functional.batch_norm( + concated_features_11, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_11 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_49 = torch.nn.functional.relu(x_48, inplace=True) + x_48 = None + bottleneck_output_11 = torch.conv2d( + x_49, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_49 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_50 = torch.nn.functional.batch_norm( + bottleneck_output_11, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_11 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_51 = torch.nn.functional.relu(x_50, inplace=True) + x_50 = None + new_features_11 = torch.conv2d( + x_51, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_51 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_12 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + ], + 1, + ) + x_52 = torch.nn.functional.batch_norm( + concated_features_12, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_12 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_53 = torch.nn.functional.relu(x_52, inplace=True) + x_52 = None + bottleneck_output_12 = torch.conv2d( + x_53, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_53 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_54 = torch.nn.functional.batch_norm( + bottleneck_output_12, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_12 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_55 = torch.nn.functional.relu(x_54, inplace=True) + x_54 = None + new_features_12 = torch.conv2d( + x_55, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_55 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_13 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + ], + 1, + ) + x_56 = torch.nn.functional.batch_norm( + concated_features_13, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_13 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_57 = torch.nn.functional.relu(x_56, inplace=True) + x_56 = None + bottleneck_output_13 = torch.conv2d( + x_57, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_57 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_58 = torch.nn.functional.batch_norm( + bottleneck_output_13, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_13 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_59 = torch.nn.functional.relu(x_58, inplace=True) + x_58 = None + new_features_13 = torch.conv2d( + x_59, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_59 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_14 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + ], + 1, + ) + x_60 = torch.nn.functional.batch_norm( + concated_features_14, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_14 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_61 = torch.nn.functional.relu(x_60, inplace=True) + x_60 = None + bottleneck_output_14 = torch.conv2d( + x_61, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_61 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_62 = torch.nn.functional.batch_norm( + bottleneck_output_14, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_14 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_63 = torch.nn.functional.relu(x_62, inplace=True) + x_62 = None + new_features_14 = torch.conv2d( + x_63, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_63 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_15 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + ], + 1, + ) + x_64 = torch.nn.functional.batch_norm( + concated_features_15, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_15 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_65 = torch.nn.functional.relu(x_64, inplace=True) + x_64 = None + bottleneck_output_15 = torch.conv2d( + x_65, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_65 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_66 = torch.nn.functional.batch_norm( + bottleneck_output_15, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_15 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_67 = torch.nn.functional.relu(x_66, inplace=True) + x_66 = None + new_features_15 = torch.conv2d( + x_67, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_67 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_16 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + ], + 1, + ) + x_68 = torch.nn.functional.batch_norm( + concated_features_16, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_16 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_69 = torch.nn.functional.relu(x_68, inplace=True) + x_68 = None + bottleneck_output_16 = torch.conv2d( + x_69, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_69 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_70 = torch.nn.functional.batch_norm( + bottleneck_output_16, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_16 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_71 = torch.nn.functional.relu(x_70, inplace=True) + x_70 = None + new_features_16 = torch.conv2d( + x_71, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_71 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_17 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + new_features_16, + ], + 1, + ) + x_72 = torch.nn.functional.batch_norm( + concated_features_17, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_17 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_73 = torch.nn.functional.relu(x_72, inplace=True) + x_72 = None + bottleneck_output_17 = torch.conv2d( + x_73, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_73 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_74 = torch.nn.functional.batch_norm( + bottleneck_output_17, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_17 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_75 = torch.nn.functional.relu(x_74, inplace=True) + x_74 = None + new_features_17 = torch.conv2d( + x_75, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_75 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + input_6 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + new_features_16, + new_features_17, + ], + 1, + ) + input_5 = ( + new_features_6 + ) = ( + new_features_7 + ) = ( + new_features_8 + ) = ( + new_features_9 + ) = ( + new_features_10 + ) = ( + new_features_11 + ) = ( + new_features_12 + ) = ( + new_features_13 + ) = new_features_14 = new_features_15 = new_features_16 = new_features_17 = None + x_76 = torch.nn.functional.batch_norm( + input_6, + l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_6 = l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_ + ) = None + x_77 = torch.nn.functional.relu(x_76, inplace=True) + x_76 = None + input_7 = torch.conv2d( + x_77, + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_77 = ( + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_ + ) = None + input_8 = torch._C._nn.avg_pool2d(input_7, 2, 2, 0, False, True, None) + input_7 = None + concated_features_18 = torch.cat([input_8], 1) + x_78 = torch.nn.functional.batch_norm( + concated_features_18, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_18 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_79 = torch.nn.functional.relu(x_78, inplace=True) + x_78 = None + bottleneck_output_18 = torch.conv2d( + x_79, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_79 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_80 = torch.nn.functional.batch_norm( + bottleneck_output_18, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_18 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_81 = torch.nn.functional.relu(x_80, inplace=True) + x_80 = None + new_features_18 = torch.conv2d( + x_81, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_81 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_19 = torch.cat([input_8, new_features_18], 1) + x_82 = torch.nn.functional.batch_norm( + concated_features_19, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_19 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_83 = torch.nn.functional.relu(x_82, inplace=True) + x_82 = None + bottleneck_output_19 = torch.conv2d( + x_83, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_83 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_84 = torch.nn.functional.batch_norm( + bottleneck_output_19, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_19 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_85 = torch.nn.functional.relu(x_84, inplace=True) + x_84 = None + new_features_19 = torch.conv2d( + x_85, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_85 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_20 = torch.cat([input_8, new_features_18, new_features_19], 1) + x_86 = torch.nn.functional.batch_norm( + concated_features_20, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_20 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_87 = torch.nn.functional.relu(x_86, inplace=True) + x_86 = None + bottleneck_output_20 = torch.conv2d( + x_87, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_87 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_88 = torch.nn.functional.batch_norm( + bottleneck_output_20, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_20 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_89 = torch.nn.functional.relu(x_88, inplace=True) + x_88 = None + new_features_20 = torch.conv2d( + x_89, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_89 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_21 = torch.cat( + [input_8, new_features_18, new_features_19, new_features_20], 1 + ) + x_90 = torch.nn.functional.batch_norm( + concated_features_21, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_21 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_91 = torch.nn.functional.relu(x_90, inplace=True) + x_90 = None + bottleneck_output_21 = torch.conv2d( + x_91, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_91 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_92 = torch.nn.functional.batch_norm( + bottleneck_output_21, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_21 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_93 = torch.nn.functional.relu(x_92, inplace=True) + x_92 = None + new_features_21 = torch.conv2d( + x_93, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_93 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_22 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + ], + 1, + ) + x_94 = torch.nn.functional.batch_norm( + concated_features_22, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_22 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_95 = torch.nn.functional.relu(x_94, inplace=True) + x_94 = None + bottleneck_output_22 = torch.conv2d( + x_95, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_95 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_96 = torch.nn.functional.batch_norm( + bottleneck_output_22, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_22 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_97 = torch.nn.functional.relu(x_96, inplace=True) + x_96 = None + new_features_22 = torch.conv2d( + x_97, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_97 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_23 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + ], + 1, + ) + x_98 = torch.nn.functional.batch_norm( + concated_features_23, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_23 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_99 = torch.nn.functional.relu(x_98, inplace=True) + x_98 = None + bottleneck_output_23 = torch.conv2d( + x_99, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_99 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_100 = torch.nn.functional.batch_norm( + bottleneck_output_23, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_23 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_101 = torch.nn.functional.relu(x_100, inplace=True) + x_100 = None + new_features_23 = torch.conv2d( + x_101, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_101 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_24 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + ], + 1, + ) + x_102 = torch.nn.functional.batch_norm( + concated_features_24, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_24 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_103 = torch.nn.functional.relu(x_102, inplace=True) + x_102 = None + bottleneck_output_24 = torch.conv2d( + x_103, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_104 = torch.nn.functional.batch_norm( + bottleneck_output_24, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_24 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_105 = torch.nn.functional.relu(x_104, inplace=True) + x_104 = None + new_features_24 = torch.conv2d( + x_105, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_105 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_25 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + ], + 1, + ) + x_106 = torch.nn.functional.batch_norm( + concated_features_25, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_25 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_107 = torch.nn.functional.relu(x_106, inplace=True) + x_106 = None + bottleneck_output_25 = torch.conv2d( + x_107, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_107 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_108 = torch.nn.functional.batch_norm( + bottleneck_output_25, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_25 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_109 = torch.nn.functional.relu(x_108, inplace=True) + x_108 = None + new_features_25 = torch.conv2d( + x_109, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_109 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_26 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + ], + 1, + ) + x_110 = torch.nn.functional.batch_norm( + concated_features_26, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_26 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_111 = torch.nn.functional.relu(x_110, inplace=True) + x_110 = None + bottleneck_output_26 = torch.conv2d( + x_111, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_112 = torch.nn.functional.batch_norm( + bottleneck_output_26, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_26 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_113 = torch.nn.functional.relu(x_112, inplace=True) + x_112 = None + new_features_26 = torch.conv2d( + x_113, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_113 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_27 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + ], + 1, + ) + x_114 = torch.nn.functional.batch_norm( + concated_features_27, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_27 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_115 = torch.nn.functional.relu(x_114, inplace=True) + x_114 = None + bottleneck_output_27 = torch.conv2d( + x_115, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_116 = torch.nn.functional.batch_norm( + bottleneck_output_27, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_27 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_117 = torch.nn.functional.relu(x_116, inplace=True) + x_116 = None + new_features_27 = torch.conv2d( + x_117, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_117 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_28 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + ], + 1, + ) + x_118 = torch.nn.functional.batch_norm( + concated_features_28, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_28 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_119 = torch.nn.functional.relu(x_118, inplace=True) + x_118 = None + bottleneck_output_28 = torch.conv2d( + x_119, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_119 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_120 = torch.nn.functional.batch_norm( + bottleneck_output_28, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_28 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_121 = torch.nn.functional.relu(x_120, inplace=True) + x_120 = None + new_features_28 = torch.conv2d( + x_121, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_121 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_29 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + ], + 1, + ) + x_122 = torch.nn.functional.batch_norm( + concated_features_29, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_29 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_123 = torch.nn.functional.relu(x_122, inplace=True) + x_122 = None + bottleneck_output_29 = torch.conv2d( + x_123, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_123 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_124 = torch.nn.functional.batch_norm( + bottleneck_output_29, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_29 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_125 = torch.nn.functional.relu(x_124, inplace=True) + x_124 = None + new_features_29 = torch.conv2d( + x_125, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_125 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + concated_features_30 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + ], + 1, + ) + x_126 = torch.nn.functional.batch_norm( + concated_features_30, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_30 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ = (None) + x_127 = torch.nn.functional.relu(x_126, inplace=True) + x_126 = None + bottleneck_output_30 = torch.conv2d( + x_127, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_127 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ = (None) + x_128 = torch.nn.functional.batch_norm( + bottleneck_output_30, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_30 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ = (None) + x_129 = torch.nn.functional.relu(x_128, inplace=True) + x_128 = None + new_features_30 = torch.conv2d( + x_129, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_129 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ = (None) + concated_features_31 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + ], + 1, + ) + x_130 = torch.nn.functional.batch_norm( + concated_features_31, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_31 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ = (None) + x_131 = torch.nn.functional.relu(x_130, inplace=True) + x_130 = None + bottleneck_output_31 = torch.conv2d( + x_131, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_131 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ = (None) + x_132 = torch.nn.functional.batch_norm( + bottleneck_output_31, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_31 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ = (None) + x_133 = torch.nn.functional.relu(x_132, inplace=True) + x_132 = None + new_features_31 = torch.conv2d( + x_133, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_133 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ = (None) + concated_features_32 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + ], + 1, + ) + x_134 = torch.nn.functional.batch_norm( + concated_features_32, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_32 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ = (None) + x_135 = torch.nn.functional.relu(x_134, inplace=True) + x_134 = None + bottleneck_output_32 = torch.conv2d( + x_135, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_135 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ = (None) + x_136 = torch.nn.functional.batch_norm( + bottleneck_output_32, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_32 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ = (None) + x_137 = torch.nn.functional.relu(x_136, inplace=True) + x_136 = None + new_features_32 = torch.conv2d( + x_137, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_137 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ = (None) + concated_features_33 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + ], + 1, + ) + x_138 = torch.nn.functional.batch_norm( + concated_features_33, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_33 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ = (None) + x_139 = torch.nn.functional.relu(x_138, inplace=True) + x_138 = None + bottleneck_output_33 = torch.conv2d( + x_139, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_139 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ = (None) + x_140 = torch.nn.functional.batch_norm( + bottleneck_output_33, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_33 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ = (None) + x_141 = torch.nn.functional.relu(x_140, inplace=True) + x_140 = None + new_features_33 = torch.conv2d( + x_141, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_141 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ = (None) + concated_features_34 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + ], + 1, + ) + x_142 = torch.nn.functional.batch_norm( + concated_features_34, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_34 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ = (None) + x_143 = torch.nn.functional.relu(x_142, inplace=True) + x_142 = None + bottleneck_output_34 = torch.conv2d( + x_143, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_143 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ = (None) + x_144 = torch.nn.functional.batch_norm( + bottleneck_output_34, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_34 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ = (None) + x_145 = torch.nn.functional.relu(x_144, inplace=True) + x_144 = None + new_features_34 = torch.conv2d( + x_145, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_145 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ = (None) + concated_features_35 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + ], + 1, + ) + x_146 = torch.nn.functional.batch_norm( + concated_features_35, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_35 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ = (None) + x_147 = torch.nn.functional.relu(x_146, inplace=True) + x_146 = None + bottleneck_output_35 = torch.conv2d( + x_147, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_147 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ = (None) + x_148 = torch.nn.functional.batch_norm( + bottleneck_output_35, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_35 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ = (None) + x_149 = torch.nn.functional.relu(x_148, inplace=True) + x_148 = None + new_features_35 = torch.conv2d( + x_149, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_149 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ = (None) + concated_features_36 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + ], + 1, + ) + x_150 = torch.nn.functional.batch_norm( + concated_features_36, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_36 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ = (None) + x_151 = torch.nn.functional.relu(x_150, inplace=True) + x_150 = None + bottleneck_output_36 = torch.conv2d( + x_151, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_151 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ = (None) + x_152 = torch.nn.functional.batch_norm( + bottleneck_output_36, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_36 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ = (None) + x_153 = torch.nn.functional.relu(x_152, inplace=True) + x_152 = None + new_features_36 = torch.conv2d( + x_153, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_153 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ = (None) + concated_features_37 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + ], + 1, + ) + x_154 = torch.nn.functional.batch_norm( + concated_features_37, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_37 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ = (None) + x_155 = torch.nn.functional.relu(x_154, inplace=True) + x_154 = None + bottleneck_output_37 = torch.conv2d( + x_155, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_155 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ = (None) + x_156 = torch.nn.functional.batch_norm( + bottleneck_output_37, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_37 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ = (None) + x_157 = torch.nn.functional.relu(x_156, inplace=True) + x_156 = None + new_features_37 = torch.conv2d( + x_157, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_157 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ = (None) + concated_features_38 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + ], + 1, + ) + x_158 = torch.nn.functional.batch_norm( + concated_features_38, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_38 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ = (None) + x_159 = torch.nn.functional.relu(x_158, inplace=True) + x_158 = None + bottleneck_output_38 = torch.conv2d( + x_159, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_159 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ = (None) + x_160 = torch.nn.functional.batch_norm( + bottleneck_output_38, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_38 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ = (None) + x_161 = torch.nn.functional.relu(x_160, inplace=True) + x_160 = None + new_features_38 = torch.conv2d( + x_161, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_161 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ = (None) + concated_features_39 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + ], + 1, + ) + x_162 = torch.nn.functional.batch_norm( + concated_features_39, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_39 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ = (None) + x_163 = torch.nn.functional.relu(x_162, inplace=True) + x_162 = None + bottleneck_output_39 = torch.conv2d( + x_163, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_163 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ = (None) + x_164 = torch.nn.functional.batch_norm( + bottleneck_output_39, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_39 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ = (None) + x_165 = torch.nn.functional.relu(x_164, inplace=True) + x_164 = None + new_features_39 = torch.conv2d( + x_165, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_165 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ = (None) + concated_features_40 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + ], + 1, + ) + x_166 = torch.nn.functional.batch_norm( + concated_features_40, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_40 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ = (None) + x_167 = torch.nn.functional.relu(x_166, inplace=True) + x_166 = None + bottleneck_output_40 = torch.conv2d( + x_167, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_167 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ = (None) + x_168 = torch.nn.functional.batch_norm( + bottleneck_output_40, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_40 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ = (None) + x_169 = torch.nn.functional.relu(x_168, inplace=True) + x_168 = None + new_features_40 = torch.conv2d( + x_169, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_169 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ = (None) + concated_features_41 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + ], + 1, + ) + x_170 = torch.nn.functional.batch_norm( + concated_features_41, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_41 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ = (None) + x_171 = torch.nn.functional.relu(x_170, inplace=True) + x_170 = None + bottleneck_output_41 = torch.conv2d( + x_171, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_171 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ = (None) + x_172 = torch.nn.functional.batch_norm( + bottleneck_output_41, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_41 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ = (None) + x_173 = torch.nn.functional.relu(x_172, inplace=True) + x_172 = None + new_features_41 = torch.conv2d( + x_173, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_173 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ = (None) + input_9 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + ], + 1, + ) + input_8 = ( + new_features_18 + ) = ( + new_features_19 + ) = ( + new_features_20 + ) = ( + new_features_21 + ) = ( + new_features_22 + ) = ( + new_features_23 + ) = ( + new_features_24 + ) = ( + new_features_25 + ) = ( + new_features_26 + ) = ( + new_features_27 + ) = ( + new_features_28 + ) = ( + new_features_29 + ) = ( + new_features_30 + ) = ( + new_features_31 + ) = ( + new_features_32 + ) = ( + new_features_33 + ) = ( + new_features_34 + ) = ( + new_features_35 + ) = ( + new_features_36 + ) = ( + new_features_37 + ) = new_features_38 = new_features_39 = new_features_40 = new_features_41 = None + x_174 = torch.nn.functional.batch_norm( + input_9, + l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_9 = l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_ + ) = None + x_175 = torch.nn.functional.relu(x_174, inplace=True) + x_174 = None + input_10 = torch.conv2d( + x_175, + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_175 = ( + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_ + ) = None + input_11 = torch._C._nn.avg_pool2d(input_10, 2, 2, 0, False, True, None) + input_10 = None + concated_features_42 = torch.cat([input_11], 1) + x_176 = torch.nn.functional.batch_norm( + concated_features_42, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_42 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_177 = torch.nn.functional.relu(x_176, inplace=True) + x_176 = None + bottleneck_output_42 = torch.conv2d( + x_177, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_177 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_178 = torch.nn.functional.batch_norm( + bottleneck_output_42, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_42 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_179 = torch.nn.functional.relu(x_178, inplace=True) + x_178 = None + new_features_42 = torch.conv2d( + x_179, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_179 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_43 = torch.cat([input_11, new_features_42], 1) + x_180 = torch.nn.functional.batch_norm( + concated_features_43, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_43 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_181 = torch.nn.functional.relu(x_180, inplace=True) + x_180 = None + bottleneck_output_43 = torch.conv2d( + x_181, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_181 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_182 = torch.nn.functional.batch_norm( + bottleneck_output_43, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_43 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_183 = torch.nn.functional.relu(x_182, inplace=True) + x_182 = None + new_features_43 = torch.conv2d( + x_183, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_183 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_44 = torch.cat( + [input_11, new_features_42, new_features_43], 1 + ) + x_184 = torch.nn.functional.batch_norm( + concated_features_44, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_44 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_185 = torch.nn.functional.relu(x_184, inplace=True) + x_184 = None + bottleneck_output_44 = torch.conv2d( + x_185, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_185 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_186 = torch.nn.functional.batch_norm( + bottleneck_output_44, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_44 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_187 = torch.nn.functional.relu(x_186, inplace=True) + x_186 = None + new_features_44 = torch.conv2d( + x_187, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_187 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_45 = torch.cat( + [input_11, new_features_42, new_features_43, new_features_44], 1 + ) + x_188 = torch.nn.functional.batch_norm( + concated_features_45, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_45 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_189 = torch.nn.functional.relu(x_188, inplace=True) + x_188 = None + bottleneck_output_45 = torch.conv2d( + x_189, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_189 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_190 = torch.nn.functional.batch_norm( + bottleneck_output_45, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_45 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_191 = torch.nn.functional.relu(x_190, inplace=True) + x_190 = None + new_features_45 = torch.conv2d( + x_191, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_191 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_46 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + ], + 1, + ) + x_192 = torch.nn.functional.batch_norm( + concated_features_46, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_46 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_193 = torch.nn.functional.relu(x_192, inplace=True) + x_192 = None + bottleneck_output_46 = torch.conv2d( + x_193, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_193 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_194 = torch.nn.functional.batch_norm( + bottleneck_output_46, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_46 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_195 = torch.nn.functional.relu(x_194, inplace=True) + x_194 = None + new_features_46 = torch.conv2d( + x_195, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_195 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_47 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + ], + 1, + ) + x_196 = torch.nn.functional.batch_norm( + concated_features_47, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_47 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_197 = torch.nn.functional.relu(x_196, inplace=True) + x_196 = None + bottleneck_output_47 = torch.conv2d( + x_197, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_197 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_198 = torch.nn.functional.batch_norm( + bottleneck_output_47, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_47 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_199 = torch.nn.functional.relu(x_198, inplace=True) + x_198 = None + new_features_47 = torch.conv2d( + x_199, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_199 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_48 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + ], + 1, + ) + x_200 = torch.nn.functional.batch_norm( + concated_features_48, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_48 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_201 = torch.nn.functional.relu(x_200, inplace=True) + x_200 = None + bottleneck_output_48 = torch.conv2d( + x_201, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_201 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_202 = torch.nn.functional.batch_norm( + bottleneck_output_48, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_48 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_203 = torch.nn.functional.relu(x_202, inplace=True) + x_202 = None + new_features_48 = torch.conv2d( + x_203, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_203 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_49 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + ], + 1, + ) + x_204 = torch.nn.functional.batch_norm( + concated_features_49, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_49 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_205 = torch.nn.functional.relu(x_204, inplace=True) + x_204 = None + bottleneck_output_49 = torch.conv2d( + x_205, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_205 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_206 = torch.nn.functional.batch_norm( + bottleneck_output_49, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_49 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_207 = torch.nn.functional.relu(x_206, inplace=True) + x_206 = None + new_features_49 = torch.conv2d( + x_207, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_207 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_50 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + ], + 1, + ) + x_208 = torch.nn.functional.batch_norm( + concated_features_50, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_50 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_209 = torch.nn.functional.relu(x_208, inplace=True) + x_208 = None + bottleneck_output_50 = torch.conv2d( + x_209, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_209 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_210 = torch.nn.functional.batch_norm( + bottleneck_output_50, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_50 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_211 = torch.nn.functional.relu(x_210, inplace=True) + x_210 = None + new_features_50 = torch.conv2d( + x_211, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_211 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_51 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + ], + 1, + ) + x_212 = torch.nn.functional.batch_norm( + concated_features_51, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_51 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_213 = torch.nn.functional.relu(x_212, inplace=True) + x_212 = None + bottleneck_output_51 = torch.conv2d( + x_213, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_213 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_214 = torch.nn.functional.batch_norm( + bottleneck_output_51, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_51 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_215 = torch.nn.functional.relu(x_214, inplace=True) + x_214 = None + new_features_51 = torch.conv2d( + x_215, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_215 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_52 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + ], + 1, + ) + x_216 = torch.nn.functional.batch_norm( + concated_features_52, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_52 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_217 = torch.nn.functional.relu(x_216, inplace=True) + x_216 = None + bottleneck_output_52 = torch.conv2d( + x_217, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_217 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_218 = torch.nn.functional.batch_norm( + bottleneck_output_52, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_52 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_219 = torch.nn.functional.relu(x_218, inplace=True) + x_218 = None + new_features_52 = torch.conv2d( + x_219, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_219 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_53 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + ], + 1, + ) + x_220 = torch.nn.functional.batch_norm( + concated_features_53, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_53 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_221 = torch.nn.functional.relu(x_220, inplace=True) + x_220 = None + bottleneck_output_53 = torch.conv2d( + x_221, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_221 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_222 = torch.nn.functional.batch_norm( + bottleneck_output_53, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_53 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_223 = torch.nn.functional.relu(x_222, inplace=True) + x_222 = None + new_features_53 = torch.conv2d( + x_223, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_223 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + concated_features_54 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + ], + 1, + ) + x_224 = torch.nn.functional.batch_norm( + concated_features_54, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_54 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ = (None) + x_225 = torch.nn.functional.relu(x_224, inplace=True) + x_224 = None + bottleneck_output_54 = torch.conv2d( + x_225, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_225 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ = (None) + x_226 = torch.nn.functional.batch_norm( + bottleneck_output_54, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_54 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ = (None) + x_227 = torch.nn.functional.relu(x_226, inplace=True) + x_226 = None + new_features_54 = torch.conv2d( + x_227, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_227 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ = (None) + concated_features_55 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + ], + 1, + ) + x_228 = torch.nn.functional.batch_norm( + concated_features_55, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_55 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ = (None) + x_229 = torch.nn.functional.relu(x_228, inplace=True) + x_228 = None + bottleneck_output_55 = torch.conv2d( + x_229, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_229 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ = (None) + x_230 = torch.nn.functional.batch_norm( + bottleneck_output_55, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_55 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ = (None) + x_231 = torch.nn.functional.relu(x_230, inplace=True) + x_230 = None + new_features_55 = torch.conv2d( + x_231, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_231 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ = (None) + concated_features_56 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + ], + 1, + ) + x_232 = torch.nn.functional.batch_norm( + concated_features_56, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_56 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ = (None) + x_233 = torch.nn.functional.relu(x_232, inplace=True) + x_232 = None + bottleneck_output_56 = torch.conv2d( + x_233, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_233 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ = (None) + x_234 = torch.nn.functional.batch_norm( + bottleneck_output_56, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_56 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ = (None) + x_235 = torch.nn.functional.relu(x_234, inplace=True) + x_234 = None + new_features_56 = torch.conv2d( + x_235, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_235 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ = (None) + concated_features_57 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + ], + 1, + ) + x_236 = torch.nn.functional.batch_norm( + concated_features_57, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_57 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ = (None) + x_237 = torch.nn.functional.relu(x_236, inplace=True) + x_236 = None + bottleneck_output_57 = torch.conv2d( + x_237, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_237 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ = (None) + x_238 = torch.nn.functional.batch_norm( + bottleneck_output_57, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_57 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ = (None) + x_239 = torch.nn.functional.relu(x_238, inplace=True) + x_238 = None + new_features_57 = torch.conv2d( + x_239, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_239 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ = (None) + input_12 = torch.cat( + [ + input_11, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + ], + 1, + ) + input_11 = ( + new_features_42 + ) = ( + new_features_43 + ) = ( + new_features_44 + ) = ( + new_features_45 + ) = ( + new_features_46 + ) = ( + new_features_47 + ) = ( + new_features_48 + ) = ( + new_features_49 + ) = ( + new_features_50 + ) = ( + new_features_51 + ) = ( + new_features_52 + ) = ( + new_features_53 + ) = new_features_54 = new_features_55 = new_features_56 = new_features_57 = None + x_240 = torch.nn.functional.batch_norm( + input_12, + l_self_modules_features_modules_norm5_buffers_running_mean_, + l_self_modules_features_modules_norm5_buffers_running_var_, + l_self_modules_features_modules_norm5_parameters_weight_, + l_self_modules_features_modules_norm5_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_12 = ( + l_self_modules_features_modules_norm5_buffers_running_mean_ + ) = ( + l_self_modules_features_modules_norm5_buffers_running_var_ + ) = ( + l_self_modules_features_modules_norm5_parameters_weight_ + ) = l_self_modules_features_modules_norm5_parameters_bias_ = None + x_241 = torch.nn.functional.relu(x_240, inplace=True) + x_240 = None + x_242 = torch.nn.functional.adaptive_avg_pool2d(x_241, 1) + x_241 = None + x_243 = x_242.flatten(1, -1) + x_242 = None + x_244 = torch.nn.functional.dropout(x_243, 0.0, False, False) + x_243 = None + x_245 = torch._C._nn.linear( + x_244, + l_self_modules_classifier_parameters_weight_, + l_self_modules_classifier_parameters_bias_, + ) + x_244 = ( + l_self_modules_classifier_parameters_weight_ + ) = l_self_modules_classifier_parameters_bias_ = None + return (x_245,) diff --git a/samples/timm/densenet121.ra_in1k/weight_meta.py b/samples/timm/densenet121.ra_in1k/weight_meta.py new file mode 100644 index 0000000000..7d3a290ff6 --- /dev/null +++ b/samples/timm/densenet121.ra_in1k/weight_meta.py @@ -0,0 +1,6090 @@ +class Program_weight_tensor_meta_L_self_modules_features_modules_conv0_parameters_weight_: + name = "L_self_modules_features_modules_conv0_parameters_weight_" + shape = [64, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.002 + std = 0.117 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.224 + std = 1.287 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_buffers_running_mean_: + name = "L_self_modules_features_modules_norm0_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_buffers_running_var_: + name = "L_self_modules_features_modules_norm0_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_parameters_weight_: + name = "L_self_modules_features_modules_norm0_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_parameters_bias_: + name = "L_self_modules_features_modules_norm0_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.143 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.101 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 224, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.094 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition1_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.103 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 224, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.094 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 288, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.084 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [128, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [128, 352, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.076 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [128, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [128, 416, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.069 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [128, 448, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.067 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [128, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition2_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition2_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition2_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 288, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 352, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.076 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 416, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.069 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [128, 448, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.067 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [128, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [128, 544, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.061 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [128, 576, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [128, 608, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.057 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_" + shape = [128, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_" + shape = [128, 672, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.054 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_" + shape = [128, 704, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_" + shape = [128, 736, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.052 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_" + shape = [128, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_" + shape = [128, 800, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.050 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_" + shape = [128, 832, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.049 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_" + shape = [128, 864, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_" + shape = [128, 896, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_" + shape = [128, 928, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_" + shape = [128, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_" + shape = [128, 992, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.045 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition3_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition3_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition3_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 544, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.060 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 576, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 608, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.057 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 672, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.054 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [128, 704, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [128, 736, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.052 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [128, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [128, 800, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.050 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [128, 832, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.049 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [128, 864, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_" + shape = [128, 896, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_" + shape = [128, 928, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_" + shape = [128, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_" + shape = [128, 992, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.045 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_buffers_running_mean_: + name = "L_self_modules_features_modules_norm5_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_buffers_running_var_: + name = "L_self_modules_features_modules_norm5_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_parameters_weight_: + name = "L_self_modules_features_modules_norm5_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_parameters_bias_: + name = "L_self_modules_features_modules_norm5_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_classifier_parameters_weight_: + name = "L_self_modules_classifier_parameters_weight_" + shape = [1000, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_classifier_parameters_bias_: + name = "L_self_modules_classifier_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/densenet161.tv_in1k/graph_hash.txt b/samples/timm/densenet161.tv_in1k/graph_hash.txt new file mode 100644 index 0000000000..5b716bb927 --- /dev/null +++ b/samples/timm/densenet161.tv_in1k/graph_hash.txt @@ -0,0 +1 @@ +4cef2f10705d139d718cf68cff2ac72ea0f705b85b19a9b403273b7308adb826 \ No newline at end of file diff --git a/samples/timm/densenet161.tv_in1k/graph_net.json b/samples/timm/densenet161.tv_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/densenet161.tv_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/densenet161.tv_in1k/input_meta.py b/samples/timm/densenet161.tv_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/densenet161.tv_in1k/input_tensor_constraints.py b/samples/timm/densenet161.tv_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/densenet161.tv_in1k/model.py b/samples/timm/densenet161.tv_in1k/model.py new file mode 100644 index 0000000000..b0f34c0e74 --- /dev/null +++ b/samples/timm/densenet161.tv_in1k/model.py @@ -0,0 +1,6987 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_features_modules_conv0_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_features_modules_norm0_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_norm0_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_norm0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm5_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_norm5_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_norm5_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm5_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_classifier_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_classifier_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_features_modules_conv0_parameters_weight_ = ( + L_self_modules_features_modules_conv0_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_features_modules_norm0_buffers_running_mean_ = ( + L_self_modules_features_modules_norm0_buffers_running_mean_ + ) + l_self_modules_features_modules_norm0_buffers_running_var_ = ( + L_self_modules_features_modules_norm0_buffers_running_var_ + ) + l_self_modules_features_modules_norm0_parameters_weight_ = ( + L_self_modules_features_modules_norm0_parameters_weight_ + ) + l_self_modules_features_modules_norm0_parameters_bias_ = ( + L_self_modules_features_modules_norm0_parameters_bias_ + ) + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition1_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition1_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition1_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition2_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition2_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition2_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition3_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition3_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition3_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_ + l_self_modules_features_modules_norm5_buffers_running_mean_ = ( + L_self_modules_features_modules_norm5_buffers_running_mean_ + ) + l_self_modules_features_modules_norm5_buffers_running_var_ = ( + L_self_modules_features_modules_norm5_buffers_running_var_ + ) + l_self_modules_features_modules_norm5_parameters_weight_ = ( + L_self_modules_features_modules_norm5_parameters_weight_ + ) + l_self_modules_features_modules_norm5_parameters_bias_ = ( + L_self_modules_features_modules_norm5_parameters_bias_ + ) + l_self_modules_classifier_parameters_weight_ = ( + L_self_modules_classifier_parameters_weight_ + ) + l_self_modules_classifier_parameters_bias_ = ( + L_self_modules_classifier_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_features_modules_conv0_parameters_weight_, + None, + (2, 2), + (3, 3), + (1, 1), + 1, + ) + l_x_ = l_self_modules_features_modules_conv0_parameters_weight_ = None + x = torch.nn.functional.batch_norm( + input_1, + l_self_modules_features_modules_norm0_buffers_running_mean_, + l_self_modules_features_modules_norm0_buffers_running_var_, + l_self_modules_features_modules_norm0_parameters_weight_, + l_self_modules_features_modules_norm0_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_1 = ( + l_self_modules_features_modules_norm0_buffers_running_mean_ + ) = ( + l_self_modules_features_modules_norm0_buffers_running_var_ + ) = ( + l_self_modules_features_modules_norm0_parameters_weight_ + ) = l_self_modules_features_modules_norm0_parameters_bias_ = None + x_1 = torch.nn.functional.relu(x, inplace=True) + x = None + input_2 = torch.nn.functional.max_pool2d( + x_1, 3, 2, 1, 1, ceil_mode=False, return_indices=False + ) + x_1 = None + concated_features = torch.cat([input_2], 1) + x_2 = torch.nn.functional.batch_norm( + concated_features, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_3 = torch.nn.functional.relu(x_2, inplace=True) + x_2 = None + bottleneck_output = torch.conv2d( + x_3, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_3 = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + bottleneck_output, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_5 = torch.nn.functional.relu(x_4, inplace=True) + x_4 = None + new_features = torch.conv2d( + x_5, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_5 = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_1 = torch.cat([input_2, new_features], 1) + x_6 = torch.nn.functional.batch_norm( + concated_features_1, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_1 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_7 = torch.nn.functional.relu(x_6, inplace=True) + x_6 = None + bottleneck_output_1 = torch.conv2d( + x_7, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_7 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_8 = torch.nn.functional.batch_norm( + bottleneck_output_1, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_1 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_9 = torch.nn.functional.relu(x_8, inplace=True) + x_8 = None + new_features_1 = torch.conv2d( + x_9, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_9 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_2 = torch.cat([input_2, new_features, new_features_1], 1) + x_10 = torch.nn.functional.batch_norm( + concated_features_2, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_2 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_11 = torch.nn.functional.relu(x_10, inplace=True) + x_10 = None + bottleneck_output_2 = torch.conv2d( + x_11, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_11 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_12 = torch.nn.functional.batch_norm( + bottleneck_output_2, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_2 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_13 = torch.nn.functional.relu(x_12, inplace=True) + x_12 = None + new_features_2 = torch.conv2d( + x_13, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_13 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_3 = torch.cat( + [input_2, new_features, new_features_1, new_features_2], 1 + ) + x_14 = torch.nn.functional.batch_norm( + concated_features_3, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_3 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_15 = torch.nn.functional.relu(x_14, inplace=True) + x_14 = None + bottleneck_output_3 = torch.conv2d( + x_15, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_15 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_16 = torch.nn.functional.batch_norm( + bottleneck_output_3, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_3 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_17 = torch.nn.functional.relu(x_16, inplace=True) + x_16 = None + new_features_3 = torch.conv2d( + x_17, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_17 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_4 = torch.cat( + [input_2, new_features, new_features_1, new_features_2, new_features_3], 1 + ) + x_18 = torch.nn.functional.batch_norm( + concated_features_4, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_4 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_19 = torch.nn.functional.relu(x_18, inplace=True) + x_18 = None + bottleneck_output_4 = torch.conv2d( + x_19, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_19 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_20 = torch.nn.functional.batch_norm( + bottleneck_output_4, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_4 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_21 = torch.nn.functional.relu(x_20, inplace=True) + x_20 = None + new_features_4 = torch.conv2d( + x_21, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_21 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_5 = torch.cat( + [ + input_2, + new_features, + new_features_1, + new_features_2, + new_features_3, + new_features_4, + ], + 1, + ) + x_22 = torch.nn.functional.batch_norm( + concated_features_5, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_5 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_23 = torch.nn.functional.relu(x_22, inplace=True) + x_22 = None + bottleneck_output_5 = torch.conv2d( + x_23, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_24 = torch.nn.functional.batch_norm( + bottleneck_output_5, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_5 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_25 = torch.nn.functional.relu(x_24, inplace=True) + x_24 = None + new_features_5 = torch.conv2d( + x_25, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_25 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + input_3 = torch.cat( + [ + input_2, + new_features, + new_features_1, + new_features_2, + new_features_3, + new_features_4, + new_features_5, + ], + 1, + ) + input_2 = ( + new_features + ) = ( + new_features_1 + ) = new_features_2 = new_features_3 = new_features_4 = new_features_5 = None + x_26 = torch.nn.functional.batch_norm( + input_3, + l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_3 = l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_ + ) = None + x_27 = torch.nn.functional.relu(x_26, inplace=True) + x_26 = None + input_4 = torch.conv2d( + x_27, + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_27 = ( + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_ + ) = None + input_5 = torch._C._nn.avg_pool2d(input_4, 2, 2, 0, False, True, None) + input_4 = None + concated_features_6 = torch.cat([input_5], 1) + x_28 = torch.nn.functional.batch_norm( + concated_features_6, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_6 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_29 = torch.nn.functional.relu(x_28, inplace=True) + x_28 = None + bottleneck_output_6 = torch.conv2d( + x_29, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_29 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_30 = torch.nn.functional.batch_norm( + bottleneck_output_6, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_6 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_31 = torch.nn.functional.relu(x_30, inplace=True) + x_30 = None + new_features_6 = torch.conv2d( + x_31, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_31 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_7 = torch.cat([input_5, new_features_6], 1) + x_32 = torch.nn.functional.batch_norm( + concated_features_7, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_7 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_33 = torch.nn.functional.relu(x_32, inplace=True) + x_32 = None + bottleneck_output_7 = torch.conv2d( + x_33, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_33 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_34 = torch.nn.functional.batch_norm( + bottleneck_output_7, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_7 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_35 = torch.nn.functional.relu(x_34, inplace=True) + x_34 = None + new_features_7 = torch.conv2d( + x_35, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_35 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_8 = torch.cat([input_5, new_features_6, new_features_7], 1) + x_36 = torch.nn.functional.batch_norm( + concated_features_8, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_8 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_37 = torch.nn.functional.relu(x_36, inplace=True) + x_36 = None + bottleneck_output_8 = torch.conv2d( + x_37, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_37 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_38 = torch.nn.functional.batch_norm( + bottleneck_output_8, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_8 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_39 = torch.nn.functional.relu(x_38, inplace=True) + x_38 = None + new_features_8 = torch.conv2d( + x_39, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_39 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_9 = torch.cat( + [input_5, new_features_6, new_features_7, new_features_8], 1 + ) + x_40 = torch.nn.functional.batch_norm( + concated_features_9, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_9 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_41 = torch.nn.functional.relu(x_40, inplace=True) + x_40 = None + bottleneck_output_9 = torch.conv2d( + x_41, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_41 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_42 = torch.nn.functional.batch_norm( + bottleneck_output_9, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_9 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_43 = torch.nn.functional.relu(x_42, inplace=True) + x_42 = None + new_features_9 = torch.conv2d( + x_43, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_43 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_10 = torch.cat( + [input_5, new_features_6, new_features_7, new_features_8, new_features_9], 1 + ) + x_44 = torch.nn.functional.batch_norm( + concated_features_10, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_10 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_45 = torch.nn.functional.relu(x_44, inplace=True) + x_44 = None + bottleneck_output_10 = torch.conv2d( + x_45, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_45 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_46 = torch.nn.functional.batch_norm( + bottleneck_output_10, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_10 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_47 = torch.nn.functional.relu(x_46, inplace=True) + x_46 = None + new_features_10 = torch.conv2d( + x_47, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_47 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_11 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + ], + 1, + ) + x_48 = torch.nn.functional.batch_norm( + concated_features_11, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_11 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_49 = torch.nn.functional.relu(x_48, inplace=True) + x_48 = None + bottleneck_output_11 = torch.conv2d( + x_49, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_49 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_50 = torch.nn.functional.batch_norm( + bottleneck_output_11, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_11 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_51 = torch.nn.functional.relu(x_50, inplace=True) + x_50 = None + new_features_11 = torch.conv2d( + x_51, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_51 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_12 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + ], + 1, + ) + x_52 = torch.nn.functional.batch_norm( + concated_features_12, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_12 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_53 = torch.nn.functional.relu(x_52, inplace=True) + x_52 = None + bottleneck_output_12 = torch.conv2d( + x_53, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_53 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_54 = torch.nn.functional.batch_norm( + bottleneck_output_12, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_12 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_55 = torch.nn.functional.relu(x_54, inplace=True) + x_54 = None + new_features_12 = torch.conv2d( + x_55, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_55 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_13 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + ], + 1, + ) + x_56 = torch.nn.functional.batch_norm( + concated_features_13, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_13 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_57 = torch.nn.functional.relu(x_56, inplace=True) + x_56 = None + bottleneck_output_13 = torch.conv2d( + x_57, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_57 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_58 = torch.nn.functional.batch_norm( + bottleneck_output_13, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_13 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_59 = torch.nn.functional.relu(x_58, inplace=True) + x_58 = None + new_features_13 = torch.conv2d( + x_59, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_59 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_14 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + ], + 1, + ) + x_60 = torch.nn.functional.batch_norm( + concated_features_14, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_14 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_61 = torch.nn.functional.relu(x_60, inplace=True) + x_60 = None + bottleneck_output_14 = torch.conv2d( + x_61, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_61 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_62 = torch.nn.functional.batch_norm( + bottleneck_output_14, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_14 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_63 = torch.nn.functional.relu(x_62, inplace=True) + x_62 = None + new_features_14 = torch.conv2d( + x_63, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_63 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_15 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + ], + 1, + ) + x_64 = torch.nn.functional.batch_norm( + concated_features_15, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_15 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_65 = torch.nn.functional.relu(x_64, inplace=True) + x_64 = None + bottleneck_output_15 = torch.conv2d( + x_65, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_65 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_66 = torch.nn.functional.batch_norm( + bottleneck_output_15, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_15 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_67 = torch.nn.functional.relu(x_66, inplace=True) + x_66 = None + new_features_15 = torch.conv2d( + x_67, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_67 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_16 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + ], + 1, + ) + x_68 = torch.nn.functional.batch_norm( + concated_features_16, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_16 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_69 = torch.nn.functional.relu(x_68, inplace=True) + x_68 = None + bottleneck_output_16 = torch.conv2d( + x_69, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_69 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_70 = torch.nn.functional.batch_norm( + bottleneck_output_16, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_16 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_71 = torch.nn.functional.relu(x_70, inplace=True) + x_70 = None + new_features_16 = torch.conv2d( + x_71, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_71 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_17 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + new_features_16, + ], + 1, + ) + x_72 = torch.nn.functional.batch_norm( + concated_features_17, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_17 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_73 = torch.nn.functional.relu(x_72, inplace=True) + x_72 = None + bottleneck_output_17 = torch.conv2d( + x_73, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_73 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_74 = torch.nn.functional.batch_norm( + bottleneck_output_17, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_17 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_75 = torch.nn.functional.relu(x_74, inplace=True) + x_74 = None + new_features_17 = torch.conv2d( + x_75, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_75 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + input_6 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + new_features_16, + new_features_17, + ], + 1, + ) + input_5 = ( + new_features_6 + ) = ( + new_features_7 + ) = ( + new_features_8 + ) = ( + new_features_9 + ) = ( + new_features_10 + ) = ( + new_features_11 + ) = ( + new_features_12 + ) = ( + new_features_13 + ) = new_features_14 = new_features_15 = new_features_16 = new_features_17 = None + x_76 = torch.nn.functional.batch_norm( + input_6, + l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_6 = l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_ + ) = None + x_77 = torch.nn.functional.relu(x_76, inplace=True) + x_76 = None + input_7 = torch.conv2d( + x_77, + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_77 = ( + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_ + ) = None + input_8 = torch._C._nn.avg_pool2d(input_7, 2, 2, 0, False, True, None) + input_7 = None + concated_features_18 = torch.cat([input_8], 1) + x_78 = torch.nn.functional.batch_norm( + concated_features_18, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_18 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_79 = torch.nn.functional.relu(x_78, inplace=True) + x_78 = None + bottleneck_output_18 = torch.conv2d( + x_79, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_79 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_80 = torch.nn.functional.batch_norm( + bottleneck_output_18, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_18 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_81 = torch.nn.functional.relu(x_80, inplace=True) + x_80 = None + new_features_18 = torch.conv2d( + x_81, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_81 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_19 = torch.cat([input_8, new_features_18], 1) + x_82 = torch.nn.functional.batch_norm( + concated_features_19, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_19 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_83 = torch.nn.functional.relu(x_82, inplace=True) + x_82 = None + bottleneck_output_19 = torch.conv2d( + x_83, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_83 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_84 = torch.nn.functional.batch_norm( + bottleneck_output_19, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_19 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_85 = torch.nn.functional.relu(x_84, inplace=True) + x_84 = None + new_features_19 = torch.conv2d( + x_85, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_85 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_20 = torch.cat([input_8, new_features_18, new_features_19], 1) + x_86 = torch.nn.functional.batch_norm( + concated_features_20, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_20 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_87 = torch.nn.functional.relu(x_86, inplace=True) + x_86 = None + bottleneck_output_20 = torch.conv2d( + x_87, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_87 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_88 = torch.nn.functional.batch_norm( + bottleneck_output_20, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_20 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_89 = torch.nn.functional.relu(x_88, inplace=True) + x_88 = None + new_features_20 = torch.conv2d( + x_89, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_89 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_21 = torch.cat( + [input_8, new_features_18, new_features_19, new_features_20], 1 + ) + x_90 = torch.nn.functional.batch_norm( + concated_features_21, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_21 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_91 = torch.nn.functional.relu(x_90, inplace=True) + x_90 = None + bottleneck_output_21 = torch.conv2d( + x_91, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_91 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_92 = torch.nn.functional.batch_norm( + bottleneck_output_21, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_21 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_93 = torch.nn.functional.relu(x_92, inplace=True) + x_92 = None + new_features_21 = torch.conv2d( + x_93, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_93 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_22 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + ], + 1, + ) + x_94 = torch.nn.functional.batch_norm( + concated_features_22, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_22 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_95 = torch.nn.functional.relu(x_94, inplace=True) + x_94 = None + bottleneck_output_22 = torch.conv2d( + x_95, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_95 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_96 = torch.nn.functional.batch_norm( + bottleneck_output_22, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_22 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_97 = torch.nn.functional.relu(x_96, inplace=True) + x_96 = None + new_features_22 = torch.conv2d( + x_97, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_97 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_23 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + ], + 1, + ) + x_98 = torch.nn.functional.batch_norm( + concated_features_23, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_23 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_99 = torch.nn.functional.relu(x_98, inplace=True) + x_98 = None + bottleneck_output_23 = torch.conv2d( + x_99, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_99 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_100 = torch.nn.functional.batch_norm( + bottleneck_output_23, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_23 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_101 = torch.nn.functional.relu(x_100, inplace=True) + x_100 = None + new_features_23 = torch.conv2d( + x_101, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_101 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_24 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + ], + 1, + ) + x_102 = torch.nn.functional.batch_norm( + concated_features_24, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_24 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_103 = torch.nn.functional.relu(x_102, inplace=True) + x_102 = None + bottleneck_output_24 = torch.conv2d( + x_103, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_104 = torch.nn.functional.batch_norm( + bottleneck_output_24, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_24 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_105 = torch.nn.functional.relu(x_104, inplace=True) + x_104 = None + new_features_24 = torch.conv2d( + x_105, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_105 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_25 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + ], + 1, + ) + x_106 = torch.nn.functional.batch_norm( + concated_features_25, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_25 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_107 = torch.nn.functional.relu(x_106, inplace=True) + x_106 = None + bottleneck_output_25 = torch.conv2d( + x_107, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_107 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_108 = torch.nn.functional.batch_norm( + bottleneck_output_25, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_25 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_109 = torch.nn.functional.relu(x_108, inplace=True) + x_108 = None + new_features_25 = torch.conv2d( + x_109, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_109 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_26 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + ], + 1, + ) + x_110 = torch.nn.functional.batch_norm( + concated_features_26, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_26 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_111 = torch.nn.functional.relu(x_110, inplace=True) + x_110 = None + bottleneck_output_26 = torch.conv2d( + x_111, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_112 = torch.nn.functional.batch_norm( + bottleneck_output_26, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_26 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_113 = torch.nn.functional.relu(x_112, inplace=True) + x_112 = None + new_features_26 = torch.conv2d( + x_113, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_113 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_27 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + ], + 1, + ) + x_114 = torch.nn.functional.batch_norm( + concated_features_27, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_27 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_115 = torch.nn.functional.relu(x_114, inplace=True) + x_114 = None + bottleneck_output_27 = torch.conv2d( + x_115, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_116 = torch.nn.functional.batch_norm( + bottleneck_output_27, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_27 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_117 = torch.nn.functional.relu(x_116, inplace=True) + x_116 = None + new_features_27 = torch.conv2d( + x_117, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_117 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_28 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + ], + 1, + ) + x_118 = torch.nn.functional.batch_norm( + concated_features_28, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_28 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_119 = torch.nn.functional.relu(x_118, inplace=True) + x_118 = None + bottleneck_output_28 = torch.conv2d( + x_119, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_119 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_120 = torch.nn.functional.batch_norm( + bottleneck_output_28, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_28 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_121 = torch.nn.functional.relu(x_120, inplace=True) + x_120 = None + new_features_28 = torch.conv2d( + x_121, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_121 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_29 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + ], + 1, + ) + x_122 = torch.nn.functional.batch_norm( + concated_features_29, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_29 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_123 = torch.nn.functional.relu(x_122, inplace=True) + x_122 = None + bottleneck_output_29 = torch.conv2d( + x_123, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_123 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_124 = torch.nn.functional.batch_norm( + bottleneck_output_29, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_29 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_125 = torch.nn.functional.relu(x_124, inplace=True) + x_124 = None + new_features_29 = torch.conv2d( + x_125, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_125 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + concated_features_30 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + ], + 1, + ) + x_126 = torch.nn.functional.batch_norm( + concated_features_30, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_30 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ = (None) + x_127 = torch.nn.functional.relu(x_126, inplace=True) + x_126 = None + bottleneck_output_30 = torch.conv2d( + x_127, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_127 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ = (None) + x_128 = torch.nn.functional.batch_norm( + bottleneck_output_30, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_30 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ = (None) + x_129 = torch.nn.functional.relu(x_128, inplace=True) + x_128 = None + new_features_30 = torch.conv2d( + x_129, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_129 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ = (None) + concated_features_31 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + ], + 1, + ) + x_130 = torch.nn.functional.batch_norm( + concated_features_31, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_31 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ = (None) + x_131 = torch.nn.functional.relu(x_130, inplace=True) + x_130 = None + bottleneck_output_31 = torch.conv2d( + x_131, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_131 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ = (None) + x_132 = torch.nn.functional.batch_norm( + bottleneck_output_31, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_31 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ = (None) + x_133 = torch.nn.functional.relu(x_132, inplace=True) + x_132 = None + new_features_31 = torch.conv2d( + x_133, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_133 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ = (None) + concated_features_32 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + ], + 1, + ) + x_134 = torch.nn.functional.batch_norm( + concated_features_32, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_32 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ = (None) + x_135 = torch.nn.functional.relu(x_134, inplace=True) + x_134 = None + bottleneck_output_32 = torch.conv2d( + x_135, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_135 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ = (None) + x_136 = torch.nn.functional.batch_norm( + bottleneck_output_32, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_32 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ = (None) + x_137 = torch.nn.functional.relu(x_136, inplace=True) + x_136 = None + new_features_32 = torch.conv2d( + x_137, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_137 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ = (None) + concated_features_33 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + ], + 1, + ) + x_138 = torch.nn.functional.batch_norm( + concated_features_33, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_33 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ = (None) + x_139 = torch.nn.functional.relu(x_138, inplace=True) + x_138 = None + bottleneck_output_33 = torch.conv2d( + x_139, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_139 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ = (None) + x_140 = torch.nn.functional.batch_norm( + bottleneck_output_33, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_33 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ = (None) + x_141 = torch.nn.functional.relu(x_140, inplace=True) + x_140 = None + new_features_33 = torch.conv2d( + x_141, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_141 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ = (None) + concated_features_34 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + ], + 1, + ) + x_142 = torch.nn.functional.batch_norm( + concated_features_34, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_34 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ = (None) + x_143 = torch.nn.functional.relu(x_142, inplace=True) + x_142 = None + bottleneck_output_34 = torch.conv2d( + x_143, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_143 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ = (None) + x_144 = torch.nn.functional.batch_norm( + bottleneck_output_34, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_34 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ = (None) + x_145 = torch.nn.functional.relu(x_144, inplace=True) + x_144 = None + new_features_34 = torch.conv2d( + x_145, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_145 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ = (None) + concated_features_35 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + ], + 1, + ) + x_146 = torch.nn.functional.batch_norm( + concated_features_35, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_35 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ = (None) + x_147 = torch.nn.functional.relu(x_146, inplace=True) + x_146 = None + bottleneck_output_35 = torch.conv2d( + x_147, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_147 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ = (None) + x_148 = torch.nn.functional.batch_norm( + bottleneck_output_35, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_35 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ = (None) + x_149 = torch.nn.functional.relu(x_148, inplace=True) + x_148 = None + new_features_35 = torch.conv2d( + x_149, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_149 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ = (None) + concated_features_36 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + ], + 1, + ) + x_150 = torch.nn.functional.batch_norm( + concated_features_36, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_36 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ = (None) + x_151 = torch.nn.functional.relu(x_150, inplace=True) + x_150 = None + bottleneck_output_36 = torch.conv2d( + x_151, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_151 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ = (None) + x_152 = torch.nn.functional.batch_norm( + bottleneck_output_36, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_36 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ = (None) + x_153 = torch.nn.functional.relu(x_152, inplace=True) + x_152 = None + new_features_36 = torch.conv2d( + x_153, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_153 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ = (None) + concated_features_37 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + ], + 1, + ) + x_154 = torch.nn.functional.batch_norm( + concated_features_37, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_37 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ = (None) + x_155 = torch.nn.functional.relu(x_154, inplace=True) + x_154 = None + bottleneck_output_37 = torch.conv2d( + x_155, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_155 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ = (None) + x_156 = torch.nn.functional.batch_norm( + bottleneck_output_37, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_37 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ = (None) + x_157 = torch.nn.functional.relu(x_156, inplace=True) + x_156 = None + new_features_37 = torch.conv2d( + x_157, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_157 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ = (None) + concated_features_38 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + ], + 1, + ) + x_158 = torch.nn.functional.batch_norm( + concated_features_38, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_38 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ = (None) + x_159 = torch.nn.functional.relu(x_158, inplace=True) + x_158 = None + bottleneck_output_38 = torch.conv2d( + x_159, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_159 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ = (None) + x_160 = torch.nn.functional.batch_norm( + bottleneck_output_38, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_38 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ = (None) + x_161 = torch.nn.functional.relu(x_160, inplace=True) + x_160 = None + new_features_38 = torch.conv2d( + x_161, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_161 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ = (None) + concated_features_39 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + ], + 1, + ) + x_162 = torch.nn.functional.batch_norm( + concated_features_39, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_39 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ = (None) + x_163 = torch.nn.functional.relu(x_162, inplace=True) + x_162 = None + bottleneck_output_39 = torch.conv2d( + x_163, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_163 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ = (None) + x_164 = torch.nn.functional.batch_norm( + bottleneck_output_39, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_39 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ = (None) + x_165 = torch.nn.functional.relu(x_164, inplace=True) + x_164 = None + new_features_39 = torch.conv2d( + x_165, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_165 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ = (None) + concated_features_40 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + ], + 1, + ) + x_166 = torch.nn.functional.batch_norm( + concated_features_40, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_40 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ = (None) + x_167 = torch.nn.functional.relu(x_166, inplace=True) + x_166 = None + bottleneck_output_40 = torch.conv2d( + x_167, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_167 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ = (None) + x_168 = torch.nn.functional.batch_norm( + bottleneck_output_40, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_40 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ = (None) + x_169 = torch.nn.functional.relu(x_168, inplace=True) + x_168 = None + new_features_40 = torch.conv2d( + x_169, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_169 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ = (None) + concated_features_41 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + ], + 1, + ) + x_170 = torch.nn.functional.batch_norm( + concated_features_41, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_41 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ = (None) + x_171 = torch.nn.functional.relu(x_170, inplace=True) + x_170 = None + bottleneck_output_41 = torch.conv2d( + x_171, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_171 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ = (None) + x_172 = torch.nn.functional.batch_norm( + bottleneck_output_41, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_41 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ = (None) + x_173 = torch.nn.functional.relu(x_172, inplace=True) + x_172 = None + new_features_41 = torch.conv2d( + x_173, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_173 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ = (None) + concated_features_42 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + ], + 1, + ) + x_174 = torch.nn.functional.batch_norm( + concated_features_42, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_42 = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_ = (None) + x_175 = torch.nn.functional.relu(x_174, inplace=True) + x_174 = None + bottleneck_output_42 = torch.conv2d( + x_175, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_175 = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_ = (None) + x_176 = torch.nn.functional.batch_norm( + bottleneck_output_42, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_42 = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_ = (None) + x_177 = torch.nn.functional.relu(x_176, inplace=True) + x_176 = None + new_features_42 = torch.conv2d( + x_177, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_177 = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_ = (None) + concated_features_43 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + ], + 1, + ) + x_178 = torch.nn.functional.batch_norm( + concated_features_43, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_43 = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_ = (None) + x_179 = torch.nn.functional.relu(x_178, inplace=True) + x_178 = None + bottleneck_output_43 = torch.conv2d( + x_179, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_179 = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_ = (None) + x_180 = torch.nn.functional.batch_norm( + bottleneck_output_43, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_43 = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_ = (None) + x_181 = torch.nn.functional.relu(x_180, inplace=True) + x_180 = None + new_features_43 = torch.conv2d( + x_181, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_181 = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_ = (None) + concated_features_44 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + ], + 1, + ) + x_182 = torch.nn.functional.batch_norm( + concated_features_44, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_44 = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_ = (None) + x_183 = torch.nn.functional.relu(x_182, inplace=True) + x_182 = None + bottleneck_output_44 = torch.conv2d( + x_183, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_183 = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_ = (None) + x_184 = torch.nn.functional.batch_norm( + bottleneck_output_44, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_44 = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_ = (None) + x_185 = torch.nn.functional.relu(x_184, inplace=True) + x_184 = None + new_features_44 = torch.conv2d( + x_185, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_185 = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_ = (None) + concated_features_45 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + ], + 1, + ) + x_186 = torch.nn.functional.batch_norm( + concated_features_45, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_45 = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_ = (None) + x_187 = torch.nn.functional.relu(x_186, inplace=True) + x_186 = None + bottleneck_output_45 = torch.conv2d( + x_187, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_187 = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_ = (None) + x_188 = torch.nn.functional.batch_norm( + bottleneck_output_45, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_45 = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_ = (None) + x_189 = torch.nn.functional.relu(x_188, inplace=True) + x_188 = None + new_features_45 = torch.conv2d( + x_189, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_189 = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_ = (None) + concated_features_46 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + ], + 1, + ) + x_190 = torch.nn.functional.batch_norm( + concated_features_46, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_46 = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_ = (None) + x_191 = torch.nn.functional.relu(x_190, inplace=True) + x_190 = None + bottleneck_output_46 = torch.conv2d( + x_191, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_191 = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_ = (None) + x_192 = torch.nn.functional.batch_norm( + bottleneck_output_46, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_46 = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_ = (None) + x_193 = torch.nn.functional.relu(x_192, inplace=True) + x_192 = None + new_features_46 = torch.conv2d( + x_193, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_193 = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_ = (None) + concated_features_47 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + ], + 1, + ) + x_194 = torch.nn.functional.batch_norm( + concated_features_47, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_47 = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_ = (None) + x_195 = torch.nn.functional.relu(x_194, inplace=True) + x_194 = None + bottleneck_output_47 = torch.conv2d( + x_195, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_195 = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_ = (None) + x_196 = torch.nn.functional.batch_norm( + bottleneck_output_47, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_47 = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_ = (None) + x_197 = torch.nn.functional.relu(x_196, inplace=True) + x_196 = None + new_features_47 = torch.conv2d( + x_197, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_197 = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_ = (None) + concated_features_48 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + ], + 1, + ) + x_198 = torch.nn.functional.batch_norm( + concated_features_48, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_48 = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_ = (None) + x_199 = torch.nn.functional.relu(x_198, inplace=True) + x_198 = None + bottleneck_output_48 = torch.conv2d( + x_199, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_199 = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_ = (None) + x_200 = torch.nn.functional.batch_norm( + bottleneck_output_48, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_48 = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_ = (None) + x_201 = torch.nn.functional.relu(x_200, inplace=True) + x_200 = None + new_features_48 = torch.conv2d( + x_201, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_201 = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_ = (None) + concated_features_49 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + ], + 1, + ) + x_202 = torch.nn.functional.batch_norm( + concated_features_49, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_49 = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_ = (None) + x_203 = torch.nn.functional.relu(x_202, inplace=True) + x_202 = None + bottleneck_output_49 = torch.conv2d( + x_203, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_203 = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_ = (None) + x_204 = torch.nn.functional.batch_norm( + bottleneck_output_49, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_49 = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_ = (None) + x_205 = torch.nn.functional.relu(x_204, inplace=True) + x_204 = None + new_features_49 = torch.conv2d( + x_205, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_205 = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_ = (None) + concated_features_50 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + ], + 1, + ) + x_206 = torch.nn.functional.batch_norm( + concated_features_50, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_50 = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_ = (None) + x_207 = torch.nn.functional.relu(x_206, inplace=True) + x_206 = None + bottleneck_output_50 = torch.conv2d( + x_207, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_207 = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_ = (None) + x_208 = torch.nn.functional.batch_norm( + bottleneck_output_50, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_50 = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_ = (None) + x_209 = torch.nn.functional.relu(x_208, inplace=True) + x_208 = None + new_features_50 = torch.conv2d( + x_209, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_209 = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_ = (None) + concated_features_51 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + ], + 1, + ) + x_210 = torch.nn.functional.batch_norm( + concated_features_51, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_51 = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_ = (None) + x_211 = torch.nn.functional.relu(x_210, inplace=True) + x_210 = None + bottleneck_output_51 = torch.conv2d( + x_211, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_211 = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_ = (None) + x_212 = torch.nn.functional.batch_norm( + bottleneck_output_51, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_51 = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_ = (None) + x_213 = torch.nn.functional.relu(x_212, inplace=True) + x_212 = None + new_features_51 = torch.conv2d( + x_213, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_213 = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_ = (None) + concated_features_52 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + ], + 1, + ) + x_214 = torch.nn.functional.batch_norm( + concated_features_52, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_52 = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_ = (None) + x_215 = torch.nn.functional.relu(x_214, inplace=True) + x_214 = None + bottleneck_output_52 = torch.conv2d( + x_215, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_215 = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_ = (None) + x_216 = torch.nn.functional.batch_norm( + bottleneck_output_52, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_52 = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_ = (None) + x_217 = torch.nn.functional.relu(x_216, inplace=True) + x_216 = None + new_features_52 = torch.conv2d( + x_217, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_217 = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_ = (None) + concated_features_53 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + ], + 1, + ) + x_218 = torch.nn.functional.batch_norm( + concated_features_53, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_53 = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_ = (None) + x_219 = torch.nn.functional.relu(x_218, inplace=True) + x_218 = None + bottleneck_output_53 = torch.conv2d( + x_219, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_219 = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_ = (None) + x_220 = torch.nn.functional.batch_norm( + bottleneck_output_53, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_53 = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_ = (None) + x_221 = torch.nn.functional.relu(x_220, inplace=True) + x_220 = None + new_features_53 = torch.conv2d( + x_221, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_221 = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_ = (None) + input_9 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + ], + 1, + ) + input_8 = ( + new_features_18 + ) = ( + new_features_19 + ) = ( + new_features_20 + ) = ( + new_features_21 + ) = ( + new_features_22 + ) = ( + new_features_23 + ) = ( + new_features_24 + ) = ( + new_features_25 + ) = ( + new_features_26 + ) = ( + new_features_27 + ) = ( + new_features_28 + ) = ( + new_features_29 + ) = ( + new_features_30 + ) = ( + new_features_31 + ) = ( + new_features_32 + ) = ( + new_features_33 + ) = ( + new_features_34 + ) = ( + new_features_35 + ) = ( + new_features_36 + ) = ( + new_features_37 + ) = ( + new_features_38 + ) = ( + new_features_39 + ) = ( + new_features_40 + ) = ( + new_features_41 + ) = ( + new_features_42 + ) = ( + new_features_43 + ) = ( + new_features_44 + ) = ( + new_features_45 + ) = ( + new_features_46 + ) = ( + new_features_47 + ) = ( + new_features_48 + ) = ( + new_features_49 + ) = new_features_50 = new_features_51 = new_features_52 = new_features_53 = None + x_222 = torch.nn.functional.batch_norm( + input_9, + l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_9 = l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_ + ) = None + x_223 = torch.nn.functional.relu(x_222, inplace=True) + x_222 = None + input_10 = torch.conv2d( + x_223, + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_223 = ( + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_ + ) = None + input_11 = torch._C._nn.avg_pool2d(input_10, 2, 2, 0, False, True, None) + input_10 = None + concated_features_54 = torch.cat([input_11], 1) + x_224 = torch.nn.functional.batch_norm( + concated_features_54, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_54 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_225 = torch.nn.functional.relu(x_224, inplace=True) + x_224 = None + bottleneck_output_54 = torch.conv2d( + x_225, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_225 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_226 = torch.nn.functional.batch_norm( + bottleneck_output_54, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_54 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_227 = torch.nn.functional.relu(x_226, inplace=True) + x_226 = None + new_features_54 = torch.conv2d( + x_227, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_227 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_55 = torch.cat([input_11, new_features_54], 1) + x_228 = torch.nn.functional.batch_norm( + concated_features_55, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_55 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_229 = torch.nn.functional.relu(x_228, inplace=True) + x_228 = None + bottleneck_output_55 = torch.conv2d( + x_229, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_229 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_230 = torch.nn.functional.batch_norm( + bottleneck_output_55, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_55 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_231 = torch.nn.functional.relu(x_230, inplace=True) + x_230 = None + new_features_55 = torch.conv2d( + x_231, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_231 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_56 = torch.cat( + [input_11, new_features_54, new_features_55], 1 + ) + x_232 = torch.nn.functional.batch_norm( + concated_features_56, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_56 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_233 = torch.nn.functional.relu(x_232, inplace=True) + x_232 = None + bottleneck_output_56 = torch.conv2d( + x_233, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_233 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_234 = torch.nn.functional.batch_norm( + bottleneck_output_56, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_56 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_235 = torch.nn.functional.relu(x_234, inplace=True) + x_234 = None + new_features_56 = torch.conv2d( + x_235, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_235 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_57 = torch.cat( + [input_11, new_features_54, new_features_55, new_features_56], 1 + ) + x_236 = torch.nn.functional.batch_norm( + concated_features_57, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_57 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_237 = torch.nn.functional.relu(x_236, inplace=True) + x_236 = None + bottleneck_output_57 = torch.conv2d( + x_237, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_237 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_238 = torch.nn.functional.batch_norm( + bottleneck_output_57, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_57 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_239 = torch.nn.functional.relu(x_238, inplace=True) + x_238 = None + new_features_57 = torch.conv2d( + x_239, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_239 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_58 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + ], + 1, + ) + x_240 = torch.nn.functional.batch_norm( + concated_features_58, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_58 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_241 = torch.nn.functional.relu(x_240, inplace=True) + x_240 = None + bottleneck_output_58 = torch.conv2d( + x_241, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_241 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_242 = torch.nn.functional.batch_norm( + bottleneck_output_58, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_58 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_243 = torch.nn.functional.relu(x_242, inplace=True) + x_242 = None + new_features_58 = torch.conv2d( + x_243, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_243 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_59 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + ], + 1, + ) + x_244 = torch.nn.functional.batch_norm( + concated_features_59, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_59 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_245 = torch.nn.functional.relu(x_244, inplace=True) + x_244 = None + bottleneck_output_59 = torch.conv2d( + x_245, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_245 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_246 = torch.nn.functional.batch_norm( + bottleneck_output_59, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_59 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_247 = torch.nn.functional.relu(x_246, inplace=True) + x_246 = None + new_features_59 = torch.conv2d( + x_247, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_247 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_60 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + ], + 1, + ) + x_248 = torch.nn.functional.batch_norm( + concated_features_60, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_60 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_249 = torch.nn.functional.relu(x_248, inplace=True) + x_248 = None + bottleneck_output_60 = torch.conv2d( + x_249, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_249 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_250 = torch.nn.functional.batch_norm( + bottleneck_output_60, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_60 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_251 = torch.nn.functional.relu(x_250, inplace=True) + x_250 = None + new_features_60 = torch.conv2d( + x_251, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_251 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_61 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + ], + 1, + ) + x_252 = torch.nn.functional.batch_norm( + concated_features_61, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_61 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_253 = torch.nn.functional.relu(x_252, inplace=True) + x_252 = None + bottleneck_output_61 = torch.conv2d( + x_253, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_253 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_254 = torch.nn.functional.batch_norm( + bottleneck_output_61, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_61 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_255 = torch.nn.functional.relu(x_254, inplace=True) + x_254 = None + new_features_61 = torch.conv2d( + x_255, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_255 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_62 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + ], + 1, + ) + x_256 = torch.nn.functional.batch_norm( + concated_features_62, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_62 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_257 = torch.nn.functional.relu(x_256, inplace=True) + x_256 = None + bottleneck_output_62 = torch.conv2d( + x_257, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_257 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_258 = torch.nn.functional.batch_norm( + bottleneck_output_62, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_62 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_259 = torch.nn.functional.relu(x_258, inplace=True) + x_258 = None + new_features_62 = torch.conv2d( + x_259, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_259 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_63 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + ], + 1, + ) + x_260 = torch.nn.functional.batch_norm( + concated_features_63, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_63 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_261 = torch.nn.functional.relu(x_260, inplace=True) + x_260 = None + bottleneck_output_63 = torch.conv2d( + x_261, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_261 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_262 = torch.nn.functional.batch_norm( + bottleneck_output_63, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_63 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_263 = torch.nn.functional.relu(x_262, inplace=True) + x_262 = None + new_features_63 = torch.conv2d( + x_263, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_263 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_64 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + ], + 1, + ) + x_264 = torch.nn.functional.batch_norm( + concated_features_64, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_64 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_265 = torch.nn.functional.relu(x_264, inplace=True) + x_264 = None + bottleneck_output_64 = torch.conv2d( + x_265, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_265 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_266 = torch.nn.functional.batch_norm( + bottleneck_output_64, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_64 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_267 = torch.nn.functional.relu(x_266, inplace=True) + x_266 = None + new_features_64 = torch.conv2d( + x_267, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_267 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_65 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + ], + 1, + ) + x_268 = torch.nn.functional.batch_norm( + concated_features_65, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_65 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_269 = torch.nn.functional.relu(x_268, inplace=True) + x_268 = None + bottleneck_output_65 = torch.conv2d( + x_269, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_269 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_270 = torch.nn.functional.batch_norm( + bottleneck_output_65, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_65 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_271 = torch.nn.functional.relu(x_270, inplace=True) + x_270 = None + new_features_65 = torch.conv2d( + x_271, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_271 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + concated_features_66 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + ], + 1, + ) + x_272 = torch.nn.functional.batch_norm( + concated_features_66, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_66 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ = (None) + x_273 = torch.nn.functional.relu(x_272, inplace=True) + x_272 = None + bottleneck_output_66 = torch.conv2d( + x_273, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_273 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ = (None) + x_274 = torch.nn.functional.batch_norm( + bottleneck_output_66, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_66 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ = (None) + x_275 = torch.nn.functional.relu(x_274, inplace=True) + x_274 = None + new_features_66 = torch.conv2d( + x_275, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_275 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ = (None) + concated_features_67 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + ], + 1, + ) + x_276 = torch.nn.functional.batch_norm( + concated_features_67, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_67 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ = (None) + x_277 = torch.nn.functional.relu(x_276, inplace=True) + x_276 = None + bottleneck_output_67 = torch.conv2d( + x_277, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_277 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ = (None) + x_278 = torch.nn.functional.batch_norm( + bottleneck_output_67, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_67 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ = (None) + x_279 = torch.nn.functional.relu(x_278, inplace=True) + x_278 = None + new_features_67 = torch.conv2d( + x_279, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_279 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ = (None) + concated_features_68 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + ], + 1, + ) + x_280 = torch.nn.functional.batch_norm( + concated_features_68, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_68 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ = (None) + x_281 = torch.nn.functional.relu(x_280, inplace=True) + x_280 = None + bottleneck_output_68 = torch.conv2d( + x_281, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_281 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ = (None) + x_282 = torch.nn.functional.batch_norm( + bottleneck_output_68, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_68 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ = (None) + x_283 = torch.nn.functional.relu(x_282, inplace=True) + x_282 = None + new_features_68 = torch.conv2d( + x_283, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_283 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ = (None) + concated_features_69 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + ], + 1, + ) + x_284 = torch.nn.functional.batch_norm( + concated_features_69, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_69 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ = (None) + x_285 = torch.nn.functional.relu(x_284, inplace=True) + x_284 = None + bottleneck_output_69 = torch.conv2d( + x_285, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_285 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ = (None) + x_286 = torch.nn.functional.batch_norm( + bottleneck_output_69, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_69 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ = (None) + x_287 = torch.nn.functional.relu(x_286, inplace=True) + x_286 = None + new_features_69 = torch.conv2d( + x_287, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_287 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ = (None) + concated_features_70 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + ], + 1, + ) + x_288 = torch.nn.functional.batch_norm( + concated_features_70, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_70 = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_ = (None) + x_289 = torch.nn.functional.relu(x_288, inplace=True) + x_288 = None + bottleneck_output_70 = torch.conv2d( + x_289, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_289 = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_ = (None) + x_290 = torch.nn.functional.batch_norm( + bottleneck_output_70, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_70 = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_ = (None) + x_291 = torch.nn.functional.relu(x_290, inplace=True) + x_290 = None + new_features_70 = torch.conv2d( + x_291, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_291 = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_ = (None) + concated_features_71 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + ], + 1, + ) + x_292 = torch.nn.functional.batch_norm( + concated_features_71, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_71 = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_ = (None) + x_293 = torch.nn.functional.relu(x_292, inplace=True) + x_292 = None + bottleneck_output_71 = torch.conv2d( + x_293, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_293 = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_ = (None) + x_294 = torch.nn.functional.batch_norm( + bottleneck_output_71, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_71 = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_ = (None) + x_295 = torch.nn.functional.relu(x_294, inplace=True) + x_294 = None + new_features_71 = torch.conv2d( + x_295, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_295 = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_ = (None) + concated_features_72 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + ], + 1, + ) + x_296 = torch.nn.functional.batch_norm( + concated_features_72, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_72 = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_ = (None) + x_297 = torch.nn.functional.relu(x_296, inplace=True) + x_296 = None + bottleneck_output_72 = torch.conv2d( + x_297, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_297 = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_ = (None) + x_298 = torch.nn.functional.batch_norm( + bottleneck_output_72, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_72 = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_ = (None) + x_299 = torch.nn.functional.relu(x_298, inplace=True) + x_298 = None + new_features_72 = torch.conv2d( + x_299, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_299 = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_ = (None) + concated_features_73 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + ], + 1, + ) + x_300 = torch.nn.functional.batch_norm( + concated_features_73, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_73 = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_ = (None) + x_301 = torch.nn.functional.relu(x_300, inplace=True) + x_300 = None + bottleneck_output_73 = torch.conv2d( + x_301, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_301 = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_ = (None) + x_302 = torch.nn.functional.batch_norm( + bottleneck_output_73, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_73 = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_ = (None) + x_303 = torch.nn.functional.relu(x_302, inplace=True) + x_302 = None + new_features_73 = torch.conv2d( + x_303, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_303 = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_ = (None) + concated_features_74 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + ], + 1, + ) + x_304 = torch.nn.functional.batch_norm( + concated_features_74, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_74 = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_ = (None) + x_305 = torch.nn.functional.relu(x_304, inplace=True) + x_304 = None + bottleneck_output_74 = torch.conv2d( + x_305, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_305 = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_ = (None) + x_306 = torch.nn.functional.batch_norm( + bottleneck_output_74, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_74 = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_ = (None) + x_307 = torch.nn.functional.relu(x_306, inplace=True) + x_306 = None + new_features_74 = torch.conv2d( + x_307, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_307 = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_ = (None) + concated_features_75 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + ], + 1, + ) + x_308 = torch.nn.functional.batch_norm( + concated_features_75, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_75 = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_ = (None) + x_309 = torch.nn.functional.relu(x_308, inplace=True) + x_308 = None + bottleneck_output_75 = torch.conv2d( + x_309, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_309 = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_ = (None) + x_310 = torch.nn.functional.batch_norm( + bottleneck_output_75, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_75 = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_ = (None) + x_311 = torch.nn.functional.relu(x_310, inplace=True) + x_310 = None + new_features_75 = torch.conv2d( + x_311, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_311 = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_ = (None) + concated_features_76 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + ], + 1, + ) + x_312 = torch.nn.functional.batch_norm( + concated_features_76, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_76 = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_ = (None) + x_313 = torch.nn.functional.relu(x_312, inplace=True) + x_312 = None + bottleneck_output_76 = torch.conv2d( + x_313, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_313 = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_ = (None) + x_314 = torch.nn.functional.batch_norm( + bottleneck_output_76, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_76 = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_ = (None) + x_315 = torch.nn.functional.relu(x_314, inplace=True) + x_314 = None + new_features_76 = torch.conv2d( + x_315, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_315 = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_ = (None) + concated_features_77 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + ], + 1, + ) + x_316 = torch.nn.functional.batch_norm( + concated_features_77, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_77 = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_ = (None) + x_317 = torch.nn.functional.relu(x_316, inplace=True) + x_316 = None + bottleneck_output_77 = torch.conv2d( + x_317, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_317 = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_ = (None) + x_318 = torch.nn.functional.batch_norm( + bottleneck_output_77, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_77 = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_ = (None) + x_319 = torch.nn.functional.relu(x_318, inplace=True) + x_318 = None + new_features_77 = torch.conv2d( + x_319, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_319 = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_ = (None) + input_12 = torch.cat( + [ + input_11, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + ], + 1, + ) + input_11 = ( + new_features_54 + ) = ( + new_features_55 + ) = ( + new_features_56 + ) = ( + new_features_57 + ) = ( + new_features_58 + ) = ( + new_features_59 + ) = ( + new_features_60 + ) = ( + new_features_61 + ) = ( + new_features_62 + ) = ( + new_features_63 + ) = ( + new_features_64 + ) = ( + new_features_65 + ) = ( + new_features_66 + ) = ( + new_features_67 + ) = ( + new_features_68 + ) = ( + new_features_69 + ) = ( + new_features_70 + ) = ( + new_features_71 + ) = ( + new_features_72 + ) = ( + new_features_73 + ) = new_features_74 = new_features_75 = new_features_76 = new_features_77 = None + x_320 = torch.nn.functional.batch_norm( + input_12, + l_self_modules_features_modules_norm5_buffers_running_mean_, + l_self_modules_features_modules_norm5_buffers_running_var_, + l_self_modules_features_modules_norm5_parameters_weight_, + l_self_modules_features_modules_norm5_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_12 = ( + l_self_modules_features_modules_norm5_buffers_running_mean_ + ) = ( + l_self_modules_features_modules_norm5_buffers_running_var_ + ) = ( + l_self_modules_features_modules_norm5_parameters_weight_ + ) = l_self_modules_features_modules_norm5_parameters_bias_ = None + x_321 = torch.nn.functional.relu(x_320, inplace=True) + x_320 = None + x_322 = torch.nn.functional.adaptive_avg_pool2d(x_321, 1) + x_321 = None + x_323 = x_322.flatten(1, -1) + x_322 = None + x_324 = torch.nn.functional.dropout(x_323, 0.0, False, False) + x_323 = None + x_325 = torch._C._nn.linear( + x_324, + l_self_modules_classifier_parameters_weight_, + l_self_modules_classifier_parameters_bias_, + ) + x_324 = ( + l_self_modules_classifier_parameters_weight_ + ) = l_self_modules_classifier_parameters_bias_ = None + return (x_325,) diff --git a/samples/timm/densenet161.tv_in1k/weight_meta.py b/samples/timm/densenet161.tv_in1k/weight_meta.py new file mode 100644 index 0000000000..7efccecee1 --- /dev/null +++ b/samples/timm/densenet161.tv_in1k/weight_meta.py @@ -0,0 +1,8090 @@ +class Program_weight_tensor_meta_L_self_modules_features_modules_conv0_parameters_weight_: + name = "L_self_modules_features_modules_conv0_parameters_weight_" + shape = [96, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.117 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.230 + std = 1.283 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_buffers_running_mean_: + name = "L_self_modules_features_modules_norm0_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_buffers_running_var_: + name = "L_self_modules_features_modules_norm0_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_parameters_weight_: + name = "L_self_modules_features_modules_norm0_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_parameters_bias_: + name = "L_self_modules_features_modules_norm0_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [192, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.143 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [144] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [144] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [144] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [192, 144, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.118 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [192, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.091 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [192, 288, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [336] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [336] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [336] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [336] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [192, 336, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.077 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_" + ) + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition1_modules_norm_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition1_modules_norm_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition1_modules_conv_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [192, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.103 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [240] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [192, 240, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.091 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [192, 288, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.084 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [336] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [336] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [336] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [336] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [192, 336, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.077 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [432] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [432] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [432] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [432] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [192, 432, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.068 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [192, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [528] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [528] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [528] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [528] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [192, 528, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.061 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [192, 576, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [624] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [624] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [624] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [624] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [192, 624, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.057 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [192, 672, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.055 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [720] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [720] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [720] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [720] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [192, 720, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_" + ) + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition2_modules_norm_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition2_modules_norm_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition2_modules_conv_parameters_weight_" + shape = [384, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [192, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [432] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [432] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [432] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [432] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [192, 432, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.068 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [192, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [528] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [528] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [528] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [528] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [192, 528, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.061 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [192, 576, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [624] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [624] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [624] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [624] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [192, 624, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.057 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [192, 672, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.055 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [720] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [720] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [720] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [720] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [192, 720, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [192, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [816] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [816] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [816] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [192, 816, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.049 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [192, 864, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [912] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [912] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [912] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [912] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [192, 912, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_" + shape = [192, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.045 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_" + shape = [1008] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_" + shape = [1008] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_" + shape = [1008] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_" + shape = [1008] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_" + shape = [192, 1008, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.045 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_" + shape = [192, 1056, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_" + shape = [1104] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_" + shape = [1104] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_" + shape = [1104] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_" + shape = [1104] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_" + shape = [192, 1104, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_" + shape = [192, 1152, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_" + shape = [1200] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_" + shape = [1200] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_" + shape = [1200] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_" + shape = [1200] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_" + shape = [192, 1200, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_" + shape = [192, 1248, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_" + shape = [1296] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_" + shape = [1296] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_" + shape = [1296] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_" + shape = [1296] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_" + shape = [192, 1296, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.039 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_" + shape = [192, 1344, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.039 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_" + shape = [1392] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_" + shape = [1392] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_" + shape = [1392] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_" + shape = [1392] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_" + shape = [192, 1392, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.038 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_" + shape = [192, 1440, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_" + shape = [1488] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_" + shape = [1488] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_" + shape = [1488] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_" + shape = [1488] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_" + shape = [192, 1488, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_" + shape = [192, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_" + shape = [1584] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_" + shape = [1584] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_" + shape = [1584] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_" + shape = [1584] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_" + shape = [192, 1584, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_" + shape = [192, 1632, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.035 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_" + shape = [1680] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_" + shape = [1680] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_" + shape = [1680] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_" + shape = [1680] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_" + shape = [192, 1680, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_" + shape = [192, 1728, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_" + shape = [1776] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_" + shape = [1776] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_" + shape = [1776] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_" + shape = [1776] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_" + shape = [192, 1776, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_" + shape = [1824] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_" + shape = [1824] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_" + shape = [1824] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_" + shape = [1824] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_" + shape = [192, 1824, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.033 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_" + shape = [1872] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_" + shape = [1872] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_" + shape = [1872] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_" + shape = [1872] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_" + shape = [192, 1872, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.033 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_" + shape = [1920] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_" + shape = [1920] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_" + shape = [1920] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_" + shape = [1920] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_" + shape = [192, 1920, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.032 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_" + shape = [1968] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_" + shape = [1968] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_" + shape = [1968] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_" + shape = [1968] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_" + shape = [192, 1968, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.032 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_" + shape = [2016] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_" + shape = [2016] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_" + shape = [2016] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_" + shape = [2016] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_" + shape = [192, 2016, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.031 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_" + shape = [2064] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_" + shape = [2064] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_" + shape = [2064] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_" + shape = [2064] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_" + shape = [192, 2064, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.031 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_" + ) + shape = [2112] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_" + ) + shape = [2112] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition3_modules_norm_parameters_weight_" + shape = [2112] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition3_modules_norm_parameters_bias_" + shape = [2112] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition3_modules_conv_parameters_weight_" + shape = [1056, 2112, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.031 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [192, 1056, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [1104] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [1104] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [1104] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [1104] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [192, 1104, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [192, 1152, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [1200] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [1200] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [1200] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [1200] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [192, 1200, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [192, 1248, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [1296] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [1296] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [1296] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [1296] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [192, 1296, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.039 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [192, 1344, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.039 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [1392] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [1392] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [1392] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [1392] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [192, 1392, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.038 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [192, 1440, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [1488] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [1488] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [1488] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [1488] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [192, 1488, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [192, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [1584] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [1584] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [1584] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [1584] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [192, 1584, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_" + shape = [192, 1632, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.035 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_" + shape = [1680] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_" + shape = [1680] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_" + shape = [1680] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_" + shape = [1680] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_" + shape = [192, 1680, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_" + shape = [192, 1728, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_" + shape = [1776] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_" + shape = [1776] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_" + shape = [1776] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_" + shape = [1776] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_" + shape = [192, 1776, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.033 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_" + shape = [1824] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_" + shape = [1824] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_" + shape = [1824] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_" + shape = [1824] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_" + shape = [192, 1824, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.033 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_" + shape = [1872] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_" + shape = [1872] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_" + shape = [1872] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_" + shape = [1872] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_" + shape = [192, 1872, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.033 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_" + shape = [1920] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_" + shape = [1920] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_" + shape = [1920] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_" + shape = [1920] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_" + shape = [192, 1920, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.032 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_" + shape = [1968] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_" + shape = [1968] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_" + shape = [1968] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_" + shape = [1968] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_" + shape = [192, 1968, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.032 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_" + shape = [2016] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_" + shape = [2016] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_" + shape = [2016] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_" + shape = [2016] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_" + shape = [192, 2016, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.031 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_" + shape = [2064] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_" + shape = [2064] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_" + shape = [2064] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_" + shape = [2064] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_" + shape = [192, 2064, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.031 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_" + shape = [2112] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_" + shape = [2112] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_" + shape = [2112] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_" + shape = [2112] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_" + shape = [192, 2112, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.031 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_" + shape = [2160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_" + shape = [2160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_" + shape = [2160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_" + shape = [2160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_" + shape = [192, 2160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_" + shape = [48, 192, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_buffers_running_mean_: + name = "L_self_modules_features_modules_norm5_buffers_running_mean_" + shape = [2208] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_buffers_running_var_: + name = "L_self_modules_features_modules_norm5_buffers_running_var_" + shape = [2208] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_parameters_weight_: + name = "L_self_modules_features_modules_norm5_parameters_weight_" + shape = [2208] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_parameters_bias_: + name = "L_self_modules_features_modules_norm5_parameters_bias_" + shape = [2208] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_classifier_parameters_weight_: + name = "L_self_modules_classifier_parameters_weight_" + shape = [1000, 2208] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.012 + data = None + + +class Program_weight_tensor_meta_L_self_modules_classifier_parameters_bias_: + name = "L_self_modules_classifier_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/densenet169.tv_in1k/graph_hash.txt b/samples/timm/densenet169.tv_in1k/graph_hash.txt new file mode 100644 index 0000000000..9ccec065de --- /dev/null +++ b/samples/timm/densenet169.tv_in1k/graph_hash.txt @@ -0,0 +1 @@ +20ec56b7a5ce2dcf23a0114ba99191eaa451499b4406b1eec91ea0b1bd44dfac \ No newline at end of file diff --git a/samples/timm/densenet169.tv_in1k/graph_net.json b/samples/timm/densenet169.tv_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/densenet169.tv_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/densenet169.tv_in1k/input_meta.py b/samples/timm/densenet169.tv_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/densenet169.tv_in1k/input_tensor_constraints.py b/samples/timm/densenet169.tv_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/densenet169.tv_in1k/model.py b/samples/timm/densenet169.tv_in1k/model.py new file mode 100644 index 0000000000..ab882eb052 --- /dev/null +++ b/samples/timm/densenet169.tv_in1k/model.py @@ -0,0 +1,7373 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_features_modules_conv0_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_features_modules_norm0_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_norm0_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_norm0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm5_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_norm5_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_norm5_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm5_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_classifier_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_classifier_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_features_modules_conv0_parameters_weight_ = ( + L_self_modules_features_modules_conv0_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_features_modules_norm0_buffers_running_mean_ = ( + L_self_modules_features_modules_norm0_buffers_running_mean_ + ) + l_self_modules_features_modules_norm0_buffers_running_var_ = ( + L_self_modules_features_modules_norm0_buffers_running_var_ + ) + l_self_modules_features_modules_norm0_parameters_weight_ = ( + L_self_modules_features_modules_norm0_parameters_weight_ + ) + l_self_modules_features_modules_norm0_parameters_bias_ = ( + L_self_modules_features_modules_norm0_parameters_bias_ + ) + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition1_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition1_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition1_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition2_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition2_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition2_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition3_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition3_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition3_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_ + l_self_modules_features_modules_norm5_buffers_running_mean_ = ( + L_self_modules_features_modules_norm5_buffers_running_mean_ + ) + l_self_modules_features_modules_norm5_buffers_running_var_ = ( + L_self_modules_features_modules_norm5_buffers_running_var_ + ) + l_self_modules_features_modules_norm5_parameters_weight_ = ( + L_self_modules_features_modules_norm5_parameters_weight_ + ) + l_self_modules_features_modules_norm5_parameters_bias_ = ( + L_self_modules_features_modules_norm5_parameters_bias_ + ) + l_self_modules_classifier_parameters_weight_ = ( + L_self_modules_classifier_parameters_weight_ + ) + l_self_modules_classifier_parameters_bias_ = ( + L_self_modules_classifier_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_features_modules_conv0_parameters_weight_, + None, + (2, 2), + (3, 3), + (1, 1), + 1, + ) + l_x_ = l_self_modules_features_modules_conv0_parameters_weight_ = None + x = torch.nn.functional.batch_norm( + input_1, + l_self_modules_features_modules_norm0_buffers_running_mean_, + l_self_modules_features_modules_norm0_buffers_running_var_, + l_self_modules_features_modules_norm0_parameters_weight_, + l_self_modules_features_modules_norm0_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_1 = ( + l_self_modules_features_modules_norm0_buffers_running_mean_ + ) = ( + l_self_modules_features_modules_norm0_buffers_running_var_ + ) = ( + l_self_modules_features_modules_norm0_parameters_weight_ + ) = l_self_modules_features_modules_norm0_parameters_bias_ = None + x_1 = torch.nn.functional.relu(x, inplace=True) + x = None + input_2 = torch.nn.functional.max_pool2d( + x_1, 3, 2, 1, 1, ceil_mode=False, return_indices=False + ) + x_1 = None + concated_features = torch.cat([input_2], 1) + x_2 = torch.nn.functional.batch_norm( + concated_features, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_3 = torch.nn.functional.relu(x_2, inplace=True) + x_2 = None + bottleneck_output = torch.conv2d( + x_3, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_3 = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + bottleneck_output, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_5 = torch.nn.functional.relu(x_4, inplace=True) + x_4 = None + new_features = torch.conv2d( + x_5, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_5 = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_1 = torch.cat([input_2, new_features], 1) + x_6 = torch.nn.functional.batch_norm( + concated_features_1, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_1 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_7 = torch.nn.functional.relu(x_6, inplace=True) + x_6 = None + bottleneck_output_1 = torch.conv2d( + x_7, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_7 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_8 = torch.nn.functional.batch_norm( + bottleneck_output_1, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_1 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_9 = torch.nn.functional.relu(x_8, inplace=True) + x_8 = None + new_features_1 = torch.conv2d( + x_9, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_9 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_2 = torch.cat([input_2, new_features, new_features_1], 1) + x_10 = torch.nn.functional.batch_norm( + concated_features_2, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_2 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_11 = torch.nn.functional.relu(x_10, inplace=True) + x_10 = None + bottleneck_output_2 = torch.conv2d( + x_11, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_11 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_12 = torch.nn.functional.batch_norm( + bottleneck_output_2, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_2 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_13 = torch.nn.functional.relu(x_12, inplace=True) + x_12 = None + new_features_2 = torch.conv2d( + x_13, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_13 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_3 = torch.cat( + [input_2, new_features, new_features_1, new_features_2], 1 + ) + x_14 = torch.nn.functional.batch_norm( + concated_features_3, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_3 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_15 = torch.nn.functional.relu(x_14, inplace=True) + x_14 = None + bottleneck_output_3 = torch.conv2d( + x_15, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_15 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_16 = torch.nn.functional.batch_norm( + bottleneck_output_3, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_3 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_17 = torch.nn.functional.relu(x_16, inplace=True) + x_16 = None + new_features_3 = torch.conv2d( + x_17, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_17 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_4 = torch.cat( + [input_2, new_features, new_features_1, new_features_2, new_features_3], 1 + ) + x_18 = torch.nn.functional.batch_norm( + concated_features_4, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_4 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_19 = torch.nn.functional.relu(x_18, inplace=True) + x_18 = None + bottleneck_output_4 = torch.conv2d( + x_19, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_19 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_20 = torch.nn.functional.batch_norm( + bottleneck_output_4, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_4 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_21 = torch.nn.functional.relu(x_20, inplace=True) + x_20 = None + new_features_4 = torch.conv2d( + x_21, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_21 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_5 = torch.cat( + [ + input_2, + new_features, + new_features_1, + new_features_2, + new_features_3, + new_features_4, + ], + 1, + ) + x_22 = torch.nn.functional.batch_norm( + concated_features_5, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_5 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_23 = torch.nn.functional.relu(x_22, inplace=True) + x_22 = None + bottleneck_output_5 = torch.conv2d( + x_23, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_24 = torch.nn.functional.batch_norm( + bottleneck_output_5, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_5 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_25 = torch.nn.functional.relu(x_24, inplace=True) + x_24 = None + new_features_5 = torch.conv2d( + x_25, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_25 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + input_3 = torch.cat( + [ + input_2, + new_features, + new_features_1, + new_features_2, + new_features_3, + new_features_4, + new_features_5, + ], + 1, + ) + input_2 = ( + new_features + ) = ( + new_features_1 + ) = new_features_2 = new_features_3 = new_features_4 = new_features_5 = None + x_26 = torch.nn.functional.batch_norm( + input_3, + l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_3 = l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_ + ) = None + x_27 = torch.nn.functional.relu(x_26, inplace=True) + x_26 = None + input_4 = torch.conv2d( + x_27, + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_27 = ( + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_ + ) = None + input_5 = torch._C._nn.avg_pool2d(input_4, 2, 2, 0, False, True, None) + input_4 = None + concated_features_6 = torch.cat([input_5], 1) + x_28 = torch.nn.functional.batch_norm( + concated_features_6, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_6 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_29 = torch.nn.functional.relu(x_28, inplace=True) + x_28 = None + bottleneck_output_6 = torch.conv2d( + x_29, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_29 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_30 = torch.nn.functional.batch_norm( + bottleneck_output_6, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_6 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_31 = torch.nn.functional.relu(x_30, inplace=True) + x_30 = None + new_features_6 = torch.conv2d( + x_31, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_31 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_7 = torch.cat([input_5, new_features_6], 1) + x_32 = torch.nn.functional.batch_norm( + concated_features_7, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_7 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_33 = torch.nn.functional.relu(x_32, inplace=True) + x_32 = None + bottleneck_output_7 = torch.conv2d( + x_33, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_33 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_34 = torch.nn.functional.batch_norm( + bottleneck_output_7, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_7 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_35 = torch.nn.functional.relu(x_34, inplace=True) + x_34 = None + new_features_7 = torch.conv2d( + x_35, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_35 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_8 = torch.cat([input_5, new_features_6, new_features_7], 1) + x_36 = torch.nn.functional.batch_norm( + concated_features_8, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_8 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_37 = torch.nn.functional.relu(x_36, inplace=True) + x_36 = None + bottleneck_output_8 = torch.conv2d( + x_37, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_37 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_38 = torch.nn.functional.batch_norm( + bottleneck_output_8, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_8 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_39 = torch.nn.functional.relu(x_38, inplace=True) + x_38 = None + new_features_8 = torch.conv2d( + x_39, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_39 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_9 = torch.cat( + [input_5, new_features_6, new_features_7, new_features_8], 1 + ) + x_40 = torch.nn.functional.batch_norm( + concated_features_9, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_9 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_41 = torch.nn.functional.relu(x_40, inplace=True) + x_40 = None + bottleneck_output_9 = torch.conv2d( + x_41, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_41 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_42 = torch.nn.functional.batch_norm( + bottleneck_output_9, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_9 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_43 = torch.nn.functional.relu(x_42, inplace=True) + x_42 = None + new_features_9 = torch.conv2d( + x_43, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_43 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_10 = torch.cat( + [input_5, new_features_6, new_features_7, new_features_8, new_features_9], 1 + ) + x_44 = torch.nn.functional.batch_norm( + concated_features_10, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_10 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_45 = torch.nn.functional.relu(x_44, inplace=True) + x_44 = None + bottleneck_output_10 = torch.conv2d( + x_45, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_45 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_46 = torch.nn.functional.batch_norm( + bottleneck_output_10, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_10 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_47 = torch.nn.functional.relu(x_46, inplace=True) + x_46 = None + new_features_10 = torch.conv2d( + x_47, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_47 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_11 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + ], + 1, + ) + x_48 = torch.nn.functional.batch_norm( + concated_features_11, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_11 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_49 = torch.nn.functional.relu(x_48, inplace=True) + x_48 = None + bottleneck_output_11 = torch.conv2d( + x_49, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_49 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_50 = torch.nn.functional.batch_norm( + bottleneck_output_11, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_11 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_51 = torch.nn.functional.relu(x_50, inplace=True) + x_50 = None + new_features_11 = torch.conv2d( + x_51, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_51 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_12 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + ], + 1, + ) + x_52 = torch.nn.functional.batch_norm( + concated_features_12, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_12 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_53 = torch.nn.functional.relu(x_52, inplace=True) + x_52 = None + bottleneck_output_12 = torch.conv2d( + x_53, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_53 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_54 = torch.nn.functional.batch_norm( + bottleneck_output_12, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_12 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_55 = torch.nn.functional.relu(x_54, inplace=True) + x_54 = None + new_features_12 = torch.conv2d( + x_55, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_55 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_13 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + ], + 1, + ) + x_56 = torch.nn.functional.batch_norm( + concated_features_13, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_13 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_57 = torch.nn.functional.relu(x_56, inplace=True) + x_56 = None + bottleneck_output_13 = torch.conv2d( + x_57, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_57 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_58 = torch.nn.functional.batch_norm( + bottleneck_output_13, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_13 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_59 = torch.nn.functional.relu(x_58, inplace=True) + x_58 = None + new_features_13 = torch.conv2d( + x_59, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_59 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_14 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + ], + 1, + ) + x_60 = torch.nn.functional.batch_norm( + concated_features_14, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_14 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_61 = torch.nn.functional.relu(x_60, inplace=True) + x_60 = None + bottleneck_output_14 = torch.conv2d( + x_61, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_61 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_62 = torch.nn.functional.batch_norm( + bottleneck_output_14, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_14 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_63 = torch.nn.functional.relu(x_62, inplace=True) + x_62 = None + new_features_14 = torch.conv2d( + x_63, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_63 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_15 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + ], + 1, + ) + x_64 = torch.nn.functional.batch_norm( + concated_features_15, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_15 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_65 = torch.nn.functional.relu(x_64, inplace=True) + x_64 = None + bottleneck_output_15 = torch.conv2d( + x_65, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_65 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_66 = torch.nn.functional.batch_norm( + bottleneck_output_15, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_15 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_67 = torch.nn.functional.relu(x_66, inplace=True) + x_66 = None + new_features_15 = torch.conv2d( + x_67, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_67 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_16 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + ], + 1, + ) + x_68 = torch.nn.functional.batch_norm( + concated_features_16, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_16 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_69 = torch.nn.functional.relu(x_68, inplace=True) + x_68 = None + bottleneck_output_16 = torch.conv2d( + x_69, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_69 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_70 = torch.nn.functional.batch_norm( + bottleneck_output_16, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_16 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_71 = torch.nn.functional.relu(x_70, inplace=True) + x_70 = None + new_features_16 = torch.conv2d( + x_71, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_71 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_17 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + new_features_16, + ], + 1, + ) + x_72 = torch.nn.functional.batch_norm( + concated_features_17, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_17 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_73 = torch.nn.functional.relu(x_72, inplace=True) + x_72 = None + bottleneck_output_17 = torch.conv2d( + x_73, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_73 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_74 = torch.nn.functional.batch_norm( + bottleneck_output_17, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_17 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_75 = torch.nn.functional.relu(x_74, inplace=True) + x_74 = None + new_features_17 = torch.conv2d( + x_75, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_75 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + input_6 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + new_features_16, + new_features_17, + ], + 1, + ) + input_5 = ( + new_features_6 + ) = ( + new_features_7 + ) = ( + new_features_8 + ) = ( + new_features_9 + ) = ( + new_features_10 + ) = ( + new_features_11 + ) = ( + new_features_12 + ) = ( + new_features_13 + ) = new_features_14 = new_features_15 = new_features_16 = new_features_17 = None + x_76 = torch.nn.functional.batch_norm( + input_6, + l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_6 = l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_ + ) = None + x_77 = torch.nn.functional.relu(x_76, inplace=True) + x_76 = None + input_7 = torch.conv2d( + x_77, + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_77 = ( + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_ + ) = None + input_8 = torch._C._nn.avg_pool2d(input_7, 2, 2, 0, False, True, None) + input_7 = None + concated_features_18 = torch.cat([input_8], 1) + x_78 = torch.nn.functional.batch_norm( + concated_features_18, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_18 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_79 = torch.nn.functional.relu(x_78, inplace=True) + x_78 = None + bottleneck_output_18 = torch.conv2d( + x_79, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_79 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_80 = torch.nn.functional.batch_norm( + bottleneck_output_18, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_18 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_81 = torch.nn.functional.relu(x_80, inplace=True) + x_80 = None + new_features_18 = torch.conv2d( + x_81, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_81 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_19 = torch.cat([input_8, new_features_18], 1) + x_82 = torch.nn.functional.batch_norm( + concated_features_19, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_19 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_83 = torch.nn.functional.relu(x_82, inplace=True) + x_82 = None + bottleneck_output_19 = torch.conv2d( + x_83, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_83 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_84 = torch.nn.functional.batch_norm( + bottleneck_output_19, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_19 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_85 = torch.nn.functional.relu(x_84, inplace=True) + x_84 = None + new_features_19 = torch.conv2d( + x_85, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_85 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_20 = torch.cat([input_8, new_features_18, new_features_19], 1) + x_86 = torch.nn.functional.batch_norm( + concated_features_20, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_20 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_87 = torch.nn.functional.relu(x_86, inplace=True) + x_86 = None + bottleneck_output_20 = torch.conv2d( + x_87, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_87 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_88 = torch.nn.functional.batch_norm( + bottleneck_output_20, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_20 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_89 = torch.nn.functional.relu(x_88, inplace=True) + x_88 = None + new_features_20 = torch.conv2d( + x_89, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_89 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_21 = torch.cat( + [input_8, new_features_18, new_features_19, new_features_20], 1 + ) + x_90 = torch.nn.functional.batch_norm( + concated_features_21, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_21 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_91 = torch.nn.functional.relu(x_90, inplace=True) + x_90 = None + bottleneck_output_21 = torch.conv2d( + x_91, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_91 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_92 = torch.nn.functional.batch_norm( + bottleneck_output_21, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_21 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_93 = torch.nn.functional.relu(x_92, inplace=True) + x_92 = None + new_features_21 = torch.conv2d( + x_93, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_93 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_22 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + ], + 1, + ) + x_94 = torch.nn.functional.batch_norm( + concated_features_22, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_22 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_95 = torch.nn.functional.relu(x_94, inplace=True) + x_94 = None + bottleneck_output_22 = torch.conv2d( + x_95, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_95 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_96 = torch.nn.functional.batch_norm( + bottleneck_output_22, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_22 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_97 = torch.nn.functional.relu(x_96, inplace=True) + x_96 = None + new_features_22 = torch.conv2d( + x_97, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_97 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_23 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + ], + 1, + ) + x_98 = torch.nn.functional.batch_norm( + concated_features_23, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_23 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_99 = torch.nn.functional.relu(x_98, inplace=True) + x_98 = None + bottleneck_output_23 = torch.conv2d( + x_99, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_99 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_100 = torch.nn.functional.batch_norm( + bottleneck_output_23, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_23 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_101 = torch.nn.functional.relu(x_100, inplace=True) + x_100 = None + new_features_23 = torch.conv2d( + x_101, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_101 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_24 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + ], + 1, + ) + x_102 = torch.nn.functional.batch_norm( + concated_features_24, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_24 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_103 = torch.nn.functional.relu(x_102, inplace=True) + x_102 = None + bottleneck_output_24 = torch.conv2d( + x_103, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_104 = torch.nn.functional.batch_norm( + bottleneck_output_24, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_24 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_105 = torch.nn.functional.relu(x_104, inplace=True) + x_104 = None + new_features_24 = torch.conv2d( + x_105, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_105 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_25 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + ], + 1, + ) + x_106 = torch.nn.functional.batch_norm( + concated_features_25, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_25 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_107 = torch.nn.functional.relu(x_106, inplace=True) + x_106 = None + bottleneck_output_25 = torch.conv2d( + x_107, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_107 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_108 = torch.nn.functional.batch_norm( + bottleneck_output_25, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_25 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_109 = torch.nn.functional.relu(x_108, inplace=True) + x_108 = None + new_features_25 = torch.conv2d( + x_109, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_109 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_26 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + ], + 1, + ) + x_110 = torch.nn.functional.batch_norm( + concated_features_26, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_26 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_111 = torch.nn.functional.relu(x_110, inplace=True) + x_110 = None + bottleneck_output_26 = torch.conv2d( + x_111, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_112 = torch.nn.functional.batch_norm( + bottleneck_output_26, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_26 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_113 = torch.nn.functional.relu(x_112, inplace=True) + x_112 = None + new_features_26 = torch.conv2d( + x_113, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_113 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_27 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + ], + 1, + ) + x_114 = torch.nn.functional.batch_norm( + concated_features_27, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_27 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_115 = torch.nn.functional.relu(x_114, inplace=True) + x_114 = None + bottleneck_output_27 = torch.conv2d( + x_115, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_116 = torch.nn.functional.batch_norm( + bottleneck_output_27, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_27 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_117 = torch.nn.functional.relu(x_116, inplace=True) + x_116 = None + new_features_27 = torch.conv2d( + x_117, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_117 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_28 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + ], + 1, + ) + x_118 = torch.nn.functional.batch_norm( + concated_features_28, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_28 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_119 = torch.nn.functional.relu(x_118, inplace=True) + x_118 = None + bottleneck_output_28 = torch.conv2d( + x_119, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_119 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_120 = torch.nn.functional.batch_norm( + bottleneck_output_28, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_28 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_121 = torch.nn.functional.relu(x_120, inplace=True) + x_120 = None + new_features_28 = torch.conv2d( + x_121, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_121 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_29 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + ], + 1, + ) + x_122 = torch.nn.functional.batch_norm( + concated_features_29, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_29 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_123 = torch.nn.functional.relu(x_122, inplace=True) + x_122 = None + bottleneck_output_29 = torch.conv2d( + x_123, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_123 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_124 = torch.nn.functional.batch_norm( + bottleneck_output_29, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_29 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_125 = torch.nn.functional.relu(x_124, inplace=True) + x_124 = None + new_features_29 = torch.conv2d( + x_125, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_125 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + concated_features_30 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + ], + 1, + ) + x_126 = torch.nn.functional.batch_norm( + concated_features_30, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_30 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ = (None) + x_127 = torch.nn.functional.relu(x_126, inplace=True) + x_126 = None + bottleneck_output_30 = torch.conv2d( + x_127, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_127 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ = (None) + x_128 = torch.nn.functional.batch_norm( + bottleneck_output_30, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_30 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ = (None) + x_129 = torch.nn.functional.relu(x_128, inplace=True) + x_128 = None + new_features_30 = torch.conv2d( + x_129, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_129 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ = (None) + concated_features_31 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + ], + 1, + ) + x_130 = torch.nn.functional.batch_norm( + concated_features_31, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_31 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ = (None) + x_131 = torch.nn.functional.relu(x_130, inplace=True) + x_130 = None + bottleneck_output_31 = torch.conv2d( + x_131, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_131 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ = (None) + x_132 = torch.nn.functional.batch_norm( + bottleneck_output_31, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_31 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ = (None) + x_133 = torch.nn.functional.relu(x_132, inplace=True) + x_132 = None + new_features_31 = torch.conv2d( + x_133, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_133 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ = (None) + concated_features_32 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + ], + 1, + ) + x_134 = torch.nn.functional.batch_norm( + concated_features_32, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_32 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ = (None) + x_135 = torch.nn.functional.relu(x_134, inplace=True) + x_134 = None + bottleneck_output_32 = torch.conv2d( + x_135, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_135 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ = (None) + x_136 = torch.nn.functional.batch_norm( + bottleneck_output_32, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_32 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ = (None) + x_137 = torch.nn.functional.relu(x_136, inplace=True) + x_136 = None + new_features_32 = torch.conv2d( + x_137, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_137 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ = (None) + concated_features_33 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + ], + 1, + ) + x_138 = torch.nn.functional.batch_norm( + concated_features_33, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_33 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ = (None) + x_139 = torch.nn.functional.relu(x_138, inplace=True) + x_138 = None + bottleneck_output_33 = torch.conv2d( + x_139, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_139 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ = (None) + x_140 = torch.nn.functional.batch_norm( + bottleneck_output_33, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_33 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ = (None) + x_141 = torch.nn.functional.relu(x_140, inplace=True) + x_140 = None + new_features_33 = torch.conv2d( + x_141, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_141 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ = (None) + concated_features_34 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + ], + 1, + ) + x_142 = torch.nn.functional.batch_norm( + concated_features_34, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_34 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ = (None) + x_143 = torch.nn.functional.relu(x_142, inplace=True) + x_142 = None + bottleneck_output_34 = torch.conv2d( + x_143, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_143 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ = (None) + x_144 = torch.nn.functional.batch_norm( + bottleneck_output_34, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_34 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ = (None) + x_145 = torch.nn.functional.relu(x_144, inplace=True) + x_144 = None + new_features_34 = torch.conv2d( + x_145, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_145 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ = (None) + concated_features_35 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + ], + 1, + ) + x_146 = torch.nn.functional.batch_norm( + concated_features_35, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_35 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ = (None) + x_147 = torch.nn.functional.relu(x_146, inplace=True) + x_146 = None + bottleneck_output_35 = torch.conv2d( + x_147, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_147 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ = (None) + x_148 = torch.nn.functional.batch_norm( + bottleneck_output_35, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_35 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ = (None) + x_149 = torch.nn.functional.relu(x_148, inplace=True) + x_148 = None + new_features_35 = torch.conv2d( + x_149, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_149 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ = (None) + concated_features_36 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + ], + 1, + ) + x_150 = torch.nn.functional.batch_norm( + concated_features_36, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_36 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ = (None) + x_151 = torch.nn.functional.relu(x_150, inplace=True) + x_150 = None + bottleneck_output_36 = torch.conv2d( + x_151, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_151 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ = (None) + x_152 = torch.nn.functional.batch_norm( + bottleneck_output_36, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_36 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ = (None) + x_153 = torch.nn.functional.relu(x_152, inplace=True) + x_152 = None + new_features_36 = torch.conv2d( + x_153, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_153 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ = (None) + concated_features_37 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + ], + 1, + ) + x_154 = torch.nn.functional.batch_norm( + concated_features_37, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_37 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ = (None) + x_155 = torch.nn.functional.relu(x_154, inplace=True) + x_154 = None + bottleneck_output_37 = torch.conv2d( + x_155, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_155 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ = (None) + x_156 = torch.nn.functional.batch_norm( + bottleneck_output_37, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_37 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ = (None) + x_157 = torch.nn.functional.relu(x_156, inplace=True) + x_156 = None + new_features_37 = torch.conv2d( + x_157, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_157 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ = (None) + concated_features_38 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + ], + 1, + ) + x_158 = torch.nn.functional.batch_norm( + concated_features_38, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_38 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ = (None) + x_159 = torch.nn.functional.relu(x_158, inplace=True) + x_158 = None + bottleneck_output_38 = torch.conv2d( + x_159, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_159 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ = (None) + x_160 = torch.nn.functional.batch_norm( + bottleneck_output_38, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_38 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ = (None) + x_161 = torch.nn.functional.relu(x_160, inplace=True) + x_160 = None + new_features_38 = torch.conv2d( + x_161, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_161 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ = (None) + concated_features_39 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + ], + 1, + ) + x_162 = torch.nn.functional.batch_norm( + concated_features_39, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_39 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ = (None) + x_163 = torch.nn.functional.relu(x_162, inplace=True) + x_162 = None + bottleneck_output_39 = torch.conv2d( + x_163, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_163 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ = (None) + x_164 = torch.nn.functional.batch_norm( + bottleneck_output_39, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_39 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ = (None) + x_165 = torch.nn.functional.relu(x_164, inplace=True) + x_164 = None + new_features_39 = torch.conv2d( + x_165, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_165 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ = (None) + concated_features_40 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + ], + 1, + ) + x_166 = torch.nn.functional.batch_norm( + concated_features_40, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_40 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ = (None) + x_167 = torch.nn.functional.relu(x_166, inplace=True) + x_166 = None + bottleneck_output_40 = torch.conv2d( + x_167, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_167 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ = (None) + x_168 = torch.nn.functional.batch_norm( + bottleneck_output_40, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_40 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ = (None) + x_169 = torch.nn.functional.relu(x_168, inplace=True) + x_168 = None + new_features_40 = torch.conv2d( + x_169, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_169 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ = (None) + concated_features_41 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + ], + 1, + ) + x_170 = torch.nn.functional.batch_norm( + concated_features_41, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_41 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ = (None) + x_171 = torch.nn.functional.relu(x_170, inplace=True) + x_170 = None + bottleneck_output_41 = torch.conv2d( + x_171, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_171 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ = (None) + x_172 = torch.nn.functional.batch_norm( + bottleneck_output_41, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_41 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ = (None) + x_173 = torch.nn.functional.relu(x_172, inplace=True) + x_172 = None + new_features_41 = torch.conv2d( + x_173, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_173 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ = (None) + concated_features_42 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + ], + 1, + ) + x_174 = torch.nn.functional.batch_norm( + concated_features_42, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_42 = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_ = (None) + x_175 = torch.nn.functional.relu(x_174, inplace=True) + x_174 = None + bottleneck_output_42 = torch.conv2d( + x_175, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_175 = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_ = (None) + x_176 = torch.nn.functional.batch_norm( + bottleneck_output_42, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_42 = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_ = (None) + x_177 = torch.nn.functional.relu(x_176, inplace=True) + x_176 = None + new_features_42 = torch.conv2d( + x_177, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_177 = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_ = (None) + concated_features_43 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + ], + 1, + ) + x_178 = torch.nn.functional.batch_norm( + concated_features_43, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_43 = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_ = (None) + x_179 = torch.nn.functional.relu(x_178, inplace=True) + x_178 = None + bottleneck_output_43 = torch.conv2d( + x_179, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_179 = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_ = (None) + x_180 = torch.nn.functional.batch_norm( + bottleneck_output_43, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_43 = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_ = (None) + x_181 = torch.nn.functional.relu(x_180, inplace=True) + x_180 = None + new_features_43 = torch.conv2d( + x_181, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_181 = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_ = (None) + concated_features_44 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + ], + 1, + ) + x_182 = torch.nn.functional.batch_norm( + concated_features_44, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_44 = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_ = (None) + x_183 = torch.nn.functional.relu(x_182, inplace=True) + x_182 = None + bottleneck_output_44 = torch.conv2d( + x_183, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_183 = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_ = (None) + x_184 = torch.nn.functional.batch_norm( + bottleneck_output_44, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_44 = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_ = (None) + x_185 = torch.nn.functional.relu(x_184, inplace=True) + x_184 = None + new_features_44 = torch.conv2d( + x_185, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_185 = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_ = (None) + concated_features_45 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + ], + 1, + ) + x_186 = torch.nn.functional.batch_norm( + concated_features_45, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_45 = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_ = (None) + x_187 = torch.nn.functional.relu(x_186, inplace=True) + x_186 = None + bottleneck_output_45 = torch.conv2d( + x_187, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_187 = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_ = (None) + x_188 = torch.nn.functional.batch_norm( + bottleneck_output_45, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_45 = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_ = (None) + x_189 = torch.nn.functional.relu(x_188, inplace=True) + x_188 = None + new_features_45 = torch.conv2d( + x_189, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_189 = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_ = (None) + concated_features_46 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + ], + 1, + ) + x_190 = torch.nn.functional.batch_norm( + concated_features_46, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_46 = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_ = (None) + x_191 = torch.nn.functional.relu(x_190, inplace=True) + x_190 = None + bottleneck_output_46 = torch.conv2d( + x_191, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_191 = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_ = (None) + x_192 = torch.nn.functional.batch_norm( + bottleneck_output_46, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_46 = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_ = (None) + x_193 = torch.nn.functional.relu(x_192, inplace=True) + x_192 = None + new_features_46 = torch.conv2d( + x_193, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_193 = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_ = (None) + concated_features_47 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + ], + 1, + ) + x_194 = torch.nn.functional.batch_norm( + concated_features_47, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_47 = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_ = (None) + x_195 = torch.nn.functional.relu(x_194, inplace=True) + x_194 = None + bottleneck_output_47 = torch.conv2d( + x_195, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_195 = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_ = (None) + x_196 = torch.nn.functional.batch_norm( + bottleneck_output_47, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_47 = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_ = (None) + x_197 = torch.nn.functional.relu(x_196, inplace=True) + x_196 = None + new_features_47 = torch.conv2d( + x_197, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_197 = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_ = (None) + concated_features_48 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + ], + 1, + ) + x_198 = torch.nn.functional.batch_norm( + concated_features_48, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_48 = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_ = (None) + x_199 = torch.nn.functional.relu(x_198, inplace=True) + x_198 = None + bottleneck_output_48 = torch.conv2d( + x_199, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_199 = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_ = (None) + x_200 = torch.nn.functional.batch_norm( + bottleneck_output_48, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_48 = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_ = (None) + x_201 = torch.nn.functional.relu(x_200, inplace=True) + x_200 = None + new_features_48 = torch.conv2d( + x_201, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_201 = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_ = (None) + concated_features_49 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + ], + 1, + ) + x_202 = torch.nn.functional.batch_norm( + concated_features_49, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_49 = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_ = (None) + x_203 = torch.nn.functional.relu(x_202, inplace=True) + x_202 = None + bottleneck_output_49 = torch.conv2d( + x_203, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_203 = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_ = (None) + x_204 = torch.nn.functional.batch_norm( + bottleneck_output_49, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_49 = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_ = (None) + x_205 = torch.nn.functional.relu(x_204, inplace=True) + x_204 = None + new_features_49 = torch.conv2d( + x_205, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_205 = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_ = (None) + input_9 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + ], + 1, + ) + input_8 = ( + new_features_18 + ) = ( + new_features_19 + ) = ( + new_features_20 + ) = ( + new_features_21 + ) = ( + new_features_22 + ) = ( + new_features_23 + ) = ( + new_features_24 + ) = ( + new_features_25 + ) = ( + new_features_26 + ) = ( + new_features_27 + ) = ( + new_features_28 + ) = ( + new_features_29 + ) = ( + new_features_30 + ) = ( + new_features_31 + ) = ( + new_features_32 + ) = ( + new_features_33 + ) = ( + new_features_34 + ) = ( + new_features_35 + ) = ( + new_features_36 + ) = ( + new_features_37 + ) = ( + new_features_38 + ) = ( + new_features_39 + ) = ( + new_features_40 + ) = ( + new_features_41 + ) = ( + new_features_42 + ) = ( + new_features_43 + ) = ( + new_features_44 + ) = ( + new_features_45 + ) = new_features_46 = new_features_47 = new_features_48 = new_features_49 = None + x_206 = torch.nn.functional.batch_norm( + input_9, + l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_9 = l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_ + ) = None + x_207 = torch.nn.functional.relu(x_206, inplace=True) + x_206 = None + input_10 = torch.conv2d( + x_207, + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_207 = ( + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_ + ) = None + input_11 = torch._C._nn.avg_pool2d(input_10, 2, 2, 0, False, True, None) + input_10 = None + concated_features_50 = torch.cat([input_11], 1) + x_208 = torch.nn.functional.batch_norm( + concated_features_50, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_50 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_209 = torch.nn.functional.relu(x_208, inplace=True) + x_208 = None + bottleneck_output_50 = torch.conv2d( + x_209, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_209 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_210 = torch.nn.functional.batch_norm( + bottleneck_output_50, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_50 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_211 = torch.nn.functional.relu(x_210, inplace=True) + x_210 = None + new_features_50 = torch.conv2d( + x_211, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_211 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_51 = torch.cat([input_11, new_features_50], 1) + x_212 = torch.nn.functional.batch_norm( + concated_features_51, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_51 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_213 = torch.nn.functional.relu(x_212, inplace=True) + x_212 = None + bottleneck_output_51 = torch.conv2d( + x_213, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_213 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_214 = torch.nn.functional.batch_norm( + bottleneck_output_51, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_51 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_215 = torch.nn.functional.relu(x_214, inplace=True) + x_214 = None + new_features_51 = torch.conv2d( + x_215, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_215 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_52 = torch.cat( + [input_11, new_features_50, new_features_51], 1 + ) + x_216 = torch.nn.functional.batch_norm( + concated_features_52, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_52 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_217 = torch.nn.functional.relu(x_216, inplace=True) + x_216 = None + bottleneck_output_52 = torch.conv2d( + x_217, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_217 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_218 = torch.nn.functional.batch_norm( + bottleneck_output_52, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_52 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_219 = torch.nn.functional.relu(x_218, inplace=True) + x_218 = None + new_features_52 = torch.conv2d( + x_219, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_219 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_53 = torch.cat( + [input_11, new_features_50, new_features_51, new_features_52], 1 + ) + x_220 = torch.nn.functional.batch_norm( + concated_features_53, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_53 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_221 = torch.nn.functional.relu(x_220, inplace=True) + x_220 = None + bottleneck_output_53 = torch.conv2d( + x_221, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_221 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_222 = torch.nn.functional.batch_norm( + bottleneck_output_53, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_53 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_223 = torch.nn.functional.relu(x_222, inplace=True) + x_222 = None + new_features_53 = torch.conv2d( + x_223, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_223 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_54 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + ], + 1, + ) + x_224 = torch.nn.functional.batch_norm( + concated_features_54, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_54 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_225 = torch.nn.functional.relu(x_224, inplace=True) + x_224 = None + bottleneck_output_54 = torch.conv2d( + x_225, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_225 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_226 = torch.nn.functional.batch_norm( + bottleneck_output_54, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_54 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_227 = torch.nn.functional.relu(x_226, inplace=True) + x_226 = None + new_features_54 = torch.conv2d( + x_227, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_227 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_55 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + ], + 1, + ) + x_228 = torch.nn.functional.batch_norm( + concated_features_55, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_55 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_229 = torch.nn.functional.relu(x_228, inplace=True) + x_228 = None + bottleneck_output_55 = torch.conv2d( + x_229, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_229 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_230 = torch.nn.functional.batch_norm( + bottleneck_output_55, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_55 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_231 = torch.nn.functional.relu(x_230, inplace=True) + x_230 = None + new_features_55 = torch.conv2d( + x_231, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_231 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_56 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + ], + 1, + ) + x_232 = torch.nn.functional.batch_norm( + concated_features_56, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_56 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_233 = torch.nn.functional.relu(x_232, inplace=True) + x_232 = None + bottleneck_output_56 = torch.conv2d( + x_233, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_233 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_234 = torch.nn.functional.batch_norm( + bottleneck_output_56, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_56 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_235 = torch.nn.functional.relu(x_234, inplace=True) + x_234 = None + new_features_56 = torch.conv2d( + x_235, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_235 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_57 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + ], + 1, + ) + x_236 = torch.nn.functional.batch_norm( + concated_features_57, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_57 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_237 = torch.nn.functional.relu(x_236, inplace=True) + x_236 = None + bottleneck_output_57 = torch.conv2d( + x_237, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_237 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_238 = torch.nn.functional.batch_norm( + bottleneck_output_57, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_57 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_239 = torch.nn.functional.relu(x_238, inplace=True) + x_238 = None + new_features_57 = torch.conv2d( + x_239, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_239 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_58 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + ], + 1, + ) + x_240 = torch.nn.functional.batch_norm( + concated_features_58, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_58 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_241 = torch.nn.functional.relu(x_240, inplace=True) + x_240 = None + bottleneck_output_58 = torch.conv2d( + x_241, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_241 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_242 = torch.nn.functional.batch_norm( + bottleneck_output_58, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_58 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_243 = torch.nn.functional.relu(x_242, inplace=True) + x_242 = None + new_features_58 = torch.conv2d( + x_243, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_243 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_59 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + ], + 1, + ) + x_244 = torch.nn.functional.batch_norm( + concated_features_59, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_59 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_245 = torch.nn.functional.relu(x_244, inplace=True) + x_244 = None + bottleneck_output_59 = torch.conv2d( + x_245, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_245 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_246 = torch.nn.functional.batch_norm( + bottleneck_output_59, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_59 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_247 = torch.nn.functional.relu(x_246, inplace=True) + x_246 = None + new_features_59 = torch.conv2d( + x_247, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_247 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_60 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + ], + 1, + ) + x_248 = torch.nn.functional.batch_norm( + concated_features_60, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_60 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_249 = torch.nn.functional.relu(x_248, inplace=True) + x_248 = None + bottleneck_output_60 = torch.conv2d( + x_249, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_249 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_250 = torch.nn.functional.batch_norm( + bottleneck_output_60, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_60 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_251 = torch.nn.functional.relu(x_250, inplace=True) + x_250 = None + new_features_60 = torch.conv2d( + x_251, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_251 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_61 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + ], + 1, + ) + x_252 = torch.nn.functional.batch_norm( + concated_features_61, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_61 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_253 = torch.nn.functional.relu(x_252, inplace=True) + x_252 = None + bottleneck_output_61 = torch.conv2d( + x_253, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_253 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_254 = torch.nn.functional.batch_norm( + bottleneck_output_61, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_61 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_255 = torch.nn.functional.relu(x_254, inplace=True) + x_254 = None + new_features_61 = torch.conv2d( + x_255, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_255 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + concated_features_62 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + ], + 1, + ) + x_256 = torch.nn.functional.batch_norm( + concated_features_62, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_62 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ = (None) + x_257 = torch.nn.functional.relu(x_256, inplace=True) + x_256 = None + bottleneck_output_62 = torch.conv2d( + x_257, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_257 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ = (None) + x_258 = torch.nn.functional.batch_norm( + bottleneck_output_62, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_62 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ = (None) + x_259 = torch.nn.functional.relu(x_258, inplace=True) + x_258 = None + new_features_62 = torch.conv2d( + x_259, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_259 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ = (None) + concated_features_63 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + ], + 1, + ) + x_260 = torch.nn.functional.batch_norm( + concated_features_63, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_63 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ = (None) + x_261 = torch.nn.functional.relu(x_260, inplace=True) + x_260 = None + bottleneck_output_63 = torch.conv2d( + x_261, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_261 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ = (None) + x_262 = torch.nn.functional.batch_norm( + bottleneck_output_63, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_63 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ = (None) + x_263 = torch.nn.functional.relu(x_262, inplace=True) + x_262 = None + new_features_63 = torch.conv2d( + x_263, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_263 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ = (None) + concated_features_64 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + ], + 1, + ) + x_264 = torch.nn.functional.batch_norm( + concated_features_64, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_64 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ = (None) + x_265 = torch.nn.functional.relu(x_264, inplace=True) + x_264 = None + bottleneck_output_64 = torch.conv2d( + x_265, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_265 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ = (None) + x_266 = torch.nn.functional.batch_norm( + bottleneck_output_64, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_64 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ = (None) + x_267 = torch.nn.functional.relu(x_266, inplace=True) + x_266 = None + new_features_64 = torch.conv2d( + x_267, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_267 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ = (None) + concated_features_65 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + ], + 1, + ) + x_268 = torch.nn.functional.batch_norm( + concated_features_65, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_65 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ = (None) + x_269 = torch.nn.functional.relu(x_268, inplace=True) + x_268 = None + bottleneck_output_65 = torch.conv2d( + x_269, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_269 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ = (None) + x_270 = torch.nn.functional.batch_norm( + bottleneck_output_65, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_65 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ = (None) + x_271 = torch.nn.functional.relu(x_270, inplace=True) + x_270 = None + new_features_65 = torch.conv2d( + x_271, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_271 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ = (None) + concated_features_66 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + ], + 1, + ) + x_272 = torch.nn.functional.batch_norm( + concated_features_66, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_66 = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_ = (None) + x_273 = torch.nn.functional.relu(x_272, inplace=True) + x_272 = None + bottleneck_output_66 = torch.conv2d( + x_273, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_273 = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_ = (None) + x_274 = torch.nn.functional.batch_norm( + bottleneck_output_66, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_66 = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_ = (None) + x_275 = torch.nn.functional.relu(x_274, inplace=True) + x_274 = None + new_features_66 = torch.conv2d( + x_275, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_275 = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_ = (None) + concated_features_67 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + ], + 1, + ) + x_276 = torch.nn.functional.batch_norm( + concated_features_67, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_67 = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_ = (None) + x_277 = torch.nn.functional.relu(x_276, inplace=True) + x_276 = None + bottleneck_output_67 = torch.conv2d( + x_277, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_277 = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_ = (None) + x_278 = torch.nn.functional.batch_norm( + bottleneck_output_67, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_67 = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_ = (None) + x_279 = torch.nn.functional.relu(x_278, inplace=True) + x_278 = None + new_features_67 = torch.conv2d( + x_279, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_279 = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_ = (None) + concated_features_68 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + ], + 1, + ) + x_280 = torch.nn.functional.batch_norm( + concated_features_68, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_68 = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_ = (None) + x_281 = torch.nn.functional.relu(x_280, inplace=True) + x_280 = None + bottleneck_output_68 = torch.conv2d( + x_281, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_281 = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_ = (None) + x_282 = torch.nn.functional.batch_norm( + bottleneck_output_68, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_68 = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_ = (None) + x_283 = torch.nn.functional.relu(x_282, inplace=True) + x_282 = None + new_features_68 = torch.conv2d( + x_283, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_283 = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_ = (None) + concated_features_69 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + ], + 1, + ) + x_284 = torch.nn.functional.batch_norm( + concated_features_69, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_69 = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_ = (None) + x_285 = torch.nn.functional.relu(x_284, inplace=True) + x_284 = None + bottleneck_output_69 = torch.conv2d( + x_285, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_285 = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_ = (None) + x_286 = torch.nn.functional.batch_norm( + bottleneck_output_69, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_69 = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_ = (None) + x_287 = torch.nn.functional.relu(x_286, inplace=True) + x_286 = None + new_features_69 = torch.conv2d( + x_287, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_287 = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_ = (None) + concated_features_70 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + ], + 1, + ) + x_288 = torch.nn.functional.batch_norm( + concated_features_70, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_70 = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_ = (None) + x_289 = torch.nn.functional.relu(x_288, inplace=True) + x_288 = None + bottleneck_output_70 = torch.conv2d( + x_289, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_289 = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_ = (None) + x_290 = torch.nn.functional.batch_norm( + bottleneck_output_70, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_70 = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_ = (None) + x_291 = torch.nn.functional.relu(x_290, inplace=True) + x_290 = None + new_features_70 = torch.conv2d( + x_291, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_291 = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_ = (None) + concated_features_71 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + ], + 1, + ) + x_292 = torch.nn.functional.batch_norm( + concated_features_71, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_71 = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_ = (None) + x_293 = torch.nn.functional.relu(x_292, inplace=True) + x_292 = None + bottleneck_output_71 = torch.conv2d( + x_293, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_293 = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_ = (None) + x_294 = torch.nn.functional.batch_norm( + bottleneck_output_71, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_71 = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_ = (None) + x_295 = torch.nn.functional.relu(x_294, inplace=True) + x_294 = None + new_features_71 = torch.conv2d( + x_295, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_295 = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_ = (None) + concated_features_72 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + ], + 1, + ) + x_296 = torch.nn.functional.batch_norm( + concated_features_72, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_72 = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_ = (None) + x_297 = torch.nn.functional.relu(x_296, inplace=True) + x_296 = None + bottleneck_output_72 = torch.conv2d( + x_297, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_297 = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_ = (None) + x_298 = torch.nn.functional.batch_norm( + bottleneck_output_72, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_72 = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_ = (None) + x_299 = torch.nn.functional.relu(x_298, inplace=True) + x_298 = None + new_features_72 = torch.conv2d( + x_299, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_299 = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_ = (None) + concated_features_73 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + ], + 1, + ) + x_300 = torch.nn.functional.batch_norm( + concated_features_73, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_73 = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_ = (None) + x_301 = torch.nn.functional.relu(x_300, inplace=True) + x_300 = None + bottleneck_output_73 = torch.conv2d( + x_301, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_301 = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_ = (None) + x_302 = torch.nn.functional.batch_norm( + bottleneck_output_73, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_73 = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_ = (None) + x_303 = torch.nn.functional.relu(x_302, inplace=True) + x_302 = None + new_features_73 = torch.conv2d( + x_303, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_303 = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_ = (None) + concated_features_74 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + ], + 1, + ) + x_304 = torch.nn.functional.batch_norm( + concated_features_74, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_74 = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_ = (None) + x_305 = torch.nn.functional.relu(x_304, inplace=True) + x_304 = None + bottleneck_output_74 = torch.conv2d( + x_305, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_305 = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_ = (None) + x_306 = torch.nn.functional.batch_norm( + bottleneck_output_74, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_74 = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_ = (None) + x_307 = torch.nn.functional.relu(x_306, inplace=True) + x_306 = None + new_features_74 = torch.conv2d( + x_307, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_307 = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_ = (None) + concated_features_75 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + ], + 1, + ) + x_308 = torch.nn.functional.batch_norm( + concated_features_75, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_75 = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_ = (None) + x_309 = torch.nn.functional.relu(x_308, inplace=True) + x_308 = None + bottleneck_output_75 = torch.conv2d( + x_309, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_309 = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_ = (None) + x_310 = torch.nn.functional.batch_norm( + bottleneck_output_75, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_75 = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_ = (None) + x_311 = torch.nn.functional.relu(x_310, inplace=True) + x_310 = None + new_features_75 = torch.conv2d( + x_311, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_311 = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_ = (None) + concated_features_76 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + ], + 1, + ) + x_312 = torch.nn.functional.batch_norm( + concated_features_76, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_76 = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_ = (None) + x_313 = torch.nn.functional.relu(x_312, inplace=True) + x_312 = None + bottleneck_output_76 = torch.conv2d( + x_313, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_313 = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_ = (None) + x_314 = torch.nn.functional.batch_norm( + bottleneck_output_76, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_76 = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_ = (None) + x_315 = torch.nn.functional.relu(x_314, inplace=True) + x_314 = None + new_features_76 = torch.conv2d( + x_315, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_315 = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_ = (None) + concated_features_77 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + ], + 1, + ) + x_316 = torch.nn.functional.batch_norm( + concated_features_77, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_77 = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_ = (None) + x_317 = torch.nn.functional.relu(x_316, inplace=True) + x_316 = None + bottleneck_output_77 = torch.conv2d( + x_317, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_317 = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_ = (None) + x_318 = torch.nn.functional.batch_norm( + bottleneck_output_77, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_77 = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_ = (None) + x_319 = torch.nn.functional.relu(x_318, inplace=True) + x_318 = None + new_features_77 = torch.conv2d( + x_319, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_319 = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_ = (None) + concated_features_78 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + ], + 1, + ) + x_320 = torch.nn.functional.batch_norm( + concated_features_78, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_78 = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_ = (None) + x_321 = torch.nn.functional.relu(x_320, inplace=True) + x_320 = None + bottleneck_output_78 = torch.conv2d( + x_321, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_321 = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_ = (None) + x_322 = torch.nn.functional.batch_norm( + bottleneck_output_78, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_78 = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_ = (None) + x_323 = torch.nn.functional.relu(x_322, inplace=True) + x_322 = None + new_features_78 = torch.conv2d( + x_323, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_323 = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_ = (None) + concated_features_79 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + ], + 1, + ) + x_324 = torch.nn.functional.batch_norm( + concated_features_79, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_79 = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_ = (None) + x_325 = torch.nn.functional.relu(x_324, inplace=True) + x_324 = None + bottleneck_output_79 = torch.conv2d( + x_325, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_325 = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_ = (None) + x_326 = torch.nn.functional.batch_norm( + bottleneck_output_79, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_79 = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_ = (None) + x_327 = torch.nn.functional.relu(x_326, inplace=True) + x_326 = None + new_features_79 = torch.conv2d( + x_327, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_327 = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_ = (None) + concated_features_80 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + ], + 1, + ) + x_328 = torch.nn.functional.batch_norm( + concated_features_80, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_80 = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_ = (None) + x_329 = torch.nn.functional.relu(x_328, inplace=True) + x_328 = None + bottleneck_output_80 = torch.conv2d( + x_329, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_329 = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_ = (None) + x_330 = torch.nn.functional.batch_norm( + bottleneck_output_80, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_80 = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_ = (None) + x_331 = torch.nn.functional.relu(x_330, inplace=True) + x_330 = None + new_features_80 = torch.conv2d( + x_331, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_331 = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_ = (None) + concated_features_81 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + ], + 1, + ) + x_332 = torch.nn.functional.batch_norm( + concated_features_81, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_81 = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_ = (None) + x_333 = torch.nn.functional.relu(x_332, inplace=True) + x_332 = None + bottleneck_output_81 = torch.conv2d( + x_333, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_333 = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_ = (None) + x_334 = torch.nn.functional.batch_norm( + bottleneck_output_81, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_81 = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_ = (None) + x_335 = torch.nn.functional.relu(x_334, inplace=True) + x_334 = None + new_features_81 = torch.conv2d( + x_335, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_335 = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_ = (None) + input_12 = torch.cat( + [ + input_11, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + ], + 1, + ) + input_11 = ( + new_features_50 + ) = ( + new_features_51 + ) = ( + new_features_52 + ) = ( + new_features_53 + ) = ( + new_features_54 + ) = ( + new_features_55 + ) = ( + new_features_56 + ) = ( + new_features_57 + ) = ( + new_features_58 + ) = ( + new_features_59 + ) = ( + new_features_60 + ) = ( + new_features_61 + ) = ( + new_features_62 + ) = ( + new_features_63 + ) = ( + new_features_64 + ) = ( + new_features_65 + ) = ( + new_features_66 + ) = ( + new_features_67 + ) = ( + new_features_68 + ) = ( + new_features_69 + ) = ( + new_features_70 + ) = ( + new_features_71 + ) = ( + new_features_72 + ) = ( + new_features_73 + ) = ( + new_features_74 + ) = ( + new_features_75 + ) = ( + new_features_76 + ) = ( + new_features_77 + ) = new_features_78 = new_features_79 = new_features_80 = new_features_81 = None + x_336 = torch.nn.functional.batch_norm( + input_12, + l_self_modules_features_modules_norm5_buffers_running_mean_, + l_self_modules_features_modules_norm5_buffers_running_var_, + l_self_modules_features_modules_norm5_parameters_weight_, + l_self_modules_features_modules_norm5_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_12 = ( + l_self_modules_features_modules_norm5_buffers_running_mean_ + ) = ( + l_self_modules_features_modules_norm5_buffers_running_var_ + ) = ( + l_self_modules_features_modules_norm5_parameters_weight_ + ) = l_self_modules_features_modules_norm5_parameters_bias_ = None + x_337 = torch.nn.functional.relu(x_336, inplace=True) + x_336 = None + x_338 = torch.nn.functional.adaptive_avg_pool2d(x_337, 1) + x_337 = None + x_339 = x_338.flatten(1, -1) + x_338 = None + x_340 = torch.nn.functional.dropout(x_339, 0.0, False, False) + x_339 = None + x_341 = torch._C._nn.linear( + x_340, + l_self_modules_classifier_parameters_weight_, + l_self_modules_classifier_parameters_bias_, + ) + x_340 = ( + l_self_modules_classifier_parameters_weight_ + ) = l_self_modules_classifier_parameters_bias_ = None + return (x_341,) diff --git a/samples/timm/densenet169.tv_in1k/weight_meta.py b/samples/timm/densenet169.tv_in1k/weight_meta.py new file mode 100644 index 0000000000..4c5dd6babc --- /dev/null +++ b/samples/timm/densenet169.tv_in1k/weight_meta.py @@ -0,0 +1,8490 @@ +class Program_weight_tensor_meta_L_self_modules_features_modules_conv0_parameters_weight_: + name = "L_self_modules_features_modules_conv0_parameters_weight_" + shape = [64, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.119 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.227 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_buffers_running_mean_: + name = "L_self_modules_features_modules_norm0_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_buffers_running_var_: + name = "L_self_modules_features_modules_norm0_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_parameters_weight_: + name = "L_self_modules_features_modules_norm0_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_parameters_bias_: + name = "L_self_modules_features_modules_norm0_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.144 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.123 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 224, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.095 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition1_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.124 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 224, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.095 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 288, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.084 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [128, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [128, 352, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.076 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [128, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [128, 416, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.069 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [128, 448, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.067 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [128, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition2_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition2_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition2_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 288, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.084 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 352, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.075 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.073 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 416, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.069 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [128, 448, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.067 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [128, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [128, 544, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.061 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [128, 576, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [128, 608, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.057 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_" + shape = [128, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_" + shape = [128, 672, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.055 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_" + shape = [128, 704, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_" + shape = [128, 736, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.052 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_" + shape = [128, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_" + shape = [128, 800, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.050 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_" + shape = [128, 832, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.049 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_" + shape = [128, 864, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_" + shape = [128, 896, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_" + shape = [128, 928, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_" + shape = [128, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_" + shape = [128, 992, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.045 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_" + shape = [128, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_" + shape = [128, 1056, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_" + shape = [128, 1088, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_" + shape = [128, 1120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_" + shape = [128, 1152, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_" + shape = [128, 1184, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_" + shape = [128, 1216, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_" + shape = [128, 1248, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_" + ) + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_" + ) + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition3_modules_norm_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition3_modules_norm_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition3_modules_conv_parameters_weight_" + shape = [640, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.039 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 672, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.055 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 704, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 736, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.052 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 800, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.050 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [128, 832, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.049 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [128, 864, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [128, 896, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [128, 928, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [128, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [128, 992, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.045 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_" + shape = [128, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_" + shape = [128, 1056, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_" + shape = [128, 1088, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_" + shape = [128, 1120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_" + shape = [128, 1152, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_" + shape = [128, 1184, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_" + shape = [128, 1216, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_" + shape = [128, 1248, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_" + shape = [128, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_" + shape = [1312] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_" + shape = [1312] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_" + shape = [1312] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_" + shape = [1312] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_" + shape = [128, 1312, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.039 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_" + shape = [128, 1344, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.039 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_" + shape = [1376] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_" + shape = [1376] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_" + shape = [1376] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_" + shape = [1376] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_" + shape = [128, 1376, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.038 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_" + shape = [128, 1408, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.038 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_" + shape = [128, 1440, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_" + shape = [1472] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_" + shape = [1472] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_" + shape = [1472] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_" + shape = [1472] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_" + shape = [128, 1472, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_" + shape = [1504] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_" + shape = [1504] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_" + shape = [1504] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_" + shape = [1504] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_" + shape = [128, 1504, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_" + shape = [128, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_" + shape = [1568] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_" + shape = [1568] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_" + shape = [1568] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_" + shape = [1568] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_" + shape = [128, 1568, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_" + shape = [1600] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_" + shape = [1600] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_" + shape = [1600] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_" + shape = [1600] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_" + shape = [128, 1600, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.035 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_" + shape = [128, 1632, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.035 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_buffers_running_mean_: + name = "L_self_modules_features_modules_norm5_buffers_running_mean_" + shape = [1664] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_buffers_running_var_: + name = "L_self_modules_features_modules_norm5_buffers_running_var_" + shape = [1664] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_parameters_weight_: + name = "L_self_modules_features_modules_norm5_parameters_weight_" + shape = [1664] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_parameters_bias_: + name = "L_self_modules_features_modules_norm5_parameters_bias_" + shape = [1664] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_classifier_parameters_weight_: + name = "L_self_modules_classifier_parameters_weight_" + shape = [1000, 1664] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.014 + data = None + + +class Program_weight_tensor_meta_L_self_modules_classifier_parameters_bias_: + name = "L_self_modules_classifier_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/densenet201.tv_in1k/graph_hash.txt b/samples/timm/densenet201.tv_in1k/graph_hash.txt new file mode 100644 index 0000000000..d8dae08729 --- /dev/null +++ b/samples/timm/densenet201.tv_in1k/graph_hash.txt @@ -0,0 +1 @@ +36ec146b47b470780e08cd74b50727801ccadde5421d7ed89ea47ca7ff326e59 \ No newline at end of file diff --git a/samples/timm/densenet201.tv_in1k/graph_net.json b/samples/timm/densenet201.tv_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/densenet201.tv_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/densenet201.tv_in1k/input_meta.py b/samples/timm/densenet201.tv_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/densenet201.tv_in1k/input_tensor_constraints.py b/samples/timm/densenet201.tv_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/densenet201.tv_in1k/model.py b/samples/timm/densenet201.tv_in1k/model.py new file mode 100644 index 0000000000..3861018bf1 --- /dev/null +++ b/samples/timm/densenet201.tv_in1k/model.py @@ -0,0 +1,9205 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_features_modules_conv0_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_features_modules_norm0_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_norm0_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_norm0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm5_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_norm5_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_norm5_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm5_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_classifier_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_classifier_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_features_modules_conv0_parameters_weight_ = ( + L_self_modules_features_modules_conv0_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_features_modules_norm0_buffers_running_mean_ = ( + L_self_modules_features_modules_norm0_buffers_running_mean_ + ) + l_self_modules_features_modules_norm0_buffers_running_var_ = ( + L_self_modules_features_modules_norm0_buffers_running_var_ + ) + l_self_modules_features_modules_norm0_parameters_weight_ = ( + L_self_modules_features_modules_norm0_parameters_weight_ + ) + l_self_modules_features_modules_norm0_parameters_bias_ = ( + L_self_modules_features_modules_norm0_parameters_bias_ + ) + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition1_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition1_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition1_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition2_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition2_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition2_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition3_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition3_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition3_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_ + l_self_modules_features_modules_norm5_buffers_running_mean_ = ( + L_self_modules_features_modules_norm5_buffers_running_mean_ + ) + l_self_modules_features_modules_norm5_buffers_running_var_ = ( + L_self_modules_features_modules_norm5_buffers_running_var_ + ) + l_self_modules_features_modules_norm5_parameters_weight_ = ( + L_self_modules_features_modules_norm5_parameters_weight_ + ) + l_self_modules_features_modules_norm5_parameters_bias_ = ( + L_self_modules_features_modules_norm5_parameters_bias_ + ) + l_self_modules_classifier_parameters_weight_ = ( + L_self_modules_classifier_parameters_weight_ + ) + l_self_modules_classifier_parameters_bias_ = ( + L_self_modules_classifier_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_features_modules_conv0_parameters_weight_, + None, + (2, 2), + (3, 3), + (1, 1), + 1, + ) + l_x_ = l_self_modules_features_modules_conv0_parameters_weight_ = None + x = torch.nn.functional.batch_norm( + input_1, + l_self_modules_features_modules_norm0_buffers_running_mean_, + l_self_modules_features_modules_norm0_buffers_running_var_, + l_self_modules_features_modules_norm0_parameters_weight_, + l_self_modules_features_modules_norm0_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_1 = ( + l_self_modules_features_modules_norm0_buffers_running_mean_ + ) = ( + l_self_modules_features_modules_norm0_buffers_running_var_ + ) = ( + l_self_modules_features_modules_norm0_parameters_weight_ + ) = l_self_modules_features_modules_norm0_parameters_bias_ = None + x_1 = torch.nn.functional.relu(x, inplace=True) + x = None + input_2 = torch.nn.functional.max_pool2d( + x_1, 3, 2, 1, 1, ceil_mode=False, return_indices=False + ) + x_1 = None + concated_features = torch.cat([input_2], 1) + x_2 = torch.nn.functional.batch_norm( + concated_features, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_3 = torch.nn.functional.relu(x_2, inplace=True) + x_2 = None + bottleneck_output = torch.conv2d( + x_3, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_3 = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + bottleneck_output, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_5 = torch.nn.functional.relu(x_4, inplace=True) + x_4 = None + new_features = torch.conv2d( + x_5, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_5 = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_1 = torch.cat([input_2, new_features], 1) + x_6 = torch.nn.functional.batch_norm( + concated_features_1, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_1 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_7 = torch.nn.functional.relu(x_6, inplace=True) + x_6 = None + bottleneck_output_1 = torch.conv2d( + x_7, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_7 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_8 = torch.nn.functional.batch_norm( + bottleneck_output_1, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_1 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_9 = torch.nn.functional.relu(x_8, inplace=True) + x_8 = None + new_features_1 = torch.conv2d( + x_9, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_9 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_2 = torch.cat([input_2, new_features, new_features_1], 1) + x_10 = torch.nn.functional.batch_norm( + concated_features_2, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_2 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_11 = torch.nn.functional.relu(x_10, inplace=True) + x_10 = None + bottleneck_output_2 = torch.conv2d( + x_11, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_11 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_12 = torch.nn.functional.batch_norm( + bottleneck_output_2, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_2 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_13 = torch.nn.functional.relu(x_12, inplace=True) + x_12 = None + new_features_2 = torch.conv2d( + x_13, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_13 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_3 = torch.cat( + [input_2, new_features, new_features_1, new_features_2], 1 + ) + x_14 = torch.nn.functional.batch_norm( + concated_features_3, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_3 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_15 = torch.nn.functional.relu(x_14, inplace=True) + x_14 = None + bottleneck_output_3 = torch.conv2d( + x_15, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_15 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_16 = torch.nn.functional.batch_norm( + bottleneck_output_3, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_3 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_17 = torch.nn.functional.relu(x_16, inplace=True) + x_16 = None + new_features_3 = torch.conv2d( + x_17, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_17 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_4 = torch.cat( + [input_2, new_features, new_features_1, new_features_2, new_features_3], 1 + ) + x_18 = torch.nn.functional.batch_norm( + concated_features_4, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_4 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_19 = torch.nn.functional.relu(x_18, inplace=True) + x_18 = None + bottleneck_output_4 = torch.conv2d( + x_19, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_19 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_20 = torch.nn.functional.batch_norm( + bottleneck_output_4, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_4 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_21 = torch.nn.functional.relu(x_20, inplace=True) + x_20 = None + new_features_4 = torch.conv2d( + x_21, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_21 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_5 = torch.cat( + [ + input_2, + new_features, + new_features_1, + new_features_2, + new_features_3, + new_features_4, + ], + 1, + ) + x_22 = torch.nn.functional.batch_norm( + concated_features_5, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_5 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_23 = torch.nn.functional.relu(x_22, inplace=True) + x_22 = None + bottleneck_output_5 = torch.conv2d( + x_23, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_23 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_24 = torch.nn.functional.batch_norm( + bottleneck_output_5, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_5 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_25 = torch.nn.functional.relu(x_24, inplace=True) + x_24 = None + new_features_5 = torch.conv2d( + x_25, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_25 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + input_3 = torch.cat( + [ + input_2, + new_features, + new_features_1, + new_features_2, + new_features_3, + new_features_4, + new_features_5, + ], + 1, + ) + input_2 = ( + new_features + ) = ( + new_features_1 + ) = new_features_2 = new_features_3 = new_features_4 = new_features_5 = None + x_26 = torch.nn.functional.batch_norm( + input_3, + l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_3 = l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_ + ) = None + x_27 = torch.nn.functional.relu(x_26, inplace=True) + x_26 = None + input_4 = torch.conv2d( + x_27, + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_27 = ( + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_ + ) = None + input_5 = torch._C._nn.avg_pool2d(input_4, 2, 2, 0, False, True, None) + input_4 = None + concated_features_6 = torch.cat([input_5], 1) + x_28 = torch.nn.functional.batch_norm( + concated_features_6, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_6 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_29 = torch.nn.functional.relu(x_28, inplace=True) + x_28 = None + bottleneck_output_6 = torch.conv2d( + x_29, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_29 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_30 = torch.nn.functional.batch_norm( + bottleneck_output_6, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_6 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_31 = torch.nn.functional.relu(x_30, inplace=True) + x_30 = None + new_features_6 = torch.conv2d( + x_31, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_31 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_7 = torch.cat([input_5, new_features_6], 1) + x_32 = torch.nn.functional.batch_norm( + concated_features_7, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_7 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_33 = torch.nn.functional.relu(x_32, inplace=True) + x_32 = None + bottleneck_output_7 = torch.conv2d( + x_33, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_33 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_34 = torch.nn.functional.batch_norm( + bottleneck_output_7, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_7 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_35 = torch.nn.functional.relu(x_34, inplace=True) + x_34 = None + new_features_7 = torch.conv2d( + x_35, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_35 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_8 = torch.cat([input_5, new_features_6, new_features_7], 1) + x_36 = torch.nn.functional.batch_norm( + concated_features_8, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_8 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_37 = torch.nn.functional.relu(x_36, inplace=True) + x_36 = None + bottleneck_output_8 = torch.conv2d( + x_37, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_37 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_38 = torch.nn.functional.batch_norm( + bottleneck_output_8, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_8 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_39 = torch.nn.functional.relu(x_38, inplace=True) + x_38 = None + new_features_8 = torch.conv2d( + x_39, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_39 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_9 = torch.cat( + [input_5, new_features_6, new_features_7, new_features_8], 1 + ) + x_40 = torch.nn.functional.batch_norm( + concated_features_9, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_9 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_41 = torch.nn.functional.relu(x_40, inplace=True) + x_40 = None + bottleneck_output_9 = torch.conv2d( + x_41, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_41 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_42 = torch.nn.functional.batch_norm( + bottleneck_output_9, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_9 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_43 = torch.nn.functional.relu(x_42, inplace=True) + x_42 = None + new_features_9 = torch.conv2d( + x_43, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_43 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_10 = torch.cat( + [input_5, new_features_6, new_features_7, new_features_8, new_features_9], 1 + ) + x_44 = torch.nn.functional.batch_norm( + concated_features_10, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_10 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_45 = torch.nn.functional.relu(x_44, inplace=True) + x_44 = None + bottleneck_output_10 = torch.conv2d( + x_45, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_45 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_46 = torch.nn.functional.batch_norm( + bottleneck_output_10, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_10 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_47 = torch.nn.functional.relu(x_46, inplace=True) + x_46 = None + new_features_10 = torch.conv2d( + x_47, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_47 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_11 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + ], + 1, + ) + x_48 = torch.nn.functional.batch_norm( + concated_features_11, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_11 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_49 = torch.nn.functional.relu(x_48, inplace=True) + x_48 = None + bottleneck_output_11 = torch.conv2d( + x_49, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_49 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_50 = torch.nn.functional.batch_norm( + bottleneck_output_11, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_11 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_51 = torch.nn.functional.relu(x_50, inplace=True) + x_50 = None + new_features_11 = torch.conv2d( + x_51, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_51 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_12 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + ], + 1, + ) + x_52 = torch.nn.functional.batch_norm( + concated_features_12, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_12 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_53 = torch.nn.functional.relu(x_52, inplace=True) + x_52 = None + bottleneck_output_12 = torch.conv2d( + x_53, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_53 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_54 = torch.nn.functional.batch_norm( + bottleneck_output_12, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_12 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_55 = torch.nn.functional.relu(x_54, inplace=True) + x_54 = None + new_features_12 = torch.conv2d( + x_55, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_55 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_13 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + ], + 1, + ) + x_56 = torch.nn.functional.batch_norm( + concated_features_13, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_13 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_57 = torch.nn.functional.relu(x_56, inplace=True) + x_56 = None + bottleneck_output_13 = torch.conv2d( + x_57, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_57 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_58 = torch.nn.functional.batch_norm( + bottleneck_output_13, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_13 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_59 = torch.nn.functional.relu(x_58, inplace=True) + x_58 = None + new_features_13 = torch.conv2d( + x_59, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_59 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_14 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + ], + 1, + ) + x_60 = torch.nn.functional.batch_norm( + concated_features_14, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_14 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_61 = torch.nn.functional.relu(x_60, inplace=True) + x_60 = None + bottleneck_output_14 = torch.conv2d( + x_61, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_61 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_62 = torch.nn.functional.batch_norm( + bottleneck_output_14, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_14 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_63 = torch.nn.functional.relu(x_62, inplace=True) + x_62 = None + new_features_14 = torch.conv2d( + x_63, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_63 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_15 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + ], + 1, + ) + x_64 = torch.nn.functional.batch_norm( + concated_features_15, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_15 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_65 = torch.nn.functional.relu(x_64, inplace=True) + x_64 = None + bottleneck_output_15 = torch.conv2d( + x_65, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_65 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_66 = torch.nn.functional.batch_norm( + bottleneck_output_15, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_15 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_67 = torch.nn.functional.relu(x_66, inplace=True) + x_66 = None + new_features_15 = torch.conv2d( + x_67, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_67 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_16 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + ], + 1, + ) + x_68 = torch.nn.functional.batch_norm( + concated_features_16, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_16 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_69 = torch.nn.functional.relu(x_68, inplace=True) + x_68 = None + bottleneck_output_16 = torch.conv2d( + x_69, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_69 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_70 = torch.nn.functional.batch_norm( + bottleneck_output_16, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_16 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_71 = torch.nn.functional.relu(x_70, inplace=True) + x_70 = None + new_features_16 = torch.conv2d( + x_71, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_71 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_17 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + new_features_16, + ], + 1, + ) + x_72 = torch.nn.functional.batch_norm( + concated_features_17, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_17 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_73 = torch.nn.functional.relu(x_72, inplace=True) + x_72 = None + bottleneck_output_17 = torch.conv2d( + x_73, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_73 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_74 = torch.nn.functional.batch_norm( + bottleneck_output_17, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_17 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_75 = torch.nn.functional.relu(x_74, inplace=True) + x_74 = None + new_features_17 = torch.conv2d( + x_75, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_75 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + input_6 = torch.cat( + [ + input_5, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + new_features_16, + new_features_17, + ], + 1, + ) + input_5 = ( + new_features_6 + ) = ( + new_features_7 + ) = ( + new_features_8 + ) = ( + new_features_9 + ) = ( + new_features_10 + ) = ( + new_features_11 + ) = ( + new_features_12 + ) = ( + new_features_13 + ) = new_features_14 = new_features_15 = new_features_16 = new_features_17 = None + x_76 = torch.nn.functional.batch_norm( + input_6, + l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_6 = l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_ + ) = None + x_77 = torch.nn.functional.relu(x_76, inplace=True) + x_76 = None + input_7 = torch.conv2d( + x_77, + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_77 = ( + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_ + ) = None + input_8 = torch._C._nn.avg_pool2d(input_7, 2, 2, 0, False, True, None) + input_7 = None + concated_features_18 = torch.cat([input_8], 1) + x_78 = torch.nn.functional.batch_norm( + concated_features_18, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_18 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_79 = torch.nn.functional.relu(x_78, inplace=True) + x_78 = None + bottleneck_output_18 = torch.conv2d( + x_79, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_79 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_80 = torch.nn.functional.batch_norm( + bottleneck_output_18, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_18 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_81 = torch.nn.functional.relu(x_80, inplace=True) + x_80 = None + new_features_18 = torch.conv2d( + x_81, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_81 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_19 = torch.cat([input_8, new_features_18], 1) + x_82 = torch.nn.functional.batch_norm( + concated_features_19, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_19 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_83 = torch.nn.functional.relu(x_82, inplace=True) + x_82 = None + bottleneck_output_19 = torch.conv2d( + x_83, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_83 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_84 = torch.nn.functional.batch_norm( + bottleneck_output_19, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_19 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_85 = torch.nn.functional.relu(x_84, inplace=True) + x_84 = None + new_features_19 = torch.conv2d( + x_85, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_85 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_20 = torch.cat([input_8, new_features_18, new_features_19], 1) + x_86 = torch.nn.functional.batch_norm( + concated_features_20, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_20 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_87 = torch.nn.functional.relu(x_86, inplace=True) + x_86 = None + bottleneck_output_20 = torch.conv2d( + x_87, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_87 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_88 = torch.nn.functional.batch_norm( + bottleneck_output_20, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_20 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_89 = torch.nn.functional.relu(x_88, inplace=True) + x_88 = None + new_features_20 = torch.conv2d( + x_89, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_89 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_21 = torch.cat( + [input_8, new_features_18, new_features_19, new_features_20], 1 + ) + x_90 = torch.nn.functional.batch_norm( + concated_features_21, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_21 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_91 = torch.nn.functional.relu(x_90, inplace=True) + x_90 = None + bottleneck_output_21 = torch.conv2d( + x_91, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_91 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_92 = torch.nn.functional.batch_norm( + bottleneck_output_21, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_21 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_93 = torch.nn.functional.relu(x_92, inplace=True) + x_92 = None + new_features_21 = torch.conv2d( + x_93, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_93 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_22 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + ], + 1, + ) + x_94 = torch.nn.functional.batch_norm( + concated_features_22, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_22 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_95 = torch.nn.functional.relu(x_94, inplace=True) + x_94 = None + bottleneck_output_22 = torch.conv2d( + x_95, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_95 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_96 = torch.nn.functional.batch_norm( + bottleneck_output_22, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_22 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_97 = torch.nn.functional.relu(x_96, inplace=True) + x_96 = None + new_features_22 = torch.conv2d( + x_97, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_97 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_23 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + ], + 1, + ) + x_98 = torch.nn.functional.batch_norm( + concated_features_23, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_23 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_99 = torch.nn.functional.relu(x_98, inplace=True) + x_98 = None + bottleneck_output_23 = torch.conv2d( + x_99, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_99 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_100 = torch.nn.functional.batch_norm( + bottleneck_output_23, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_23 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_101 = torch.nn.functional.relu(x_100, inplace=True) + x_100 = None + new_features_23 = torch.conv2d( + x_101, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_101 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_24 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + ], + 1, + ) + x_102 = torch.nn.functional.batch_norm( + concated_features_24, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_24 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_103 = torch.nn.functional.relu(x_102, inplace=True) + x_102 = None + bottleneck_output_24 = torch.conv2d( + x_103, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_103 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_104 = torch.nn.functional.batch_norm( + bottleneck_output_24, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_24 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_105 = torch.nn.functional.relu(x_104, inplace=True) + x_104 = None + new_features_24 = torch.conv2d( + x_105, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_105 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_25 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + ], + 1, + ) + x_106 = torch.nn.functional.batch_norm( + concated_features_25, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_25 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_107 = torch.nn.functional.relu(x_106, inplace=True) + x_106 = None + bottleneck_output_25 = torch.conv2d( + x_107, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_107 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_108 = torch.nn.functional.batch_norm( + bottleneck_output_25, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_25 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_109 = torch.nn.functional.relu(x_108, inplace=True) + x_108 = None + new_features_25 = torch.conv2d( + x_109, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_109 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_26 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + ], + 1, + ) + x_110 = torch.nn.functional.batch_norm( + concated_features_26, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_26 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_111 = torch.nn.functional.relu(x_110, inplace=True) + x_110 = None + bottleneck_output_26 = torch.conv2d( + x_111, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_111 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_112 = torch.nn.functional.batch_norm( + bottleneck_output_26, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_26 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_113 = torch.nn.functional.relu(x_112, inplace=True) + x_112 = None + new_features_26 = torch.conv2d( + x_113, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_113 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_27 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + ], + 1, + ) + x_114 = torch.nn.functional.batch_norm( + concated_features_27, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_27 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_115 = torch.nn.functional.relu(x_114, inplace=True) + x_114 = None + bottleneck_output_27 = torch.conv2d( + x_115, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_115 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_116 = torch.nn.functional.batch_norm( + bottleneck_output_27, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_27 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_117 = torch.nn.functional.relu(x_116, inplace=True) + x_116 = None + new_features_27 = torch.conv2d( + x_117, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_117 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_28 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + ], + 1, + ) + x_118 = torch.nn.functional.batch_norm( + concated_features_28, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_28 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_119 = torch.nn.functional.relu(x_118, inplace=True) + x_118 = None + bottleneck_output_28 = torch.conv2d( + x_119, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_119 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_120 = torch.nn.functional.batch_norm( + bottleneck_output_28, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_28 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_121 = torch.nn.functional.relu(x_120, inplace=True) + x_120 = None + new_features_28 = torch.conv2d( + x_121, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_121 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_29 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + ], + 1, + ) + x_122 = torch.nn.functional.batch_norm( + concated_features_29, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_29 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_123 = torch.nn.functional.relu(x_122, inplace=True) + x_122 = None + bottleneck_output_29 = torch.conv2d( + x_123, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_123 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_124 = torch.nn.functional.batch_norm( + bottleneck_output_29, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_29 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_125 = torch.nn.functional.relu(x_124, inplace=True) + x_124 = None + new_features_29 = torch.conv2d( + x_125, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_125 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + concated_features_30 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + ], + 1, + ) + x_126 = torch.nn.functional.batch_norm( + concated_features_30, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_30 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ = (None) + x_127 = torch.nn.functional.relu(x_126, inplace=True) + x_126 = None + bottleneck_output_30 = torch.conv2d( + x_127, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_127 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ = (None) + x_128 = torch.nn.functional.batch_norm( + bottleneck_output_30, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_30 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ = (None) + x_129 = torch.nn.functional.relu(x_128, inplace=True) + x_128 = None + new_features_30 = torch.conv2d( + x_129, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_129 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ = (None) + concated_features_31 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + ], + 1, + ) + x_130 = torch.nn.functional.batch_norm( + concated_features_31, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_31 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ = (None) + x_131 = torch.nn.functional.relu(x_130, inplace=True) + x_130 = None + bottleneck_output_31 = torch.conv2d( + x_131, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_131 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ = (None) + x_132 = torch.nn.functional.batch_norm( + bottleneck_output_31, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_31 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ = (None) + x_133 = torch.nn.functional.relu(x_132, inplace=True) + x_132 = None + new_features_31 = torch.conv2d( + x_133, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_133 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ = (None) + concated_features_32 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + ], + 1, + ) + x_134 = torch.nn.functional.batch_norm( + concated_features_32, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_32 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ = (None) + x_135 = torch.nn.functional.relu(x_134, inplace=True) + x_134 = None + bottleneck_output_32 = torch.conv2d( + x_135, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_135 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ = (None) + x_136 = torch.nn.functional.batch_norm( + bottleneck_output_32, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_32 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ = (None) + x_137 = torch.nn.functional.relu(x_136, inplace=True) + x_136 = None + new_features_32 = torch.conv2d( + x_137, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_137 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ = (None) + concated_features_33 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + ], + 1, + ) + x_138 = torch.nn.functional.batch_norm( + concated_features_33, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_33 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ = (None) + x_139 = torch.nn.functional.relu(x_138, inplace=True) + x_138 = None + bottleneck_output_33 = torch.conv2d( + x_139, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_139 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ = (None) + x_140 = torch.nn.functional.batch_norm( + bottleneck_output_33, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_33 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ = (None) + x_141 = torch.nn.functional.relu(x_140, inplace=True) + x_140 = None + new_features_33 = torch.conv2d( + x_141, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_141 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ = (None) + concated_features_34 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + ], + 1, + ) + x_142 = torch.nn.functional.batch_norm( + concated_features_34, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_34 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ = (None) + x_143 = torch.nn.functional.relu(x_142, inplace=True) + x_142 = None + bottleneck_output_34 = torch.conv2d( + x_143, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_143 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ = (None) + x_144 = torch.nn.functional.batch_norm( + bottleneck_output_34, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_34 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ = (None) + x_145 = torch.nn.functional.relu(x_144, inplace=True) + x_144 = None + new_features_34 = torch.conv2d( + x_145, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_145 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ = (None) + concated_features_35 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + ], + 1, + ) + x_146 = torch.nn.functional.batch_norm( + concated_features_35, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_35 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ = (None) + x_147 = torch.nn.functional.relu(x_146, inplace=True) + x_146 = None + bottleneck_output_35 = torch.conv2d( + x_147, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_147 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ = (None) + x_148 = torch.nn.functional.batch_norm( + bottleneck_output_35, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_35 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ = (None) + x_149 = torch.nn.functional.relu(x_148, inplace=True) + x_148 = None + new_features_35 = torch.conv2d( + x_149, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_149 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ = (None) + concated_features_36 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + ], + 1, + ) + x_150 = torch.nn.functional.batch_norm( + concated_features_36, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_36 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ = (None) + x_151 = torch.nn.functional.relu(x_150, inplace=True) + x_150 = None + bottleneck_output_36 = torch.conv2d( + x_151, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_151 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ = (None) + x_152 = torch.nn.functional.batch_norm( + bottleneck_output_36, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_36 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ = (None) + x_153 = torch.nn.functional.relu(x_152, inplace=True) + x_152 = None + new_features_36 = torch.conv2d( + x_153, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_153 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ = (None) + concated_features_37 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + ], + 1, + ) + x_154 = torch.nn.functional.batch_norm( + concated_features_37, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_37 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ = (None) + x_155 = torch.nn.functional.relu(x_154, inplace=True) + x_154 = None + bottleneck_output_37 = torch.conv2d( + x_155, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_155 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ = (None) + x_156 = torch.nn.functional.batch_norm( + bottleneck_output_37, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_37 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ = (None) + x_157 = torch.nn.functional.relu(x_156, inplace=True) + x_156 = None + new_features_37 = torch.conv2d( + x_157, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_157 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ = (None) + concated_features_38 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + ], + 1, + ) + x_158 = torch.nn.functional.batch_norm( + concated_features_38, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_38 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ = (None) + x_159 = torch.nn.functional.relu(x_158, inplace=True) + x_158 = None + bottleneck_output_38 = torch.conv2d( + x_159, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_159 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ = (None) + x_160 = torch.nn.functional.batch_norm( + bottleneck_output_38, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_38 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ = (None) + x_161 = torch.nn.functional.relu(x_160, inplace=True) + x_160 = None + new_features_38 = torch.conv2d( + x_161, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_161 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ = (None) + concated_features_39 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + ], + 1, + ) + x_162 = torch.nn.functional.batch_norm( + concated_features_39, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_39 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ = (None) + x_163 = torch.nn.functional.relu(x_162, inplace=True) + x_162 = None + bottleneck_output_39 = torch.conv2d( + x_163, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_163 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ = (None) + x_164 = torch.nn.functional.batch_norm( + bottleneck_output_39, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_39 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ = (None) + x_165 = torch.nn.functional.relu(x_164, inplace=True) + x_164 = None + new_features_39 = torch.conv2d( + x_165, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_165 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ = (None) + concated_features_40 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + ], + 1, + ) + x_166 = torch.nn.functional.batch_norm( + concated_features_40, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_40 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ = (None) + x_167 = torch.nn.functional.relu(x_166, inplace=True) + x_166 = None + bottleneck_output_40 = torch.conv2d( + x_167, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_167 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ = (None) + x_168 = torch.nn.functional.batch_norm( + bottleneck_output_40, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_40 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ = (None) + x_169 = torch.nn.functional.relu(x_168, inplace=True) + x_168 = None + new_features_40 = torch.conv2d( + x_169, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_169 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ = (None) + concated_features_41 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + ], + 1, + ) + x_170 = torch.nn.functional.batch_norm( + concated_features_41, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_41 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ = (None) + x_171 = torch.nn.functional.relu(x_170, inplace=True) + x_170 = None + bottleneck_output_41 = torch.conv2d( + x_171, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_171 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ = (None) + x_172 = torch.nn.functional.batch_norm( + bottleneck_output_41, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_41 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ = (None) + x_173 = torch.nn.functional.relu(x_172, inplace=True) + x_172 = None + new_features_41 = torch.conv2d( + x_173, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_173 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ = (None) + concated_features_42 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + ], + 1, + ) + x_174 = torch.nn.functional.batch_norm( + concated_features_42, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_42 = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_ = (None) + x_175 = torch.nn.functional.relu(x_174, inplace=True) + x_174 = None + bottleneck_output_42 = torch.conv2d( + x_175, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_175 = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_ = (None) + x_176 = torch.nn.functional.batch_norm( + bottleneck_output_42, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_42 = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_ = (None) + x_177 = torch.nn.functional.relu(x_176, inplace=True) + x_176 = None + new_features_42 = torch.conv2d( + x_177, + l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_177 = l_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_ = (None) + concated_features_43 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + ], + 1, + ) + x_178 = torch.nn.functional.batch_norm( + concated_features_43, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_43 = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_ = (None) + x_179 = torch.nn.functional.relu(x_178, inplace=True) + x_178 = None + bottleneck_output_43 = torch.conv2d( + x_179, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_179 = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_ = (None) + x_180 = torch.nn.functional.batch_norm( + bottleneck_output_43, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_43 = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_ = (None) + x_181 = torch.nn.functional.relu(x_180, inplace=True) + x_180 = None + new_features_43 = torch.conv2d( + x_181, + l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_181 = l_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_ = (None) + concated_features_44 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + ], + 1, + ) + x_182 = torch.nn.functional.batch_norm( + concated_features_44, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_44 = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_ = (None) + x_183 = torch.nn.functional.relu(x_182, inplace=True) + x_182 = None + bottleneck_output_44 = torch.conv2d( + x_183, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_183 = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_ = (None) + x_184 = torch.nn.functional.batch_norm( + bottleneck_output_44, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_44 = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_ = (None) + x_185 = torch.nn.functional.relu(x_184, inplace=True) + x_184 = None + new_features_44 = torch.conv2d( + x_185, + l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_185 = l_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_ = (None) + concated_features_45 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + ], + 1, + ) + x_186 = torch.nn.functional.batch_norm( + concated_features_45, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_45 = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_ = (None) + x_187 = torch.nn.functional.relu(x_186, inplace=True) + x_186 = None + bottleneck_output_45 = torch.conv2d( + x_187, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_187 = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_ = (None) + x_188 = torch.nn.functional.batch_norm( + bottleneck_output_45, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_45 = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_ = (None) + x_189 = torch.nn.functional.relu(x_188, inplace=True) + x_188 = None + new_features_45 = torch.conv2d( + x_189, + l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_189 = l_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_ = (None) + concated_features_46 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + ], + 1, + ) + x_190 = torch.nn.functional.batch_norm( + concated_features_46, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_46 = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_ = (None) + x_191 = torch.nn.functional.relu(x_190, inplace=True) + x_190 = None + bottleneck_output_46 = torch.conv2d( + x_191, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_191 = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_ = (None) + x_192 = torch.nn.functional.batch_norm( + bottleneck_output_46, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_46 = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_ = (None) + x_193 = torch.nn.functional.relu(x_192, inplace=True) + x_192 = None + new_features_46 = torch.conv2d( + x_193, + l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_193 = l_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_ = (None) + concated_features_47 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + ], + 1, + ) + x_194 = torch.nn.functional.batch_norm( + concated_features_47, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_47 = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_ = (None) + x_195 = torch.nn.functional.relu(x_194, inplace=True) + x_194 = None + bottleneck_output_47 = torch.conv2d( + x_195, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_195 = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_ = (None) + x_196 = torch.nn.functional.batch_norm( + bottleneck_output_47, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_47 = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_ = (None) + x_197 = torch.nn.functional.relu(x_196, inplace=True) + x_196 = None + new_features_47 = torch.conv2d( + x_197, + l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_197 = l_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_ = (None) + concated_features_48 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + ], + 1, + ) + x_198 = torch.nn.functional.batch_norm( + concated_features_48, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_48 = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_ = (None) + x_199 = torch.nn.functional.relu(x_198, inplace=True) + x_198 = None + bottleneck_output_48 = torch.conv2d( + x_199, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_199 = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_ = (None) + x_200 = torch.nn.functional.batch_norm( + bottleneck_output_48, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_48 = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_ = (None) + x_201 = torch.nn.functional.relu(x_200, inplace=True) + x_200 = None + new_features_48 = torch.conv2d( + x_201, + l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_201 = l_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_ = (None) + concated_features_49 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + ], + 1, + ) + x_202 = torch.nn.functional.batch_norm( + concated_features_49, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_49 = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_ = (None) + x_203 = torch.nn.functional.relu(x_202, inplace=True) + x_202 = None + bottleneck_output_49 = torch.conv2d( + x_203, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_203 = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_ = (None) + x_204 = torch.nn.functional.batch_norm( + bottleneck_output_49, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_49 = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_ = (None) + x_205 = torch.nn.functional.relu(x_204, inplace=True) + x_204 = None + new_features_49 = torch.conv2d( + x_205, + l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_205 = l_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_ = (None) + concated_features_50 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + ], + 1, + ) + x_206 = torch.nn.functional.batch_norm( + concated_features_50, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_50 = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_ = (None) + x_207 = torch.nn.functional.relu(x_206, inplace=True) + x_206 = None + bottleneck_output_50 = torch.conv2d( + x_207, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_207 = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_ = (None) + x_208 = torch.nn.functional.batch_norm( + bottleneck_output_50, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_50 = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_ = (None) + x_209 = torch.nn.functional.relu(x_208, inplace=True) + x_208 = None + new_features_50 = torch.conv2d( + x_209, + l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_209 = l_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_ = (None) + concated_features_51 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + ], + 1, + ) + x_210 = torch.nn.functional.batch_norm( + concated_features_51, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_51 = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_ = (None) + x_211 = torch.nn.functional.relu(x_210, inplace=True) + x_210 = None + bottleneck_output_51 = torch.conv2d( + x_211, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_211 = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_ = (None) + x_212 = torch.nn.functional.batch_norm( + bottleneck_output_51, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_51 = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_ = (None) + x_213 = torch.nn.functional.relu(x_212, inplace=True) + x_212 = None + new_features_51 = torch.conv2d( + x_213, + l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_213 = l_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_ = (None) + concated_features_52 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + ], + 1, + ) + x_214 = torch.nn.functional.batch_norm( + concated_features_52, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_52 = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_ = (None) + x_215 = torch.nn.functional.relu(x_214, inplace=True) + x_214 = None + bottleneck_output_52 = torch.conv2d( + x_215, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_215 = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_ = (None) + x_216 = torch.nn.functional.batch_norm( + bottleneck_output_52, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_52 = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_ = (None) + x_217 = torch.nn.functional.relu(x_216, inplace=True) + x_216 = None + new_features_52 = torch.conv2d( + x_217, + l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_217 = l_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_ = (None) + concated_features_53 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + ], + 1, + ) + x_218 = torch.nn.functional.batch_norm( + concated_features_53, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_53 = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_ = (None) + x_219 = torch.nn.functional.relu(x_218, inplace=True) + x_218 = None + bottleneck_output_53 = torch.conv2d( + x_219, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_219 = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_ = (None) + x_220 = torch.nn.functional.batch_norm( + bottleneck_output_53, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_53 = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_ = (None) + x_221 = torch.nn.functional.relu(x_220, inplace=True) + x_220 = None + new_features_53 = torch.conv2d( + x_221, + l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_221 = l_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_ = (None) + concated_features_54 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + ], + 1, + ) + x_222 = torch.nn.functional.batch_norm( + concated_features_54, + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_54 = l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_bias_ = (None) + x_223 = torch.nn.functional.relu(x_222, inplace=True) + x_222 = None + bottleneck_output_54 = torch.conv2d( + x_223, + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_223 = l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv1_parameters_weight_ = (None) + x_224 = torch.nn.functional.batch_norm( + bottleneck_output_54, + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_54 = l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_bias_ = (None) + x_225 = torch.nn.functional.relu(x_224, inplace=True) + x_224 = None + new_features_54 = torch.conv2d( + x_225, + l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_225 = l_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv2_parameters_weight_ = (None) + concated_features_55 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + ], + 1, + ) + x_226 = torch.nn.functional.batch_norm( + concated_features_55, + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_55 = l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_bias_ = (None) + x_227 = torch.nn.functional.relu(x_226, inplace=True) + x_226 = None + bottleneck_output_55 = torch.conv2d( + x_227, + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_227 = l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv1_parameters_weight_ = (None) + x_228 = torch.nn.functional.batch_norm( + bottleneck_output_55, + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_55 = l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_bias_ = (None) + x_229 = torch.nn.functional.relu(x_228, inplace=True) + x_228 = None + new_features_55 = torch.conv2d( + x_229, + l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_229 = l_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv2_parameters_weight_ = (None) + concated_features_56 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + ], + 1, + ) + x_230 = torch.nn.functional.batch_norm( + concated_features_56, + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_56 = l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_bias_ = (None) + x_231 = torch.nn.functional.relu(x_230, inplace=True) + x_230 = None + bottleneck_output_56 = torch.conv2d( + x_231, + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_231 = l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv1_parameters_weight_ = (None) + x_232 = torch.nn.functional.batch_norm( + bottleneck_output_56, + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_56 = l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_bias_ = (None) + x_233 = torch.nn.functional.relu(x_232, inplace=True) + x_232 = None + new_features_56 = torch.conv2d( + x_233, + l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_233 = l_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv2_parameters_weight_ = (None) + concated_features_57 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + ], + 1, + ) + x_234 = torch.nn.functional.batch_norm( + concated_features_57, + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_57 = l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_bias_ = (None) + x_235 = torch.nn.functional.relu(x_234, inplace=True) + x_234 = None + bottleneck_output_57 = torch.conv2d( + x_235, + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_235 = l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv1_parameters_weight_ = (None) + x_236 = torch.nn.functional.batch_norm( + bottleneck_output_57, + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_57 = l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_bias_ = (None) + x_237 = torch.nn.functional.relu(x_236, inplace=True) + x_236 = None + new_features_57 = torch.conv2d( + x_237, + l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_237 = l_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv2_parameters_weight_ = (None) + concated_features_58 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + ], + 1, + ) + x_238 = torch.nn.functional.batch_norm( + concated_features_58, + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_58 = l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_bias_ = (None) + x_239 = torch.nn.functional.relu(x_238, inplace=True) + x_238 = None + bottleneck_output_58 = torch.conv2d( + x_239, + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_239 = l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv1_parameters_weight_ = (None) + x_240 = torch.nn.functional.batch_norm( + bottleneck_output_58, + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_58 = l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_bias_ = (None) + x_241 = torch.nn.functional.relu(x_240, inplace=True) + x_240 = None + new_features_58 = torch.conv2d( + x_241, + l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_241 = l_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv2_parameters_weight_ = (None) + concated_features_59 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + ], + 1, + ) + x_242 = torch.nn.functional.batch_norm( + concated_features_59, + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_59 = l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_bias_ = (None) + x_243 = torch.nn.functional.relu(x_242, inplace=True) + x_242 = None + bottleneck_output_59 = torch.conv2d( + x_243, + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_243 = l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv1_parameters_weight_ = (None) + x_244 = torch.nn.functional.batch_norm( + bottleneck_output_59, + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_59 = l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_bias_ = (None) + x_245 = torch.nn.functional.relu(x_244, inplace=True) + x_244 = None + new_features_59 = torch.conv2d( + x_245, + l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_245 = l_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv2_parameters_weight_ = (None) + concated_features_60 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + ], + 1, + ) + x_246 = torch.nn.functional.batch_norm( + concated_features_60, + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_60 = l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_bias_ = (None) + x_247 = torch.nn.functional.relu(x_246, inplace=True) + x_246 = None + bottleneck_output_60 = torch.conv2d( + x_247, + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_247 = l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv1_parameters_weight_ = (None) + x_248 = torch.nn.functional.batch_norm( + bottleneck_output_60, + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_60 = l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_bias_ = (None) + x_249 = torch.nn.functional.relu(x_248, inplace=True) + x_248 = None + new_features_60 = torch.conv2d( + x_249, + l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_249 = l_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv2_parameters_weight_ = (None) + concated_features_61 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + ], + 1, + ) + x_250 = torch.nn.functional.batch_norm( + concated_features_61, + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_61 = l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_bias_ = (None) + x_251 = torch.nn.functional.relu(x_250, inplace=True) + x_250 = None + bottleneck_output_61 = torch.conv2d( + x_251, + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_251 = l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv1_parameters_weight_ = (None) + x_252 = torch.nn.functional.batch_norm( + bottleneck_output_61, + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_61 = l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_bias_ = (None) + x_253 = torch.nn.functional.relu(x_252, inplace=True) + x_252 = None + new_features_61 = torch.conv2d( + x_253, + l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_253 = l_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv2_parameters_weight_ = (None) + concated_features_62 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + ], + 1, + ) + x_254 = torch.nn.functional.batch_norm( + concated_features_62, + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_62 = l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_bias_ = (None) + x_255 = torch.nn.functional.relu(x_254, inplace=True) + x_254 = None + bottleneck_output_62 = torch.conv2d( + x_255, + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_255 = l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv1_parameters_weight_ = (None) + x_256 = torch.nn.functional.batch_norm( + bottleneck_output_62, + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_62 = l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_bias_ = (None) + x_257 = torch.nn.functional.relu(x_256, inplace=True) + x_256 = None + new_features_62 = torch.conv2d( + x_257, + l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_257 = l_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv2_parameters_weight_ = (None) + concated_features_63 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + ], + 1, + ) + x_258 = torch.nn.functional.batch_norm( + concated_features_63, + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_63 = l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_bias_ = (None) + x_259 = torch.nn.functional.relu(x_258, inplace=True) + x_258 = None + bottleneck_output_63 = torch.conv2d( + x_259, + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_259 = l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv1_parameters_weight_ = (None) + x_260 = torch.nn.functional.batch_norm( + bottleneck_output_63, + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_63 = l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_bias_ = (None) + x_261 = torch.nn.functional.relu(x_260, inplace=True) + x_260 = None + new_features_63 = torch.conv2d( + x_261, + l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_261 = l_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv2_parameters_weight_ = (None) + concated_features_64 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + ], + 1, + ) + x_262 = torch.nn.functional.batch_norm( + concated_features_64, + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_64 = l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_bias_ = (None) + x_263 = torch.nn.functional.relu(x_262, inplace=True) + x_262 = None + bottleneck_output_64 = torch.conv2d( + x_263, + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_263 = l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv1_parameters_weight_ = (None) + x_264 = torch.nn.functional.batch_norm( + bottleneck_output_64, + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_64 = l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_bias_ = (None) + x_265 = torch.nn.functional.relu(x_264, inplace=True) + x_264 = None + new_features_64 = torch.conv2d( + x_265, + l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_265 = l_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv2_parameters_weight_ = (None) + concated_features_65 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + ], + 1, + ) + x_266 = torch.nn.functional.batch_norm( + concated_features_65, + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_65 = l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_bias_ = (None) + x_267 = torch.nn.functional.relu(x_266, inplace=True) + x_266 = None + bottleneck_output_65 = torch.conv2d( + x_267, + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_267 = l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv1_parameters_weight_ = (None) + x_268 = torch.nn.functional.batch_norm( + bottleneck_output_65, + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_65 = l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_bias_ = (None) + x_269 = torch.nn.functional.relu(x_268, inplace=True) + x_268 = None + new_features_65 = torch.conv2d( + x_269, + l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_269 = l_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv2_parameters_weight_ = (None) + input_9 = torch.cat( + [ + input_8, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + new_features_58, + new_features_59, + new_features_60, + new_features_61, + new_features_62, + new_features_63, + new_features_64, + new_features_65, + ], + 1, + ) + input_8 = ( + new_features_18 + ) = ( + new_features_19 + ) = ( + new_features_20 + ) = ( + new_features_21 + ) = ( + new_features_22 + ) = ( + new_features_23 + ) = ( + new_features_24 + ) = ( + new_features_25 + ) = ( + new_features_26 + ) = ( + new_features_27 + ) = ( + new_features_28 + ) = ( + new_features_29 + ) = ( + new_features_30 + ) = ( + new_features_31 + ) = ( + new_features_32 + ) = ( + new_features_33 + ) = ( + new_features_34 + ) = ( + new_features_35 + ) = ( + new_features_36 + ) = ( + new_features_37 + ) = ( + new_features_38 + ) = ( + new_features_39 + ) = ( + new_features_40 + ) = ( + new_features_41 + ) = ( + new_features_42 + ) = ( + new_features_43 + ) = ( + new_features_44 + ) = ( + new_features_45 + ) = ( + new_features_46 + ) = ( + new_features_47 + ) = ( + new_features_48 + ) = ( + new_features_49 + ) = ( + new_features_50 + ) = ( + new_features_51 + ) = ( + new_features_52 + ) = ( + new_features_53 + ) = ( + new_features_54 + ) = ( + new_features_55 + ) = ( + new_features_56 + ) = ( + new_features_57 + ) = ( + new_features_58 + ) = ( + new_features_59 + ) = ( + new_features_60 + ) = ( + new_features_61 + ) = new_features_62 = new_features_63 = new_features_64 = new_features_65 = None + x_270 = torch.nn.functional.batch_norm( + input_9, + l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_9 = l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_ + ) = None + x_271 = torch.nn.functional.relu(x_270, inplace=True) + x_270 = None + input_10 = torch.conv2d( + x_271, + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_271 = ( + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_ + ) = None + input_11 = torch._C._nn.avg_pool2d(input_10, 2, 2, 0, False, True, None) + input_10 = None + concated_features_66 = torch.cat([input_11], 1) + x_272 = torch.nn.functional.batch_norm( + concated_features_66, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_66 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_273 = torch.nn.functional.relu(x_272, inplace=True) + x_272 = None + bottleneck_output_66 = torch.conv2d( + x_273, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_273 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_274 = torch.nn.functional.batch_norm( + bottleneck_output_66, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_66 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_275 = torch.nn.functional.relu(x_274, inplace=True) + x_274 = None + new_features_66 = torch.conv2d( + x_275, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_275 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_67 = torch.cat([input_11, new_features_66], 1) + x_276 = torch.nn.functional.batch_norm( + concated_features_67, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_67 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_277 = torch.nn.functional.relu(x_276, inplace=True) + x_276 = None + bottleneck_output_67 = torch.conv2d( + x_277, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_277 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_278 = torch.nn.functional.batch_norm( + bottleneck_output_67, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_67 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_279 = torch.nn.functional.relu(x_278, inplace=True) + x_278 = None + new_features_67 = torch.conv2d( + x_279, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_279 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_68 = torch.cat( + [input_11, new_features_66, new_features_67], 1 + ) + x_280 = torch.nn.functional.batch_norm( + concated_features_68, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_68 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_281 = torch.nn.functional.relu(x_280, inplace=True) + x_280 = None + bottleneck_output_68 = torch.conv2d( + x_281, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_281 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_282 = torch.nn.functional.batch_norm( + bottleneck_output_68, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_68 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_283 = torch.nn.functional.relu(x_282, inplace=True) + x_282 = None + new_features_68 = torch.conv2d( + x_283, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_283 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_69 = torch.cat( + [input_11, new_features_66, new_features_67, new_features_68], 1 + ) + x_284 = torch.nn.functional.batch_norm( + concated_features_69, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_69 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_285 = torch.nn.functional.relu(x_284, inplace=True) + x_284 = None + bottleneck_output_69 = torch.conv2d( + x_285, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_285 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_286 = torch.nn.functional.batch_norm( + bottleneck_output_69, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_69 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_287 = torch.nn.functional.relu(x_286, inplace=True) + x_286 = None + new_features_69 = torch.conv2d( + x_287, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_287 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_70 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + ], + 1, + ) + x_288 = torch.nn.functional.batch_norm( + concated_features_70, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_70 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_289 = torch.nn.functional.relu(x_288, inplace=True) + x_288 = None + bottleneck_output_70 = torch.conv2d( + x_289, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_289 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_290 = torch.nn.functional.batch_norm( + bottleneck_output_70, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_70 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_291 = torch.nn.functional.relu(x_290, inplace=True) + x_290 = None + new_features_70 = torch.conv2d( + x_291, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_291 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_71 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + ], + 1, + ) + x_292 = torch.nn.functional.batch_norm( + concated_features_71, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_71 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_293 = torch.nn.functional.relu(x_292, inplace=True) + x_292 = None + bottleneck_output_71 = torch.conv2d( + x_293, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_293 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_294 = torch.nn.functional.batch_norm( + bottleneck_output_71, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_71 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_295 = torch.nn.functional.relu(x_294, inplace=True) + x_294 = None + new_features_71 = torch.conv2d( + x_295, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_295 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_72 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + ], + 1, + ) + x_296 = torch.nn.functional.batch_norm( + concated_features_72, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_72 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_297 = torch.nn.functional.relu(x_296, inplace=True) + x_296 = None + bottleneck_output_72 = torch.conv2d( + x_297, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_297 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_298 = torch.nn.functional.batch_norm( + bottleneck_output_72, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_72 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_299 = torch.nn.functional.relu(x_298, inplace=True) + x_298 = None + new_features_72 = torch.conv2d( + x_299, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_299 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_73 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + ], + 1, + ) + x_300 = torch.nn.functional.batch_norm( + concated_features_73, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_73 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_301 = torch.nn.functional.relu(x_300, inplace=True) + x_300 = None + bottleneck_output_73 = torch.conv2d( + x_301, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_301 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_302 = torch.nn.functional.batch_norm( + bottleneck_output_73, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_73 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_303 = torch.nn.functional.relu(x_302, inplace=True) + x_302 = None + new_features_73 = torch.conv2d( + x_303, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_303 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_74 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + ], + 1, + ) + x_304 = torch.nn.functional.batch_norm( + concated_features_74, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_74 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_305 = torch.nn.functional.relu(x_304, inplace=True) + x_304 = None + bottleneck_output_74 = torch.conv2d( + x_305, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_305 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_306 = torch.nn.functional.batch_norm( + bottleneck_output_74, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_74 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_307 = torch.nn.functional.relu(x_306, inplace=True) + x_306 = None + new_features_74 = torch.conv2d( + x_307, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_307 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_75 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + ], + 1, + ) + x_308 = torch.nn.functional.batch_norm( + concated_features_75, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_75 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_309 = torch.nn.functional.relu(x_308, inplace=True) + x_308 = None + bottleneck_output_75 = torch.conv2d( + x_309, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_309 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_310 = torch.nn.functional.batch_norm( + bottleneck_output_75, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_75 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_311 = torch.nn.functional.relu(x_310, inplace=True) + x_310 = None + new_features_75 = torch.conv2d( + x_311, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_311 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_76 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + ], + 1, + ) + x_312 = torch.nn.functional.batch_norm( + concated_features_76, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_76 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_313 = torch.nn.functional.relu(x_312, inplace=True) + x_312 = None + bottleneck_output_76 = torch.conv2d( + x_313, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_313 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_314 = torch.nn.functional.batch_norm( + bottleneck_output_76, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_76 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_315 = torch.nn.functional.relu(x_314, inplace=True) + x_314 = None + new_features_76 = torch.conv2d( + x_315, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_315 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_77 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + ], + 1, + ) + x_316 = torch.nn.functional.batch_norm( + concated_features_77, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_77 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_317 = torch.nn.functional.relu(x_316, inplace=True) + x_316 = None + bottleneck_output_77 = torch.conv2d( + x_317, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_317 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_318 = torch.nn.functional.batch_norm( + bottleneck_output_77, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_77 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_319 = torch.nn.functional.relu(x_318, inplace=True) + x_318 = None + new_features_77 = torch.conv2d( + x_319, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_319 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + concated_features_78 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + ], + 1, + ) + x_320 = torch.nn.functional.batch_norm( + concated_features_78, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_78 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ = (None) + x_321 = torch.nn.functional.relu(x_320, inplace=True) + x_320 = None + bottleneck_output_78 = torch.conv2d( + x_321, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_321 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ = (None) + x_322 = torch.nn.functional.batch_norm( + bottleneck_output_78, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_78 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ = (None) + x_323 = torch.nn.functional.relu(x_322, inplace=True) + x_322 = None + new_features_78 = torch.conv2d( + x_323, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_323 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ = (None) + concated_features_79 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + ], + 1, + ) + x_324 = torch.nn.functional.batch_norm( + concated_features_79, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_79 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ = (None) + x_325 = torch.nn.functional.relu(x_324, inplace=True) + x_324 = None + bottleneck_output_79 = torch.conv2d( + x_325, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_325 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ = (None) + x_326 = torch.nn.functional.batch_norm( + bottleneck_output_79, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_79 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ = (None) + x_327 = torch.nn.functional.relu(x_326, inplace=True) + x_326 = None + new_features_79 = torch.conv2d( + x_327, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_327 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ = (None) + concated_features_80 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + ], + 1, + ) + x_328 = torch.nn.functional.batch_norm( + concated_features_80, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_80 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ = (None) + x_329 = torch.nn.functional.relu(x_328, inplace=True) + x_328 = None + bottleneck_output_80 = torch.conv2d( + x_329, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_329 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ = (None) + x_330 = torch.nn.functional.batch_norm( + bottleneck_output_80, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_80 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ = (None) + x_331 = torch.nn.functional.relu(x_330, inplace=True) + x_330 = None + new_features_80 = torch.conv2d( + x_331, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_331 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ = (None) + concated_features_81 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + ], + 1, + ) + x_332 = torch.nn.functional.batch_norm( + concated_features_81, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_81 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ = (None) + x_333 = torch.nn.functional.relu(x_332, inplace=True) + x_332 = None + bottleneck_output_81 = torch.conv2d( + x_333, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_333 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ = (None) + x_334 = torch.nn.functional.batch_norm( + bottleneck_output_81, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_81 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ = (None) + x_335 = torch.nn.functional.relu(x_334, inplace=True) + x_334 = None + new_features_81 = torch.conv2d( + x_335, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_335 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ = (None) + concated_features_82 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + ], + 1, + ) + x_336 = torch.nn.functional.batch_norm( + concated_features_82, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_82 = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_ = (None) + x_337 = torch.nn.functional.relu(x_336, inplace=True) + x_336 = None + bottleneck_output_82 = torch.conv2d( + x_337, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_337 = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_ = (None) + x_338 = torch.nn.functional.batch_norm( + bottleneck_output_82, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_82 = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_ = (None) + x_339 = torch.nn.functional.relu(x_338, inplace=True) + x_338 = None + new_features_82 = torch.conv2d( + x_339, + l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_339 = l_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_ = (None) + concated_features_83 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + ], + 1, + ) + x_340 = torch.nn.functional.batch_norm( + concated_features_83, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_83 = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_ = (None) + x_341 = torch.nn.functional.relu(x_340, inplace=True) + x_340 = None + bottleneck_output_83 = torch.conv2d( + x_341, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_341 = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_ = (None) + x_342 = torch.nn.functional.batch_norm( + bottleneck_output_83, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_83 = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_ = (None) + x_343 = torch.nn.functional.relu(x_342, inplace=True) + x_342 = None + new_features_83 = torch.conv2d( + x_343, + l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_343 = l_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_ = (None) + concated_features_84 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + ], + 1, + ) + x_344 = torch.nn.functional.batch_norm( + concated_features_84, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_84 = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_ = (None) + x_345 = torch.nn.functional.relu(x_344, inplace=True) + x_344 = None + bottleneck_output_84 = torch.conv2d( + x_345, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_345 = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_ = (None) + x_346 = torch.nn.functional.batch_norm( + bottleneck_output_84, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_84 = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_ = (None) + x_347 = torch.nn.functional.relu(x_346, inplace=True) + x_346 = None + new_features_84 = torch.conv2d( + x_347, + l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_347 = l_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_ = (None) + concated_features_85 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + ], + 1, + ) + x_348 = torch.nn.functional.batch_norm( + concated_features_85, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_85 = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_ = (None) + x_349 = torch.nn.functional.relu(x_348, inplace=True) + x_348 = None + bottleneck_output_85 = torch.conv2d( + x_349, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_349 = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_ = (None) + x_350 = torch.nn.functional.batch_norm( + bottleneck_output_85, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_85 = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_ = (None) + x_351 = torch.nn.functional.relu(x_350, inplace=True) + x_350 = None + new_features_85 = torch.conv2d( + x_351, + l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_351 = l_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_ = (None) + concated_features_86 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + ], + 1, + ) + x_352 = torch.nn.functional.batch_norm( + concated_features_86, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_86 = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_ = (None) + x_353 = torch.nn.functional.relu(x_352, inplace=True) + x_352 = None + bottleneck_output_86 = torch.conv2d( + x_353, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_353 = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_ = (None) + x_354 = torch.nn.functional.batch_norm( + bottleneck_output_86, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_86 = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_ = (None) + x_355 = torch.nn.functional.relu(x_354, inplace=True) + x_354 = None + new_features_86 = torch.conv2d( + x_355, + l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_355 = l_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_ = (None) + concated_features_87 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + new_features_86, + ], + 1, + ) + x_356 = torch.nn.functional.batch_norm( + concated_features_87, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_87 = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_ = (None) + x_357 = torch.nn.functional.relu(x_356, inplace=True) + x_356 = None + bottleneck_output_87 = torch.conv2d( + x_357, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_357 = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_ = (None) + x_358 = torch.nn.functional.batch_norm( + bottleneck_output_87, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_87 = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_ = (None) + x_359 = torch.nn.functional.relu(x_358, inplace=True) + x_358 = None + new_features_87 = torch.conv2d( + x_359, + l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_359 = l_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_ = (None) + concated_features_88 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + new_features_86, + new_features_87, + ], + 1, + ) + x_360 = torch.nn.functional.batch_norm( + concated_features_88, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_88 = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_ = (None) + x_361 = torch.nn.functional.relu(x_360, inplace=True) + x_360 = None + bottleneck_output_88 = torch.conv2d( + x_361, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_361 = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_ = (None) + x_362 = torch.nn.functional.batch_norm( + bottleneck_output_88, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_88 = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_ = (None) + x_363 = torch.nn.functional.relu(x_362, inplace=True) + x_362 = None + new_features_88 = torch.conv2d( + x_363, + l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_363 = l_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_ = (None) + concated_features_89 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + new_features_86, + new_features_87, + new_features_88, + ], + 1, + ) + x_364 = torch.nn.functional.batch_norm( + concated_features_89, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_89 = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_ = (None) + x_365 = torch.nn.functional.relu(x_364, inplace=True) + x_364 = None + bottleneck_output_89 = torch.conv2d( + x_365, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_365 = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_ = (None) + x_366 = torch.nn.functional.batch_norm( + bottleneck_output_89, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_89 = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_ = (None) + x_367 = torch.nn.functional.relu(x_366, inplace=True) + x_366 = None + new_features_89 = torch.conv2d( + x_367, + l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_367 = l_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_ = (None) + concated_features_90 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + new_features_86, + new_features_87, + new_features_88, + new_features_89, + ], + 1, + ) + x_368 = torch.nn.functional.batch_norm( + concated_features_90, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_90 = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_ = (None) + x_369 = torch.nn.functional.relu(x_368, inplace=True) + x_368 = None + bottleneck_output_90 = torch.conv2d( + x_369, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_369 = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_ = (None) + x_370 = torch.nn.functional.batch_norm( + bottleneck_output_90, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_90 = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_ = (None) + x_371 = torch.nn.functional.relu(x_370, inplace=True) + x_370 = None + new_features_90 = torch.conv2d( + x_371, + l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_371 = l_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_ = (None) + concated_features_91 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + new_features_86, + new_features_87, + new_features_88, + new_features_89, + new_features_90, + ], + 1, + ) + x_372 = torch.nn.functional.batch_norm( + concated_features_91, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_91 = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_ = (None) + x_373 = torch.nn.functional.relu(x_372, inplace=True) + x_372 = None + bottleneck_output_91 = torch.conv2d( + x_373, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_373 = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_ = (None) + x_374 = torch.nn.functional.batch_norm( + bottleneck_output_91, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_91 = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_ = (None) + x_375 = torch.nn.functional.relu(x_374, inplace=True) + x_374 = None + new_features_91 = torch.conv2d( + x_375, + l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_375 = l_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_ = (None) + concated_features_92 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + new_features_86, + new_features_87, + new_features_88, + new_features_89, + new_features_90, + new_features_91, + ], + 1, + ) + x_376 = torch.nn.functional.batch_norm( + concated_features_92, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_92 = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_ = (None) + x_377 = torch.nn.functional.relu(x_376, inplace=True) + x_376 = None + bottleneck_output_92 = torch.conv2d( + x_377, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_377 = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_ = (None) + x_378 = torch.nn.functional.batch_norm( + bottleneck_output_92, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_92 = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_ = (None) + x_379 = torch.nn.functional.relu(x_378, inplace=True) + x_378 = None + new_features_92 = torch.conv2d( + x_379, + l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_379 = l_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_ = (None) + concated_features_93 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + new_features_86, + new_features_87, + new_features_88, + new_features_89, + new_features_90, + new_features_91, + new_features_92, + ], + 1, + ) + x_380 = torch.nn.functional.batch_norm( + concated_features_93, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_93 = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_ = (None) + x_381 = torch.nn.functional.relu(x_380, inplace=True) + x_380 = None + bottleneck_output_93 = torch.conv2d( + x_381, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_381 = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_ = (None) + x_382 = torch.nn.functional.batch_norm( + bottleneck_output_93, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_93 = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_ = (None) + x_383 = torch.nn.functional.relu(x_382, inplace=True) + x_382 = None + new_features_93 = torch.conv2d( + x_383, + l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_383 = l_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_ = (None) + concated_features_94 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + new_features_86, + new_features_87, + new_features_88, + new_features_89, + new_features_90, + new_features_91, + new_features_92, + new_features_93, + ], + 1, + ) + x_384 = torch.nn.functional.batch_norm( + concated_features_94, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_94 = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_ = (None) + x_385 = torch.nn.functional.relu(x_384, inplace=True) + x_384 = None + bottleneck_output_94 = torch.conv2d( + x_385, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_385 = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_ = (None) + x_386 = torch.nn.functional.batch_norm( + bottleneck_output_94, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_94 = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_ = (None) + x_387 = torch.nn.functional.relu(x_386, inplace=True) + x_386 = None + new_features_94 = torch.conv2d( + x_387, + l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_387 = l_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_ = (None) + concated_features_95 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + new_features_86, + new_features_87, + new_features_88, + new_features_89, + new_features_90, + new_features_91, + new_features_92, + new_features_93, + new_features_94, + ], + 1, + ) + x_388 = torch.nn.functional.batch_norm( + concated_features_95, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_95 = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_ = (None) + x_389 = torch.nn.functional.relu(x_388, inplace=True) + x_388 = None + bottleneck_output_95 = torch.conv2d( + x_389, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_389 = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_ = (None) + x_390 = torch.nn.functional.batch_norm( + bottleneck_output_95, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_95 = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_ = (None) + x_391 = torch.nn.functional.relu(x_390, inplace=True) + x_390 = None + new_features_95 = torch.conv2d( + x_391, + l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_391 = l_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_ = (None) + concated_features_96 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + new_features_86, + new_features_87, + new_features_88, + new_features_89, + new_features_90, + new_features_91, + new_features_92, + new_features_93, + new_features_94, + new_features_95, + ], + 1, + ) + x_392 = torch.nn.functional.batch_norm( + concated_features_96, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_96 = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_ = (None) + x_393 = torch.nn.functional.relu(x_392, inplace=True) + x_392 = None + bottleneck_output_96 = torch.conv2d( + x_393, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_393 = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_ = (None) + x_394 = torch.nn.functional.batch_norm( + bottleneck_output_96, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_96 = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_ = (None) + x_395 = torch.nn.functional.relu(x_394, inplace=True) + x_394 = None + new_features_96 = torch.conv2d( + x_395, + l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_395 = l_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_ = (None) + concated_features_97 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + new_features_86, + new_features_87, + new_features_88, + new_features_89, + new_features_90, + new_features_91, + new_features_92, + new_features_93, + new_features_94, + new_features_95, + new_features_96, + ], + 1, + ) + x_396 = torch.nn.functional.batch_norm( + concated_features_97, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_97 = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_ = (None) + x_397 = torch.nn.functional.relu(x_396, inplace=True) + x_396 = None + bottleneck_output_97 = torch.conv2d( + x_397, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_397 = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_ = (None) + x_398 = torch.nn.functional.batch_norm( + bottleneck_output_97, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_97 = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_ = (None) + x_399 = torch.nn.functional.relu(x_398, inplace=True) + x_398 = None + new_features_97 = torch.conv2d( + x_399, + l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_399 = l_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_ = (None) + input_12 = torch.cat( + [ + input_11, + new_features_66, + new_features_67, + new_features_68, + new_features_69, + new_features_70, + new_features_71, + new_features_72, + new_features_73, + new_features_74, + new_features_75, + new_features_76, + new_features_77, + new_features_78, + new_features_79, + new_features_80, + new_features_81, + new_features_82, + new_features_83, + new_features_84, + new_features_85, + new_features_86, + new_features_87, + new_features_88, + new_features_89, + new_features_90, + new_features_91, + new_features_92, + new_features_93, + new_features_94, + new_features_95, + new_features_96, + new_features_97, + ], + 1, + ) + input_11 = ( + new_features_66 + ) = ( + new_features_67 + ) = ( + new_features_68 + ) = ( + new_features_69 + ) = ( + new_features_70 + ) = ( + new_features_71 + ) = ( + new_features_72 + ) = ( + new_features_73 + ) = ( + new_features_74 + ) = ( + new_features_75 + ) = ( + new_features_76 + ) = ( + new_features_77 + ) = ( + new_features_78 + ) = ( + new_features_79 + ) = ( + new_features_80 + ) = ( + new_features_81 + ) = ( + new_features_82 + ) = ( + new_features_83 + ) = ( + new_features_84 + ) = ( + new_features_85 + ) = ( + new_features_86 + ) = ( + new_features_87 + ) = ( + new_features_88 + ) = ( + new_features_89 + ) = ( + new_features_90 + ) = ( + new_features_91 + ) = ( + new_features_92 + ) = ( + new_features_93 + ) = new_features_94 = new_features_95 = new_features_96 = new_features_97 = None + x_400 = torch.nn.functional.batch_norm( + input_12, + l_self_modules_features_modules_norm5_buffers_running_mean_, + l_self_modules_features_modules_norm5_buffers_running_var_, + l_self_modules_features_modules_norm5_parameters_weight_, + l_self_modules_features_modules_norm5_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_12 = ( + l_self_modules_features_modules_norm5_buffers_running_mean_ + ) = ( + l_self_modules_features_modules_norm5_buffers_running_var_ + ) = ( + l_self_modules_features_modules_norm5_parameters_weight_ + ) = l_self_modules_features_modules_norm5_parameters_bias_ = None + x_401 = torch.nn.functional.relu(x_400, inplace=True) + x_400 = None + x_402 = torch.nn.functional.adaptive_avg_pool2d(x_401, 1) + x_401 = None + x_403 = x_402.flatten(1, -1) + x_402 = None + x_404 = torch.nn.functional.dropout(x_403, 0.0, False, False) + x_403 = None + x_405 = torch._C._nn.linear( + x_404, + l_self_modules_classifier_parameters_weight_, + l_self_modules_classifier_parameters_bias_, + ) + x_404 = ( + l_self_modules_classifier_parameters_weight_ + ) = l_self_modules_classifier_parameters_bias_ = None + return (x_405,) diff --git a/samples/timm/densenet201.tv_in1k/weight_meta.py b/samples/timm/densenet201.tv_in1k/weight_meta.py new file mode 100644 index 0000000000..d33ca6c097 --- /dev/null +++ b/samples/timm/densenet201.tv_in1k/weight_meta.py @@ -0,0 +1,10090 @@ +class Program_weight_tensor_meta_L_self_modules_features_modules_conv0_parameters_weight_: + name = "L_self_modules_features_modules_conv0_parameters_weight_" + shape = [64, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.002 + std = 0.117 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.227 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_buffers_running_mean_: + name = "L_self_modules_features_modules_norm0_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_buffers_running_var_: + name = "L_self_modules_features_modules_norm0_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_parameters_weight_: + name = "L_self_modules_features_modules_norm0_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_parameters_bias_: + name = "L_self_modules_features_modules_norm0_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.176 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.145 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.103 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 224, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.095 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition1_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.111 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 224, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.095 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 288, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [128, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [128, 352, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.075 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [128, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [128, 416, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.069 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [128, 448, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.067 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [128, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition2_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition2_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition2_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 288, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 352, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.075 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 416, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.069 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [128, 448, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.067 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [128, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [128, 544, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.061 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [128, 576, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [128, 608, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.057 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_" + shape = [128, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_" + shape = [128, 672, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.055 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_" + shape = [128, 704, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_" + shape = [128, 736, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.052 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_" + shape = [128, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_" + shape = [128, 800, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.050 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_" + shape = [128, 832, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.049 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_" + shape = [128, 864, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_" + shape = [128, 896, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_" + shape = [128, 928, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_" + shape = [128, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_" + shape = [128, 992, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.045 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv1_parameters_weight_" + shape = [128, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer25_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_mean_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_buffers_running_var_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_weight_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm1_parameters_bias_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv1_parameters_weight_" + shape = [128, 1056, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer26_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_mean_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_buffers_running_var_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_weight_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm1_parameters_bias_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv1_parameters_weight_" + shape = [128, 1088, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer27_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_mean_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_buffers_running_var_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_weight_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm1_parameters_bias_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv1_parameters_weight_" + shape = [128, 1120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer28_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_mean_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_buffers_running_var_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_weight_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv1_parameters_weight_" + shape = [128, 1152, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer29_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_mean_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_buffers_running_var_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_weight_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm1_parameters_bias_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv1_parameters_weight_" + shape = [128, 1184, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer30_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_mean_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_buffers_running_var_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_weight_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm1_parameters_bias_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv1_parameters_weight_" + shape = [128, 1216, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer31_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_mean_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_buffers_running_var_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_weight_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm1_parameters_bias_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv1_parameters_weight_" + shape = [128, 1248, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer32_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_mean_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_buffers_running_var_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv1_parameters_weight_" + shape = [128, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.039 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer33_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_mean_" + shape = [1312] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_buffers_running_var_" + shape = [1312] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_weight_" + shape = [1312] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm1_parameters_bias_" + shape = [1312] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv1_parameters_weight_" + shape = [128, 1312, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.039 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer34_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_mean_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_buffers_running_var_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_weight_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv1_parameters_weight_" + shape = [128, 1344, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.039 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer35_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_mean_" + shape = [1376] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_buffers_running_var_" + shape = [1376] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_weight_" + shape = [1376] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm1_parameters_bias_" + shape = [1376] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv1_parameters_weight_" + shape = [128, 1376, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.038 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer36_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_mean_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_buffers_running_var_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm1_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv1_parameters_weight_" + shape = [128, 1408, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.038 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer37_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_mean_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_buffers_running_var_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_weight_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm1_parameters_bias_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv1_parameters_weight_" + shape = [128, 1440, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer38_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_mean_" + shape = [1472] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_buffers_running_var_" + shape = [1472] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_weight_" + shape = [1472] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm1_parameters_bias_" + shape = [1472] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv1_parameters_weight_" + shape = [128, 1472, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer39_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_mean_" + shape = [1504] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_buffers_running_var_" + shape = [1504] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_weight_" + shape = [1504] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm1_parameters_bias_" + shape = [1504] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv1_parameters_weight_" + shape = [128, 1504, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer40_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv1_parameters_weight_" + shape = [128, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer41_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_mean_" + shape = [1568] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_buffers_running_var_" + shape = [1568] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_weight_" + shape = [1568] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm1_parameters_bias_" + shape = [1568] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv1_parameters_weight_" + shape = [128, 1568, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer42_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_mean_" + shape = [1600] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_buffers_running_var_" + shape = [1600] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_weight_" + shape = [1600] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm1_parameters_bias_" + shape = [1600] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv1_parameters_weight_" + shape = [128, 1600, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.035 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer43_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_mean_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_buffers_running_var_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_weight_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm1_parameters_bias_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv1_parameters_weight_" + shape = [128, 1632, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.035 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer44_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_mean_" + shape = [1664] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_buffers_running_var_" + shape = [1664] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_weight_" + shape = [1664] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm1_parameters_bias_" + shape = [1664] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv1_parameters_weight_" + shape = [128, 1664, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.035 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer45_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_mean_" + shape = [1696] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_buffers_running_var_" + shape = [1696] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_weight_" + shape = [1696] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm1_parameters_bias_" + shape = [1696] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv1_parameters_weight_" + shape = [128, 1696, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer46_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_mean_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_buffers_running_var_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_weight_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm1_parameters_bias_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv1_parameters_weight_" + shape = [128, 1728, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer47_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_mean_" + shape = [1760] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_buffers_running_var_" + shape = [1760] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_weight_" + shape = [1760] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm1_parameters_bias_" + shape = [1760] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv1_parameters_weight_" + shape = [128, 1760, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer48_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_" + ) + shape = [1792] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_" + ) + shape = [1792] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition3_modules_norm_parameters_weight_" + shape = [1792] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition3_modules_norm_parameters_bias_" + shape = [1792] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition3_modules_conv_parameters_weight_" + shape = [896, 1792, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.033 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 896, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 928, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 992, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.045 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [1056] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 1056, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [1088] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [128, 1088, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.043 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [1120] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [128, 1120, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [1152] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [128, 1152, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [1184] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [128, 1184, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [1216] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [128, 1216, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [1248] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [128, 1248, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_" + shape = [1280] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_" + shape = [128, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.040 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_" + shape = [1312] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_" + shape = [1312] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_" + shape = [1312] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_" + shape = [1312] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_" + shape = [128, 1312, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.039 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_" + shape = [1344] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_" + shape = [128, 1344, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.038 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_" + shape = [1376] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_" + shape = [1376] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_" + shape = [1376] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_" + shape = [1376] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_" + shape = [128, 1376, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.038 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_mean_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_buffers_running_var_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_weight_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm1_parameters_bias_" + shape = [1408] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv1_parameters_weight_" + shape = [128, 1408, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.038 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer17_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_mean_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_buffers_running_var_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_weight_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm1_parameters_bias_" + shape = [1440] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv1_parameters_weight_" + shape = [128, 1440, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer18_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_mean_" + shape = [1472] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_buffers_running_var_" + shape = [1472] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_weight_" + shape = [1472] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm1_parameters_bias_" + shape = [1472] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv1_parameters_weight_" + shape = [128, 1472, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.037 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer19_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_mean_" + shape = [1504] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_buffers_running_var_" + shape = [1504] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_weight_" + shape = [1504] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm1_parameters_bias_" + shape = [1504] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv1_parameters_weight_" + shape = [128, 1504, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer20_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_mean_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_buffers_running_var_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_weight_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm1_parameters_bias_" + shape = [1536] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv1_parameters_weight_" + shape = [128, 1536, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer21_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_mean_" + shape = [1568] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_buffers_running_var_" + shape = [1568] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_weight_" + shape = [1568] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm1_parameters_bias_" + shape = [1568] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv1_parameters_weight_" + shape = [128, 1568, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.036 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer22_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_mean_" + shape = [1600] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_buffers_running_var_" + shape = [1600] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_weight_" + shape = [1600] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm1_parameters_bias_" + shape = [1600] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv1_parameters_weight_" + shape = [128, 1600, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.035 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer23_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_mean_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_buffers_running_var_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_weight_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm1_parameters_bias_" + shape = [1632] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv1_parameters_weight_" + shape = [128, 1632, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.035 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer24_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_mean_" + shape = [1664] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_buffers_running_var_" + shape = [1664] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_weight_" + shape = [1664] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm1_parameters_bias_" + shape = [1664] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv1_parameters_weight_" + shape = [128, 1664, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.035 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer25_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_mean_" + shape = [1696] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_buffers_running_var_" + shape = [1696] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_weight_" + shape = [1696] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm1_parameters_bias_" + shape = [1696] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv1_parameters_weight_" + shape = [128, 1696, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer26_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_mean_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_buffers_running_var_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_weight_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm1_parameters_bias_" + shape = [1728] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv1_parameters_weight_" + shape = [128, 1728, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer27_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_mean_" + shape = [1760] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_buffers_running_var_" + shape = [1760] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_weight_" + shape = [1760] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm1_parameters_bias_" + shape = [1760] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv1_parameters_weight_" + shape = [128, 1760, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.034 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer28_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_mean_" + shape = [1792] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_buffers_running_var_" + shape = [1792] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_weight_" + shape = [1792] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm1_parameters_bias_" + shape = [1792] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv1_parameters_weight_" + shape = [128, 1792, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.033 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer29_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_mean_" + shape = [1824] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_buffers_running_var_" + shape = [1824] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_weight_" + shape = [1824] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm1_parameters_bias_" + shape = [1824] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv1_parameters_weight_" + shape = [128, 1824, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.033 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer30_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_mean_" + shape = [1856] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_buffers_running_var_" + shape = [1856] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_weight_" + shape = [1856] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm1_parameters_bias_" + shape = [1856] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv1_parameters_weight_" + shape = [128, 1856, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.033 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer31_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_mean_" + shape = [1888] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_buffers_running_var_" + shape = [1888] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_weight_" + shape = [1888] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm1_parameters_bias_" + shape = [1888] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv1_parameters_weight_" + shape = [128, 1888, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.033 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer32_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_buffers_running_mean_: + name = "L_self_modules_features_modules_norm5_buffers_running_mean_" + shape = [1920] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_buffers_running_var_: + name = "L_self_modules_features_modules_norm5_buffers_running_var_" + shape = [1920] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_parameters_weight_: + name = "L_self_modules_features_modules_norm5_parameters_weight_" + shape = [1920] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_parameters_bias_: + name = "L_self_modules_features_modules_norm5_parameters_bias_" + shape = [1920] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_classifier_parameters_weight_: + name = "L_self_modules_classifier_parameters_weight_" + shape = [1000, 1920] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.013 + data = None + + +class Program_weight_tensor_meta_L_self_modules_classifier_parameters_bias_: + name = "L_self_modules_classifier_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/densenetblur121d.ra_in1k/graph_hash.txt b/samples/timm/densenetblur121d.ra_in1k/graph_hash.txt new file mode 100644 index 0000000000..f267737c70 --- /dev/null +++ b/samples/timm/densenetblur121d.ra_in1k/graph_hash.txt @@ -0,0 +1 @@ +c19519861ce0a98446d72201993ea0c8e7662560a288da6ea9997a3e4ddae2c2 \ No newline at end of file diff --git a/samples/timm/densenetblur121d.ra_in1k/graph_net.json b/samples/timm/densenetblur121d.ra_in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/densenetblur121d.ra_in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/densenetblur121d.ra_in1k/input_meta.py b/samples/timm/densenetblur121d.ra_in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/densenetblur121d.ra_in1k/input_tensor_constraints.py b/samples/timm/densenetblur121d.ra_in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/densenetblur121d.ra_in1k/model.py b/samples/timm/densenetblur121d.ra_in1k/model.py new file mode 100644 index 0000000000..087a06325c --- /dev/null +++ b/samples/timm/densenetblur121d.ra_in1k/model.py @@ -0,0 +1,5090 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_features_modules_conv0_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_features_modules_norm0_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_norm0_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_norm0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm0_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_pool0_modules_1_buffers_filt_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition1_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition2_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_transition3_modules_norm_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_norm_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_transition3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm5_buffers_running_mean_: torch.Tensor, + L_self_modules_features_modules_norm5_buffers_running_var_: torch.Tensor, + L_self_modules_features_modules_norm5_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_features_modules_norm5_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_classifier_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_classifier_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_features_modules_conv0_parameters_weight_ = ( + L_self_modules_features_modules_conv0_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_features_modules_norm0_buffers_running_mean_ = ( + L_self_modules_features_modules_norm0_buffers_running_mean_ + ) + l_self_modules_features_modules_norm0_buffers_running_var_ = ( + L_self_modules_features_modules_norm0_buffers_running_var_ + ) + l_self_modules_features_modules_norm0_parameters_weight_ = ( + L_self_modules_features_modules_norm0_parameters_weight_ + ) + l_self_modules_features_modules_norm0_parameters_bias_ = ( + L_self_modules_features_modules_norm0_parameters_bias_ + ) + l_self_modules_features_modules_conv1_parameters_weight_ = ( + L_self_modules_features_modules_conv1_parameters_weight_ + ) + l_self_modules_features_modules_norm1_buffers_running_mean_ = ( + L_self_modules_features_modules_norm1_buffers_running_mean_ + ) + l_self_modules_features_modules_norm1_buffers_running_var_ = ( + L_self_modules_features_modules_norm1_buffers_running_var_ + ) + l_self_modules_features_modules_norm1_parameters_weight_ = ( + L_self_modules_features_modules_norm1_parameters_weight_ + ) + l_self_modules_features_modules_norm1_parameters_bias_ = ( + L_self_modules_features_modules_norm1_parameters_bias_ + ) + l_self_modules_features_modules_conv2_parameters_weight_ = ( + L_self_modules_features_modules_conv2_parameters_weight_ + ) + l_self_modules_features_modules_norm2_buffers_running_mean_ = ( + L_self_modules_features_modules_norm2_buffers_running_mean_ + ) + l_self_modules_features_modules_norm2_buffers_running_var_ = ( + L_self_modules_features_modules_norm2_buffers_running_var_ + ) + l_self_modules_features_modules_norm2_parameters_weight_ = ( + L_self_modules_features_modules_norm2_parameters_weight_ + ) + l_self_modules_features_modules_norm2_parameters_bias_ = ( + L_self_modules_features_modules_norm2_parameters_bias_ + ) + l_self_modules_features_modules_pool0_modules_1_buffers_filt_ = ( + L_self_modules_features_modules_pool0_modules_1_buffers_filt_ + ) + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition1_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition1_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition1_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition2_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition2_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition2_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ + l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ = L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ + l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ = L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_ = ( + L_self_modules_features_modules_transition3_modules_norm_parameters_weight_ + ) + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_ = ( + L_self_modules_features_modules_transition3_modules_norm_parameters_bias_ + ) + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_ = ( + L_self_modules_features_modules_transition3_modules_conv_parameters_weight_ + ) + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ = L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ + l_self_modules_features_modules_norm5_buffers_running_mean_ = ( + L_self_modules_features_modules_norm5_buffers_running_mean_ + ) + l_self_modules_features_modules_norm5_buffers_running_var_ = ( + L_self_modules_features_modules_norm5_buffers_running_var_ + ) + l_self_modules_features_modules_norm5_parameters_weight_ = ( + L_self_modules_features_modules_norm5_parameters_weight_ + ) + l_self_modules_features_modules_norm5_parameters_bias_ = ( + L_self_modules_features_modules_norm5_parameters_bias_ + ) + l_self_modules_classifier_parameters_weight_ = ( + L_self_modules_classifier_parameters_weight_ + ) + l_self_modules_classifier_parameters_bias_ = ( + L_self_modules_classifier_parameters_bias_ + ) + input_1 = torch.conv2d( + l_x_, + l_self_modules_features_modules_conv0_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + l_x_ = l_self_modules_features_modules_conv0_parameters_weight_ = None + x = torch.nn.functional.batch_norm( + input_1, + l_self_modules_features_modules_norm0_buffers_running_mean_, + l_self_modules_features_modules_norm0_buffers_running_var_, + l_self_modules_features_modules_norm0_parameters_weight_, + l_self_modules_features_modules_norm0_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_1 = ( + l_self_modules_features_modules_norm0_buffers_running_mean_ + ) = ( + l_self_modules_features_modules_norm0_buffers_running_var_ + ) = ( + l_self_modules_features_modules_norm0_parameters_weight_ + ) = l_self_modules_features_modules_norm0_parameters_bias_ = None + x_1 = torch.nn.functional.relu(x, inplace=True) + x = None + input_2 = torch.conv2d( + x_1, + l_self_modules_features_modules_conv1_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_1 = l_self_modules_features_modules_conv1_parameters_weight_ = None + x_2 = torch.nn.functional.batch_norm( + input_2, + l_self_modules_features_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_norm1_parameters_weight_, + l_self_modules_features_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_2 = ( + l_self_modules_features_modules_norm1_buffers_running_mean_ + ) = ( + l_self_modules_features_modules_norm1_buffers_running_var_ + ) = ( + l_self_modules_features_modules_norm1_parameters_weight_ + ) = l_self_modules_features_modules_norm1_parameters_bias_ = None + x_3 = torch.nn.functional.relu(x_2, inplace=True) + x_2 = None + input_3 = torch.conv2d( + x_3, + l_self_modules_features_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_3 = l_self_modules_features_modules_conv2_parameters_weight_ = None + x_4 = torch.nn.functional.batch_norm( + input_3, + l_self_modules_features_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_norm2_parameters_weight_, + l_self_modules_features_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_3 = ( + l_self_modules_features_modules_norm2_buffers_running_mean_ + ) = ( + l_self_modules_features_modules_norm2_buffers_running_var_ + ) = ( + l_self_modules_features_modules_norm2_parameters_weight_ + ) = l_self_modules_features_modules_norm2_parameters_bias_ = None + x_5 = torch.nn.functional.relu(x_4, inplace=True) + x_4 = None + input_4 = torch.nn.functional.max_pool2d( + x_5, 3, 1, 1, 1, ceil_mode=False, return_indices=False + ) + x_5 = None + x_6 = torch._C._nn.pad(input_4, [1, 1, 1, 1], "reflect", None) + input_4 = None + input_5 = torch.conv2d( + x_6, + l_self_modules_features_modules_pool0_modules_1_buffers_filt_, + stride=2, + groups=64, + ) + x_6 = l_self_modules_features_modules_pool0_modules_1_buffers_filt_ = None + concated_features = torch.cat([input_5], 1) + x_7 = torch.nn.functional.batch_norm( + concated_features, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_8 = torch.nn.functional.relu(x_7, inplace=True) + x_7 = None + bottleneck_output = torch.conv2d( + x_8, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_8 = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_9 = torch.nn.functional.batch_norm( + bottleneck_output, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_10 = torch.nn.functional.relu(x_9, inplace=True) + x_9 = None + new_features = torch.conv2d( + x_10, + l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_10 = l_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_1 = torch.cat([input_5, new_features], 1) + x_11 = torch.nn.functional.batch_norm( + concated_features_1, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_1 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_12 = torch.nn.functional.relu(x_11, inplace=True) + x_11 = None + bottleneck_output_1 = torch.conv2d( + x_12, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_12 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_13 = torch.nn.functional.batch_norm( + bottleneck_output_1, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_1 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_14 = torch.nn.functional.relu(x_13, inplace=True) + x_13 = None + new_features_1 = torch.conv2d( + x_14, + l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_14 = l_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_2 = torch.cat([input_5, new_features, new_features_1], 1) + x_15 = torch.nn.functional.batch_norm( + concated_features_2, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_2 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_16 = torch.nn.functional.relu(x_15, inplace=True) + x_15 = None + bottleneck_output_2 = torch.conv2d( + x_16, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_16 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_17 = torch.nn.functional.batch_norm( + bottleneck_output_2, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_2 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_18 = torch.nn.functional.relu(x_17, inplace=True) + x_17 = None + new_features_2 = torch.conv2d( + x_18, + l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_18 = l_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_3 = torch.cat( + [input_5, new_features, new_features_1, new_features_2], 1 + ) + x_19 = torch.nn.functional.batch_norm( + concated_features_3, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_3 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_20 = torch.nn.functional.relu(x_19, inplace=True) + x_19 = None + bottleneck_output_3 = torch.conv2d( + x_20, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_20 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_21 = torch.nn.functional.batch_norm( + bottleneck_output_3, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_3 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_22 = torch.nn.functional.relu(x_21, inplace=True) + x_21 = None + new_features_3 = torch.conv2d( + x_22, + l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_22 = l_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_4 = torch.cat( + [input_5, new_features, new_features_1, new_features_2, new_features_3], 1 + ) + x_23 = torch.nn.functional.batch_norm( + concated_features_4, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_4 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_24 = torch.nn.functional.relu(x_23, inplace=True) + x_23 = None + bottleneck_output_4 = torch.conv2d( + x_24, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_24 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_25 = torch.nn.functional.batch_norm( + bottleneck_output_4, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_4 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_26 = torch.nn.functional.relu(x_25, inplace=True) + x_25 = None + new_features_4 = torch.conv2d( + x_26, + l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_26 = l_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_5 = torch.cat( + [ + input_5, + new_features, + new_features_1, + new_features_2, + new_features_3, + new_features_4, + ], + 1, + ) + x_27 = torch.nn.functional.batch_norm( + concated_features_5, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_5 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_28 = torch.nn.functional.relu(x_27, inplace=True) + x_27 = None + bottleneck_output_5 = torch.conv2d( + x_28, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_28 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_29 = torch.nn.functional.batch_norm( + bottleneck_output_5, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_5 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_30 = torch.nn.functional.relu(x_29, inplace=True) + x_29 = None + new_features_5 = torch.conv2d( + x_30, + l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_30 = l_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + input_6 = torch.cat( + [ + input_5, + new_features, + new_features_1, + new_features_2, + new_features_3, + new_features_4, + new_features_5, + ], + 1, + ) + input_5 = ( + new_features + ) = ( + new_features_1 + ) = new_features_2 = new_features_3 = new_features_4 = new_features_5 = None + x_31 = torch.nn.functional.batch_norm( + input_6, + l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_6 = l_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition1_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition1_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition1_modules_norm_parameters_bias_ + ) = None + x_32 = torch.nn.functional.relu(x_31, inplace=True) + x_31 = None + input_7 = torch.conv2d( + x_32, + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_32 = ( + l_self_modules_features_modules_transition1_modules_conv_parameters_weight_ + ) = None + input_8 = torch._C._nn.avg_pool2d(input_7, 2, 2, 0, False, True, None) + input_7 = None + concated_features_6 = torch.cat([input_8], 1) + x_33 = torch.nn.functional.batch_norm( + concated_features_6, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_6 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_34 = torch.nn.functional.relu(x_33, inplace=True) + x_33 = None + bottleneck_output_6 = torch.conv2d( + x_34, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_34 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_35 = torch.nn.functional.batch_norm( + bottleneck_output_6, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_6 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_36 = torch.nn.functional.relu(x_35, inplace=True) + x_35 = None + new_features_6 = torch.conv2d( + x_36, + l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_36 = l_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_7 = torch.cat([input_8, new_features_6], 1) + x_37 = torch.nn.functional.batch_norm( + concated_features_7, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_7 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_38 = torch.nn.functional.relu(x_37, inplace=True) + x_37 = None + bottleneck_output_7 = torch.conv2d( + x_38, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_38 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_39 = torch.nn.functional.batch_norm( + bottleneck_output_7, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_7 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_40 = torch.nn.functional.relu(x_39, inplace=True) + x_39 = None + new_features_7 = torch.conv2d( + x_40, + l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_40 = l_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_8 = torch.cat([input_8, new_features_6, new_features_7], 1) + x_41 = torch.nn.functional.batch_norm( + concated_features_8, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_8 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_42 = torch.nn.functional.relu(x_41, inplace=True) + x_41 = None + bottleneck_output_8 = torch.conv2d( + x_42, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_42 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_43 = torch.nn.functional.batch_norm( + bottleneck_output_8, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_8 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_44 = torch.nn.functional.relu(x_43, inplace=True) + x_43 = None + new_features_8 = torch.conv2d( + x_44, + l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_44 = l_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_9 = torch.cat( + [input_8, new_features_6, new_features_7, new_features_8], 1 + ) + x_45 = torch.nn.functional.batch_norm( + concated_features_9, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_9 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_46 = torch.nn.functional.relu(x_45, inplace=True) + x_45 = None + bottleneck_output_9 = torch.conv2d( + x_46, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_46 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_47 = torch.nn.functional.batch_norm( + bottleneck_output_9, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_9 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_48 = torch.nn.functional.relu(x_47, inplace=True) + x_47 = None + new_features_9 = torch.conv2d( + x_48, + l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_48 = l_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_10 = torch.cat( + [input_8, new_features_6, new_features_7, new_features_8, new_features_9], 1 + ) + x_49 = torch.nn.functional.batch_norm( + concated_features_10, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_10 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_50 = torch.nn.functional.relu(x_49, inplace=True) + x_49 = None + bottleneck_output_10 = torch.conv2d( + x_50, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_50 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_51 = torch.nn.functional.batch_norm( + bottleneck_output_10, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_10 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_52 = torch.nn.functional.relu(x_51, inplace=True) + x_51 = None + new_features_10 = torch.conv2d( + x_52, + l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_52 = l_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_11 = torch.cat( + [ + input_8, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + ], + 1, + ) + x_53 = torch.nn.functional.batch_norm( + concated_features_11, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_11 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_54 = torch.nn.functional.relu(x_53, inplace=True) + x_53 = None + bottleneck_output_11 = torch.conv2d( + x_54, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_54 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_55 = torch.nn.functional.batch_norm( + bottleneck_output_11, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_11 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_56 = torch.nn.functional.relu(x_55, inplace=True) + x_55 = None + new_features_11 = torch.conv2d( + x_56, + l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_56 = l_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_12 = torch.cat( + [ + input_8, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + ], + 1, + ) + x_57 = torch.nn.functional.batch_norm( + concated_features_12, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_12 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_58 = torch.nn.functional.relu(x_57, inplace=True) + x_57 = None + bottleneck_output_12 = torch.conv2d( + x_58, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_58 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_59 = torch.nn.functional.batch_norm( + bottleneck_output_12, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_12 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_60 = torch.nn.functional.relu(x_59, inplace=True) + x_59 = None + new_features_12 = torch.conv2d( + x_60, + l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_60 = l_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_13 = torch.cat( + [ + input_8, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + ], + 1, + ) + x_61 = torch.nn.functional.batch_norm( + concated_features_13, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_13 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_62 = torch.nn.functional.relu(x_61, inplace=True) + x_61 = None + bottleneck_output_13 = torch.conv2d( + x_62, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_62 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_63 = torch.nn.functional.batch_norm( + bottleneck_output_13, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_13 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_64 = torch.nn.functional.relu(x_63, inplace=True) + x_63 = None + new_features_13 = torch.conv2d( + x_64, + l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_64 = l_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_14 = torch.cat( + [ + input_8, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + ], + 1, + ) + x_65 = torch.nn.functional.batch_norm( + concated_features_14, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_14 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_66 = torch.nn.functional.relu(x_65, inplace=True) + x_65 = None + bottleneck_output_14 = torch.conv2d( + x_66, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_66 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_67 = torch.nn.functional.batch_norm( + bottleneck_output_14, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_14 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_68 = torch.nn.functional.relu(x_67, inplace=True) + x_67 = None + new_features_14 = torch.conv2d( + x_68, + l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_68 = l_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_15 = torch.cat( + [ + input_8, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + ], + 1, + ) + x_69 = torch.nn.functional.batch_norm( + concated_features_15, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_15 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_70 = torch.nn.functional.relu(x_69, inplace=True) + x_69 = None + bottleneck_output_15 = torch.conv2d( + x_70, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_70 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_71 = torch.nn.functional.batch_norm( + bottleneck_output_15, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_15 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_72 = torch.nn.functional.relu(x_71, inplace=True) + x_71 = None + new_features_15 = torch.conv2d( + x_72, + l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_72 = l_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_16 = torch.cat( + [ + input_8, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + ], + 1, + ) + x_73 = torch.nn.functional.batch_norm( + concated_features_16, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_16 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_74 = torch.nn.functional.relu(x_73, inplace=True) + x_73 = None + bottleneck_output_16 = torch.conv2d( + x_74, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_74 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_75 = torch.nn.functional.batch_norm( + bottleneck_output_16, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_16 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_76 = torch.nn.functional.relu(x_75, inplace=True) + x_75 = None + new_features_16 = torch.conv2d( + x_76, + l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_76 = l_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_17 = torch.cat( + [ + input_8, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + new_features_16, + ], + 1, + ) + x_77 = torch.nn.functional.batch_norm( + concated_features_17, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_17 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_78 = torch.nn.functional.relu(x_77, inplace=True) + x_77 = None + bottleneck_output_17 = torch.conv2d( + x_78, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_78 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_79 = torch.nn.functional.batch_norm( + bottleneck_output_17, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_17 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_80 = torch.nn.functional.relu(x_79, inplace=True) + x_79 = None + new_features_17 = torch.conv2d( + x_80, + l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_80 = l_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + input_9 = torch.cat( + [ + input_8, + new_features_6, + new_features_7, + new_features_8, + new_features_9, + new_features_10, + new_features_11, + new_features_12, + new_features_13, + new_features_14, + new_features_15, + new_features_16, + new_features_17, + ], + 1, + ) + input_8 = ( + new_features_6 + ) = ( + new_features_7 + ) = ( + new_features_8 + ) = ( + new_features_9 + ) = ( + new_features_10 + ) = ( + new_features_11 + ) = ( + new_features_12 + ) = ( + new_features_13 + ) = new_features_14 = new_features_15 = new_features_16 = new_features_17 = None + x_81 = torch.nn.functional.batch_norm( + input_9, + l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_9 = l_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition2_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition2_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition2_modules_norm_parameters_bias_ + ) = None + x_82 = torch.nn.functional.relu(x_81, inplace=True) + x_81 = None + input_10 = torch.conv2d( + x_82, + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_82 = ( + l_self_modules_features_modules_transition2_modules_conv_parameters_weight_ + ) = None + input_11 = torch._C._nn.avg_pool2d(input_10, 2, 2, 0, False, True, None) + input_10 = None + concated_features_18 = torch.cat([input_11], 1) + x_83 = torch.nn.functional.batch_norm( + concated_features_18, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_18 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_84 = torch.nn.functional.relu(x_83, inplace=True) + x_83 = None + bottleneck_output_18 = torch.conv2d( + x_84, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_84 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_85 = torch.nn.functional.batch_norm( + bottleneck_output_18, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_18 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_86 = torch.nn.functional.relu(x_85, inplace=True) + x_85 = None + new_features_18 = torch.conv2d( + x_86, + l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_86 = l_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_19 = torch.cat([input_11, new_features_18], 1) + x_87 = torch.nn.functional.batch_norm( + concated_features_19, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_19 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_88 = torch.nn.functional.relu(x_87, inplace=True) + x_87 = None + bottleneck_output_19 = torch.conv2d( + x_88, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_88 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_89 = torch.nn.functional.batch_norm( + bottleneck_output_19, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_19 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_90 = torch.nn.functional.relu(x_89, inplace=True) + x_89 = None + new_features_19 = torch.conv2d( + x_90, + l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_90 = l_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_20 = torch.cat( + [input_11, new_features_18, new_features_19], 1 + ) + x_91 = torch.nn.functional.batch_norm( + concated_features_20, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_20 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_92 = torch.nn.functional.relu(x_91, inplace=True) + x_91 = None + bottleneck_output_20 = torch.conv2d( + x_92, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_92 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_93 = torch.nn.functional.batch_norm( + bottleneck_output_20, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_20 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_94 = torch.nn.functional.relu(x_93, inplace=True) + x_93 = None + new_features_20 = torch.conv2d( + x_94, + l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_94 = l_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_21 = torch.cat( + [input_11, new_features_18, new_features_19, new_features_20], 1 + ) + x_95 = torch.nn.functional.batch_norm( + concated_features_21, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_21 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_96 = torch.nn.functional.relu(x_95, inplace=True) + x_95 = None + bottleneck_output_21 = torch.conv2d( + x_96, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_96 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_97 = torch.nn.functional.batch_norm( + bottleneck_output_21, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_21 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_98 = torch.nn.functional.relu(x_97, inplace=True) + x_97 = None + new_features_21 = torch.conv2d( + x_98, + l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_98 = l_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_22 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + ], + 1, + ) + x_99 = torch.nn.functional.batch_norm( + concated_features_22, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_22 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_100 = torch.nn.functional.relu(x_99, inplace=True) + x_99 = None + bottleneck_output_22 = torch.conv2d( + x_100, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_100 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_101 = torch.nn.functional.batch_norm( + bottleneck_output_22, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_22 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_102 = torch.nn.functional.relu(x_101, inplace=True) + x_101 = None + new_features_22 = torch.conv2d( + x_102, + l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_102 = l_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_23 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + ], + 1, + ) + x_103 = torch.nn.functional.batch_norm( + concated_features_23, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_23 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_104 = torch.nn.functional.relu(x_103, inplace=True) + x_103 = None + bottleneck_output_23 = torch.conv2d( + x_104, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_104 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_105 = torch.nn.functional.batch_norm( + bottleneck_output_23, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_23 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_106 = torch.nn.functional.relu(x_105, inplace=True) + x_105 = None + new_features_23 = torch.conv2d( + x_106, + l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_106 = l_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_24 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + ], + 1, + ) + x_107 = torch.nn.functional.batch_norm( + concated_features_24, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_24 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_108 = torch.nn.functional.relu(x_107, inplace=True) + x_107 = None + bottleneck_output_24 = torch.conv2d( + x_108, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_108 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_109 = torch.nn.functional.batch_norm( + bottleneck_output_24, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_24 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_110 = torch.nn.functional.relu(x_109, inplace=True) + x_109 = None + new_features_24 = torch.conv2d( + x_110, + l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_110 = l_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_25 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + ], + 1, + ) + x_111 = torch.nn.functional.batch_norm( + concated_features_25, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_25 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_112 = torch.nn.functional.relu(x_111, inplace=True) + x_111 = None + bottleneck_output_25 = torch.conv2d( + x_112, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_112 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_113 = torch.nn.functional.batch_norm( + bottleneck_output_25, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_25 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_114 = torch.nn.functional.relu(x_113, inplace=True) + x_113 = None + new_features_25 = torch.conv2d( + x_114, + l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_114 = l_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_26 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + ], + 1, + ) + x_115 = torch.nn.functional.batch_norm( + concated_features_26, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_26 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_116 = torch.nn.functional.relu(x_115, inplace=True) + x_115 = None + bottleneck_output_26 = torch.conv2d( + x_116, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_116 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_117 = torch.nn.functional.batch_norm( + bottleneck_output_26, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_26 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_118 = torch.nn.functional.relu(x_117, inplace=True) + x_117 = None + new_features_26 = torch.conv2d( + x_118, + l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_118 = l_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_27 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + ], + 1, + ) + x_119 = torch.nn.functional.batch_norm( + concated_features_27, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_27 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_120 = torch.nn.functional.relu(x_119, inplace=True) + x_119 = None + bottleneck_output_27 = torch.conv2d( + x_120, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_120 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_121 = torch.nn.functional.batch_norm( + bottleneck_output_27, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_27 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_122 = torch.nn.functional.relu(x_121, inplace=True) + x_121 = None + new_features_27 = torch.conv2d( + x_122, + l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_122 = l_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_28 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + ], + 1, + ) + x_123 = torch.nn.functional.batch_norm( + concated_features_28, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_28 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_124 = torch.nn.functional.relu(x_123, inplace=True) + x_123 = None + bottleneck_output_28 = torch.conv2d( + x_124, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_124 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_125 = torch.nn.functional.batch_norm( + bottleneck_output_28, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_28 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_126 = torch.nn.functional.relu(x_125, inplace=True) + x_125 = None + new_features_28 = torch.conv2d( + x_126, + l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_126 = l_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_29 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + ], + 1, + ) + x_127 = torch.nn.functional.batch_norm( + concated_features_29, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_29 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_128 = torch.nn.functional.relu(x_127, inplace=True) + x_127 = None + bottleneck_output_29 = torch.conv2d( + x_128, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_128 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_129 = torch.nn.functional.batch_norm( + bottleneck_output_29, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_29 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_130 = torch.nn.functional.relu(x_129, inplace=True) + x_129 = None + new_features_29 = torch.conv2d( + x_130, + l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_130 = l_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + concated_features_30 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + ], + 1, + ) + x_131 = torch.nn.functional.batch_norm( + concated_features_30, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_30 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_ = (None) + x_132 = torch.nn.functional.relu(x_131, inplace=True) + x_131 = None + bottleneck_output_30 = torch.conv2d( + x_132, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_132 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_ = (None) + x_133 = torch.nn.functional.batch_norm( + bottleneck_output_30, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_30 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_ = (None) + x_134 = torch.nn.functional.relu(x_133, inplace=True) + x_133 = None + new_features_30 = torch.conv2d( + x_134, + l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_134 = l_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_ = (None) + concated_features_31 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + ], + 1, + ) + x_135 = torch.nn.functional.batch_norm( + concated_features_31, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_31 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_ = (None) + x_136 = torch.nn.functional.relu(x_135, inplace=True) + x_135 = None + bottleneck_output_31 = torch.conv2d( + x_136, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_136 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_ = (None) + x_137 = torch.nn.functional.batch_norm( + bottleneck_output_31, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_31 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_ = (None) + x_138 = torch.nn.functional.relu(x_137, inplace=True) + x_137 = None + new_features_31 = torch.conv2d( + x_138, + l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_138 = l_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_ = (None) + concated_features_32 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + ], + 1, + ) + x_139 = torch.nn.functional.batch_norm( + concated_features_32, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_32 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_ = (None) + x_140 = torch.nn.functional.relu(x_139, inplace=True) + x_139 = None + bottleneck_output_32 = torch.conv2d( + x_140, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_140 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_ = (None) + x_141 = torch.nn.functional.batch_norm( + bottleneck_output_32, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_32 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_ = (None) + x_142 = torch.nn.functional.relu(x_141, inplace=True) + x_141 = None + new_features_32 = torch.conv2d( + x_142, + l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_142 = l_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_ = (None) + concated_features_33 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + ], + 1, + ) + x_143 = torch.nn.functional.batch_norm( + concated_features_33, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_33 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_ = (None) + x_144 = torch.nn.functional.relu(x_143, inplace=True) + x_143 = None + bottleneck_output_33 = torch.conv2d( + x_144, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_144 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_ = (None) + x_145 = torch.nn.functional.batch_norm( + bottleneck_output_33, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_33 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_ = (None) + x_146 = torch.nn.functional.relu(x_145, inplace=True) + x_145 = None + new_features_33 = torch.conv2d( + x_146, + l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_146 = l_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_ = (None) + concated_features_34 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + ], + 1, + ) + x_147 = torch.nn.functional.batch_norm( + concated_features_34, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_34 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_ = (None) + x_148 = torch.nn.functional.relu(x_147, inplace=True) + x_147 = None + bottleneck_output_34 = torch.conv2d( + x_148, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_148 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_ = (None) + x_149 = torch.nn.functional.batch_norm( + bottleneck_output_34, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_34 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_ = (None) + x_150 = torch.nn.functional.relu(x_149, inplace=True) + x_149 = None + new_features_34 = torch.conv2d( + x_150, + l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_150 = l_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_ = (None) + concated_features_35 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + ], + 1, + ) + x_151 = torch.nn.functional.batch_norm( + concated_features_35, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_35 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_ = (None) + x_152 = torch.nn.functional.relu(x_151, inplace=True) + x_151 = None + bottleneck_output_35 = torch.conv2d( + x_152, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_152 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_ = (None) + x_153 = torch.nn.functional.batch_norm( + bottleneck_output_35, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_35 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_ = (None) + x_154 = torch.nn.functional.relu(x_153, inplace=True) + x_153 = None + new_features_35 = torch.conv2d( + x_154, + l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_154 = l_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_ = (None) + concated_features_36 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + ], + 1, + ) + x_155 = torch.nn.functional.batch_norm( + concated_features_36, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_36 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_ = (None) + x_156 = torch.nn.functional.relu(x_155, inplace=True) + x_155 = None + bottleneck_output_36 = torch.conv2d( + x_156, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_156 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_ = (None) + x_157 = torch.nn.functional.batch_norm( + bottleneck_output_36, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_36 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_ = (None) + x_158 = torch.nn.functional.relu(x_157, inplace=True) + x_157 = None + new_features_36 = torch.conv2d( + x_158, + l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_158 = l_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_ = (None) + concated_features_37 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + ], + 1, + ) + x_159 = torch.nn.functional.batch_norm( + concated_features_37, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_37 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_ = (None) + x_160 = torch.nn.functional.relu(x_159, inplace=True) + x_159 = None + bottleneck_output_37 = torch.conv2d( + x_160, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_160 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_ = (None) + x_161 = torch.nn.functional.batch_norm( + bottleneck_output_37, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_37 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_ = (None) + x_162 = torch.nn.functional.relu(x_161, inplace=True) + x_161 = None + new_features_37 = torch.conv2d( + x_162, + l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_162 = l_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_ = (None) + concated_features_38 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + ], + 1, + ) + x_163 = torch.nn.functional.batch_norm( + concated_features_38, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_38 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_ = (None) + x_164 = torch.nn.functional.relu(x_163, inplace=True) + x_163 = None + bottleneck_output_38 = torch.conv2d( + x_164, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_164 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_ = (None) + x_165 = torch.nn.functional.batch_norm( + bottleneck_output_38, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_38 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_ = (None) + x_166 = torch.nn.functional.relu(x_165, inplace=True) + x_165 = None + new_features_38 = torch.conv2d( + x_166, + l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_166 = l_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_ = (None) + concated_features_39 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + ], + 1, + ) + x_167 = torch.nn.functional.batch_norm( + concated_features_39, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_39 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_ = (None) + x_168 = torch.nn.functional.relu(x_167, inplace=True) + x_167 = None + bottleneck_output_39 = torch.conv2d( + x_168, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_168 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_ = (None) + x_169 = torch.nn.functional.batch_norm( + bottleneck_output_39, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_39 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_ = (None) + x_170 = torch.nn.functional.relu(x_169, inplace=True) + x_169 = None + new_features_39 = torch.conv2d( + x_170, + l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_170 = l_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_ = (None) + concated_features_40 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + ], + 1, + ) + x_171 = torch.nn.functional.batch_norm( + concated_features_40, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_40 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_ = (None) + x_172 = torch.nn.functional.relu(x_171, inplace=True) + x_171 = None + bottleneck_output_40 = torch.conv2d( + x_172, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_172 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_ = (None) + x_173 = torch.nn.functional.batch_norm( + bottleneck_output_40, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_40 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_ = (None) + x_174 = torch.nn.functional.relu(x_173, inplace=True) + x_173 = None + new_features_40 = torch.conv2d( + x_174, + l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_174 = l_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_ = (None) + concated_features_41 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + ], + 1, + ) + x_175 = torch.nn.functional.batch_norm( + concated_features_41, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_41 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_ = (None) + x_176 = torch.nn.functional.relu(x_175, inplace=True) + x_175 = None + bottleneck_output_41 = torch.conv2d( + x_176, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_176 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_ = (None) + x_177 = torch.nn.functional.batch_norm( + bottleneck_output_41, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_41 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_ = (None) + x_178 = torch.nn.functional.relu(x_177, inplace=True) + x_177 = None + new_features_41 = torch.conv2d( + x_178, + l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_178 = l_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_ = (None) + input_12 = torch.cat( + [ + input_11, + new_features_18, + new_features_19, + new_features_20, + new_features_21, + new_features_22, + new_features_23, + new_features_24, + new_features_25, + new_features_26, + new_features_27, + new_features_28, + new_features_29, + new_features_30, + new_features_31, + new_features_32, + new_features_33, + new_features_34, + new_features_35, + new_features_36, + new_features_37, + new_features_38, + new_features_39, + new_features_40, + new_features_41, + ], + 1, + ) + input_11 = ( + new_features_18 + ) = ( + new_features_19 + ) = ( + new_features_20 + ) = ( + new_features_21 + ) = ( + new_features_22 + ) = ( + new_features_23 + ) = ( + new_features_24 + ) = ( + new_features_25 + ) = ( + new_features_26 + ) = ( + new_features_27 + ) = ( + new_features_28 + ) = ( + new_features_29 + ) = ( + new_features_30 + ) = ( + new_features_31 + ) = ( + new_features_32 + ) = ( + new_features_33 + ) = ( + new_features_34 + ) = ( + new_features_35 + ) = ( + new_features_36 + ) = ( + new_features_37 + ) = new_features_38 = new_features_39 = new_features_40 = new_features_41 = None + x_179 = torch.nn.functional.batch_norm( + input_12, + l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_, + l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_, + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_, + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_12 = l_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_ = l_self_modules_features_modules_transition3_modules_norm_buffers_running_var_ = ( + l_self_modules_features_modules_transition3_modules_norm_parameters_weight_ + ) = ( + l_self_modules_features_modules_transition3_modules_norm_parameters_bias_ + ) = None + x_180 = torch.nn.functional.relu(x_179, inplace=True) + x_179 = None + input_13 = torch.conv2d( + x_180, + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_180 = ( + l_self_modules_features_modules_transition3_modules_conv_parameters_weight_ + ) = None + input_14 = torch._C._nn.avg_pool2d(input_13, 2, 2, 0, False, True, None) + input_13 = None + concated_features_42 = torch.cat([input_14], 1) + x_181 = torch.nn.functional.batch_norm( + concated_features_42, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_42 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_ = (None) + x_182 = torch.nn.functional.relu(x_181, inplace=True) + x_181 = None + bottleneck_output_42 = torch.conv2d( + x_182, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_182 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_ = (None) + x_183 = torch.nn.functional.batch_norm( + bottleneck_output_42, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_42 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_ = (None) + x_184 = torch.nn.functional.relu(x_183, inplace=True) + x_183 = None + new_features_42 = torch.conv2d( + x_184, + l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_184 = l_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_ = (None) + concated_features_43 = torch.cat([input_14, new_features_42], 1) + x_185 = torch.nn.functional.batch_norm( + concated_features_43, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_43 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_ = (None) + x_186 = torch.nn.functional.relu(x_185, inplace=True) + x_185 = None + bottleneck_output_43 = torch.conv2d( + x_186, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_186 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_ = (None) + x_187 = torch.nn.functional.batch_norm( + bottleneck_output_43, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_43 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_ = (None) + x_188 = torch.nn.functional.relu(x_187, inplace=True) + x_187 = None + new_features_43 = torch.conv2d( + x_188, + l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_188 = l_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_ = (None) + concated_features_44 = torch.cat( + [input_14, new_features_42, new_features_43], 1 + ) + x_189 = torch.nn.functional.batch_norm( + concated_features_44, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_44 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_ = (None) + x_190 = torch.nn.functional.relu(x_189, inplace=True) + x_189 = None + bottleneck_output_44 = torch.conv2d( + x_190, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_190 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_ = (None) + x_191 = torch.nn.functional.batch_norm( + bottleneck_output_44, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_44 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_ = (None) + x_192 = torch.nn.functional.relu(x_191, inplace=True) + x_191 = None + new_features_44 = torch.conv2d( + x_192, + l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_192 = l_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_ = (None) + concated_features_45 = torch.cat( + [input_14, new_features_42, new_features_43, new_features_44], 1 + ) + x_193 = torch.nn.functional.batch_norm( + concated_features_45, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_45 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_ = (None) + x_194 = torch.nn.functional.relu(x_193, inplace=True) + x_193 = None + bottleneck_output_45 = torch.conv2d( + x_194, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_194 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_ = (None) + x_195 = torch.nn.functional.batch_norm( + bottleneck_output_45, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_45 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_ = (None) + x_196 = torch.nn.functional.relu(x_195, inplace=True) + x_195 = None + new_features_45 = torch.conv2d( + x_196, + l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_196 = l_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_ = (None) + concated_features_46 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + ], + 1, + ) + x_197 = torch.nn.functional.batch_norm( + concated_features_46, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_46 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_ = (None) + x_198 = torch.nn.functional.relu(x_197, inplace=True) + x_197 = None + bottleneck_output_46 = torch.conv2d( + x_198, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_198 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_ = (None) + x_199 = torch.nn.functional.batch_norm( + bottleneck_output_46, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_46 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_ = (None) + x_200 = torch.nn.functional.relu(x_199, inplace=True) + x_199 = None + new_features_46 = torch.conv2d( + x_200, + l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_200 = l_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_ = (None) + concated_features_47 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + ], + 1, + ) + x_201 = torch.nn.functional.batch_norm( + concated_features_47, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_47 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_ = (None) + x_202 = torch.nn.functional.relu(x_201, inplace=True) + x_201 = None + bottleneck_output_47 = torch.conv2d( + x_202, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_202 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_ = (None) + x_203 = torch.nn.functional.batch_norm( + bottleneck_output_47, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_47 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_ = (None) + x_204 = torch.nn.functional.relu(x_203, inplace=True) + x_203 = None + new_features_47 = torch.conv2d( + x_204, + l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_204 = l_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_ = (None) + concated_features_48 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + ], + 1, + ) + x_205 = torch.nn.functional.batch_norm( + concated_features_48, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_48 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_ = (None) + x_206 = torch.nn.functional.relu(x_205, inplace=True) + x_205 = None + bottleneck_output_48 = torch.conv2d( + x_206, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_206 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_ = (None) + x_207 = torch.nn.functional.batch_norm( + bottleneck_output_48, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_48 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_ = (None) + x_208 = torch.nn.functional.relu(x_207, inplace=True) + x_207 = None + new_features_48 = torch.conv2d( + x_208, + l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_208 = l_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_ = (None) + concated_features_49 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + ], + 1, + ) + x_209 = torch.nn.functional.batch_norm( + concated_features_49, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_49 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_ = (None) + x_210 = torch.nn.functional.relu(x_209, inplace=True) + x_209 = None + bottleneck_output_49 = torch.conv2d( + x_210, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_210 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_ = (None) + x_211 = torch.nn.functional.batch_norm( + bottleneck_output_49, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_49 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_ = (None) + x_212 = torch.nn.functional.relu(x_211, inplace=True) + x_211 = None + new_features_49 = torch.conv2d( + x_212, + l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_212 = l_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_ = (None) + concated_features_50 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + ], + 1, + ) + x_213 = torch.nn.functional.batch_norm( + concated_features_50, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_50 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_ = (None) + x_214 = torch.nn.functional.relu(x_213, inplace=True) + x_213 = None + bottleneck_output_50 = torch.conv2d( + x_214, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_214 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_ = (None) + x_215 = torch.nn.functional.batch_norm( + bottleneck_output_50, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_50 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_ = (None) + x_216 = torch.nn.functional.relu(x_215, inplace=True) + x_215 = None + new_features_50 = torch.conv2d( + x_216, + l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_216 = l_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_ = (None) + concated_features_51 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + ], + 1, + ) + x_217 = torch.nn.functional.batch_norm( + concated_features_51, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_51 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_ = (None) + x_218 = torch.nn.functional.relu(x_217, inplace=True) + x_217 = None + bottleneck_output_51 = torch.conv2d( + x_218, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_218 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_ = (None) + x_219 = torch.nn.functional.batch_norm( + bottleneck_output_51, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_51 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_ = (None) + x_220 = torch.nn.functional.relu(x_219, inplace=True) + x_219 = None + new_features_51 = torch.conv2d( + x_220, + l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_220 = l_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_ = (None) + concated_features_52 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + ], + 1, + ) + x_221 = torch.nn.functional.batch_norm( + concated_features_52, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_52 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_ = (None) + x_222 = torch.nn.functional.relu(x_221, inplace=True) + x_221 = None + bottleneck_output_52 = torch.conv2d( + x_222, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_222 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_ = (None) + x_223 = torch.nn.functional.batch_norm( + bottleneck_output_52, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_52 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_ = (None) + x_224 = torch.nn.functional.relu(x_223, inplace=True) + x_223 = None + new_features_52 = torch.conv2d( + x_224, + l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_224 = l_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_ = (None) + concated_features_53 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + ], + 1, + ) + x_225 = torch.nn.functional.batch_norm( + concated_features_53, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_53 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_ = (None) + x_226 = torch.nn.functional.relu(x_225, inplace=True) + x_225 = None + bottleneck_output_53 = torch.conv2d( + x_226, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_226 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_ = (None) + x_227 = torch.nn.functional.batch_norm( + bottleneck_output_53, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_53 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_ = (None) + x_228 = torch.nn.functional.relu(x_227, inplace=True) + x_227 = None + new_features_53 = torch.conv2d( + x_228, + l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_228 = l_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_ = (None) + concated_features_54 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + ], + 1, + ) + x_229 = torch.nn.functional.batch_norm( + concated_features_54, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_54 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_ = (None) + x_230 = torch.nn.functional.relu(x_229, inplace=True) + x_229 = None + bottleneck_output_54 = torch.conv2d( + x_230, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_230 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_ = (None) + x_231 = torch.nn.functional.batch_norm( + bottleneck_output_54, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_54 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_ = (None) + x_232 = torch.nn.functional.relu(x_231, inplace=True) + x_231 = None + new_features_54 = torch.conv2d( + x_232, + l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_232 = l_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_ = (None) + concated_features_55 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + ], + 1, + ) + x_233 = torch.nn.functional.batch_norm( + concated_features_55, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_55 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_ = (None) + x_234 = torch.nn.functional.relu(x_233, inplace=True) + x_233 = None + bottleneck_output_55 = torch.conv2d( + x_234, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_234 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_ = (None) + x_235 = torch.nn.functional.batch_norm( + bottleneck_output_55, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_55 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_ = (None) + x_236 = torch.nn.functional.relu(x_235, inplace=True) + x_235 = None + new_features_55 = torch.conv2d( + x_236, + l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_236 = l_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_ = (None) + concated_features_56 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + ], + 1, + ) + x_237 = torch.nn.functional.batch_norm( + concated_features_56, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_56 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_ = (None) + x_238 = torch.nn.functional.relu(x_237, inplace=True) + x_237 = None + bottleneck_output_56 = torch.conv2d( + x_238, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_238 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_ = (None) + x_239 = torch.nn.functional.batch_norm( + bottleneck_output_56, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_56 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_ = (None) + x_240 = torch.nn.functional.relu(x_239, inplace=True) + x_239 = None + new_features_56 = torch.conv2d( + x_240, + l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_240 = l_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_ = (None) + concated_features_57 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + ], + 1, + ) + x_241 = torch.nn.functional.batch_norm( + concated_features_57, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + concated_features_57 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_ = (None) + x_242 = torch.nn.functional.relu(x_241, inplace=True) + x_241 = None + bottleneck_output_57 = torch.conv2d( + x_242, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_242 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_ = (None) + x_243 = torch.nn.functional.batch_norm( + bottleneck_output_57, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + bottleneck_output_57 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_ = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_ = (None) + x_244 = torch.nn.functional.relu(x_243, inplace=True) + x_243 = None + new_features_57 = torch.conv2d( + x_244, + l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + x_244 = l_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_ = (None) + input_15 = torch.cat( + [ + input_14, + new_features_42, + new_features_43, + new_features_44, + new_features_45, + new_features_46, + new_features_47, + new_features_48, + new_features_49, + new_features_50, + new_features_51, + new_features_52, + new_features_53, + new_features_54, + new_features_55, + new_features_56, + new_features_57, + ], + 1, + ) + input_14 = ( + new_features_42 + ) = ( + new_features_43 + ) = ( + new_features_44 + ) = ( + new_features_45 + ) = ( + new_features_46 + ) = ( + new_features_47 + ) = ( + new_features_48 + ) = ( + new_features_49 + ) = ( + new_features_50 + ) = ( + new_features_51 + ) = ( + new_features_52 + ) = ( + new_features_53 + ) = new_features_54 = new_features_55 = new_features_56 = new_features_57 = None + x_245 = torch.nn.functional.batch_norm( + input_15, + l_self_modules_features_modules_norm5_buffers_running_mean_, + l_self_modules_features_modules_norm5_buffers_running_var_, + l_self_modules_features_modules_norm5_parameters_weight_, + l_self_modules_features_modules_norm5_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_15 = ( + l_self_modules_features_modules_norm5_buffers_running_mean_ + ) = ( + l_self_modules_features_modules_norm5_buffers_running_var_ + ) = ( + l_self_modules_features_modules_norm5_parameters_weight_ + ) = l_self_modules_features_modules_norm5_parameters_bias_ = None + x_246 = torch.nn.functional.relu(x_245, inplace=True) + x_245 = None + x_247 = torch.nn.functional.adaptive_avg_pool2d(x_246, 1) + x_246 = None + x_248 = x_247.flatten(1, -1) + x_247 = None + x_249 = torch.nn.functional.dropout(x_248, 0.0, False, False) + x_248 = None + x_250 = torch._C._nn.linear( + x_249, + l_self_modules_classifier_parameters_weight_, + l_self_modules_classifier_parameters_bias_, + ) + x_249 = ( + l_self_modules_classifier_parameters_weight_ + ) = l_self_modules_classifier_parameters_bias_ = None + return (x_250,) diff --git a/samples/timm/densenetblur121d.ra_in1k/weight_meta.py b/samples/timm/densenetblur121d.ra_in1k/weight_meta.py new file mode 100644 index 0000000000..487168616c --- /dev/null +++ b/samples/timm/densenetblur121d.ra_in1k/weight_meta.py @@ -0,0 +1,6200 @@ +class Program_weight_tensor_meta_L_self_modules_features_modules_conv0_parameters_weight_: + name = "L_self_modules_features_modules_conv0_parameters_weight_" + shape = [32, 3, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.282 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.224 + std = 1.286 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_buffers_running_mean_: + name = "L_self_modules_features_modules_norm0_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_buffers_running_var_: + name = "L_self_modules_features_modules_norm0_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_parameters_weight_: + name = "L_self_modules_features_modules_norm0_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm0_parameters_bias_: + name = "L_self_modules_features_modules_norm0_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_conv1_parameters_weight_" + shape = [32, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_norm1_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_norm1_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_norm1_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_norm1_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_conv2_parameters_weight_" + shape = [64, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_norm2_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_norm2_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_norm2_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_norm2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_pool0_modules_1_buffers_filt_: + name = "L_self_modules_features_modules_pool0_modules_1_buffers_filt_" + shape = [64, 1, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.111 + std = 0.057 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.178 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [96] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 96, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.144 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.113 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 224, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.095 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock1_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition1_modules_norm_buffers_running_mean_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition1_modules_norm_buffers_running_var_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition1_modules_norm_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition1_modules_norm_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition1_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition1_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [160] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 160, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.112 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [192] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 192, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.102 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 224, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.094 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.089 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 288, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [128, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [128, 352, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.076 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [128, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [128, 416, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.069 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [128, 448, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.067 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [128, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.065 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock2_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition2_modules_norm_buffers_running_mean_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition2_modules_norm_buffers_running_var_" + ) + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition2_modules_norm_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition2_modules_norm_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition2_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition2_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [288] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 288, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [320] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 320, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.079 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [352] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 352, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.075 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [384] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 384, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.072 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [416] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 416, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.069 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [448] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [128, 448, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.067 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [480] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [128, 480, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.064 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [128, 544, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.061 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [128, 576, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [128, 608, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.057 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv1_parameters_weight_" + shape = [128, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer13_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_mean_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_buffers_running_var_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_weight_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv1_parameters_weight_" + shape = [128, 672, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.054 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer14_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_mean_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_buffers_running_var_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm1_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv1_parameters_weight_" + shape = [128, 704, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer15_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_mean_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_buffers_running_var_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_weight_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm1_parameters_bias_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv1_parameters_weight_" + shape = [128, 736, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.052 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer16_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv1_parameters_weight_" + shape = [128, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer17_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_mean_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_buffers_running_var_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_weight_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm1_parameters_bias_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv1_parameters_weight_" + shape = [128, 800, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.050 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer18_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_mean_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_buffers_running_var_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_weight_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm1_parameters_bias_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv1_parameters_weight_" + shape = [128, 832, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.049 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer19_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_mean_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_buffers_running_var_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_weight_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm1_parameters_bias_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv1_parameters_weight_" + shape = [128, 864, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer20_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_mean_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_buffers_running_var_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_weight_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm1_parameters_bias_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv1_parameters_weight_" + shape = [128, 896, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer21_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_mean_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_buffers_running_var_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_weight_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm1_parameters_bias_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv1_parameters_weight_" + shape = [128, 928, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer22_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm1_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv1_parameters_weight_" + shape = [128, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer23_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_mean_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_buffers_running_var_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_weight_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm1_parameters_bias_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv1_parameters_weight_" + shape = [128, 992, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.045 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock3_modules_denselayer24_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_: + name = ( + "L_self_modules_features_modules_transition3_modules_norm_buffers_running_mean_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_: + name = ( + "L_self_modules_features_modules_transition3_modules_norm_buffers_running_var_" + ) + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_parameters_weight_: + name = "L_self_modules_features_modules_transition3_modules_norm_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_norm_parameters_bias_: + name = "L_self_modules_features_modules_transition3_modules_norm_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_transition3_modules_conv_parameters_weight_: + name = "L_self_modules_features_modules_transition3_modules_conv_parameters_weight_" + shape = [512, 1024, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.044 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv1_parameters_weight_" + shape = [128, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer1_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_mean_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_buffers_running_var_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_weight_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm1_parameters_bias_" + shape = [544] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv1_parameters_weight_" + shape = [128, 544, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.061 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer2_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_mean_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_buffers_running_var_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_weight_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm1_parameters_bias_" + shape = [576] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv1_parameters_weight_" + shape = [128, 576, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer3_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_mean_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_buffers_running_var_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_weight_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm1_parameters_bias_" + shape = [608] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv1_parameters_weight_" + shape = [128, 608, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.058 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer4_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_mean_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_buffers_running_var_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_weight_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm1_parameters_bias_" + shape = [640] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv1_parameters_weight_" + shape = [128, 640, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.056 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer5_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_mean_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_buffers_running_var_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_weight_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm1_parameters_bias_" + shape = [672] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv1_parameters_weight_" + shape = [128, 672, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.055 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer6_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_mean_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_buffers_running_var_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_weight_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm1_parameters_bias_" + shape = [704] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv1_parameters_weight_" + shape = [128, 704, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.053 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer7_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_mean_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_buffers_running_var_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_weight_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm1_parameters_bias_" + shape = [736] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv1_parameters_weight_" + shape = [128, 736, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.052 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer8_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.041 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_mean_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_buffers_running_var_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_weight_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm1_parameters_bias_" + shape = [768] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv1_parameters_weight_" + shape = [128, 768, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.051 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer9_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_mean_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_buffers_running_var_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_weight_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm1_parameters_bias_" + shape = [800] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv1_parameters_weight_" + shape = [128, 800, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.050 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer10_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_mean_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_buffers_running_var_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_weight_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm1_parameters_bias_" + shape = [832] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv1_parameters_weight_" + shape = [128, 832, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.049 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer11_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_mean_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_buffers_running_var_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_weight_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm1_parameters_bias_" + shape = [864] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv1_parameters_weight_" + shape = [128, 864, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.048 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer12_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_mean_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_buffers_running_var_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_weight_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm1_parameters_bias_" + shape = [896] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv1_parameters_weight_" + shape = [128, 896, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.047 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer13_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_mean_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_buffers_running_var_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_weight_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm1_parameters_bias_" + shape = [928] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv1_parameters_weight_" + shape = [128, 928, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer14_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_mean_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_buffers_running_var_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_weight_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm1_parameters_bias_" + shape = [960] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv1_parameters_weight_" + shape = [128, 960, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.046 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer15_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_mean_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_buffers_running_var_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_weight_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm1_parameters_bias_" + shape = [992] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv1_parameters_weight_" + shape = [128, 992, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.045 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_norm2_parameters_bias_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_: + name = "L_self_modules_features_modules_denseblock4_modules_denselayer16_modules_conv2_parameters_weight_" + shape = [32, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_buffers_running_mean_: + name = "L_self_modules_features_modules_norm5_buffers_running_mean_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_buffers_running_var_: + name = "L_self_modules_features_modules_norm5_buffers_running_var_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_parameters_weight_: + name = "L_self_modules_features_modules_norm5_parameters_weight_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_features_modules_norm5_parameters_bias_: + name = "L_self_modules_features_modules_norm5_parameters_bias_" + shape = [1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_classifier_parameters_weight_: + name = "L_self_modules_classifier_parameters_weight_" + shape = [1000, 1024] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.018 + data = None + + +class Program_weight_tensor_meta_L_self_modules_classifier_parameters_bias_: + name = "L_self_modules_classifier_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None diff --git a/samples/timm/dla34.in1k/graph_hash.txt b/samples/timm/dla34.in1k/graph_hash.txt new file mode 100644 index 0000000000..7938093160 --- /dev/null +++ b/samples/timm/dla34.in1k/graph_hash.txt @@ -0,0 +1 @@ +13e26b9a256b6c4b6ff13fdc717e0de35f09711d723103383755a556a61ce457 \ No newline at end of file diff --git a/samples/timm/dla34.in1k/graph_net.json b/samples/timm/dla34.in1k/graph_net.json new file mode 100644 index 0000000000..1373fe3b58 --- /dev/null +++ b/samples/timm/dla34.in1k/graph_net.json @@ -0,0 +1,5 @@ +{ + "framework": "torch", + "num_devices_required": 1, + "num_nodes_required": 1 +} \ No newline at end of file diff --git a/samples/timm/dla34.in1k/input_meta.py b/samples/timm/dla34.in1k/input_meta.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/dla34.in1k/input_tensor_constraints.py b/samples/timm/dla34.in1k/input_tensor_constraints.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/samples/timm/dla34.in1k/model.py b/samples/timm/dla34.in1k/model.py new file mode 100644 index 0000000000..21ca6551ec --- /dev/null +++ b/samples/timm/dla34.in1k/model.py @@ -0,0 +1,1596 @@ +import torch + + +class GraphModule(torch.nn.Module): + def forward( + self, + L_self_modules_base_layer_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + s1: torch.SymInt, + L_x_: torch.Tensor, + L_self_modules_base_layer_modules_1_buffers_running_mean_: torch.Tensor, + L_self_modules_base_layer_modules_1_buffers_running_var_: torch.Tensor, + L_self_modules_base_layer_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_base_layer_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level0_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level0_modules_1_buffers_running_mean_: torch.Tensor, + L_self_modules_level0_modules_1_buffers_running_var_: torch.Tensor, + L_self_modules_level0_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level0_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level1_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level1_modules_1_buffers_running_mean_: torch.Tensor, + L_self_modules_level1_modules_1_buffers_running_var_: torch.Tensor, + L_self_modules_level1_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level1_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_project_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_project_modules_1_buffers_running_mean_: torch.Tensor, + L_self_modules_level2_modules_project_modules_1_buffers_running_var_: torch.Tensor, + L_self_modules_level2_modules_project_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_project_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_tree1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_tree1_modules_bn1_buffers_running_mean_: torch.Tensor, + L_self_modules_level2_modules_tree1_modules_bn1_buffers_running_var_: torch.Tensor, + L_self_modules_level2_modules_tree1_modules_bn1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_tree1_modules_bn1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_tree1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_tree1_modules_bn2_buffers_running_mean_: torch.Tensor, + L_self_modules_level2_modules_tree1_modules_bn2_buffers_running_var_: torch.Tensor, + L_self_modules_level2_modules_tree1_modules_bn2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_tree1_modules_bn2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_tree2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_tree2_modules_bn1_buffers_running_mean_: torch.Tensor, + L_self_modules_level2_modules_tree2_modules_bn1_buffers_running_var_: torch.Tensor, + L_self_modules_level2_modules_tree2_modules_bn1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_tree2_modules_bn1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_tree2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_tree2_modules_bn2_buffers_running_mean_: torch.Tensor, + L_self_modules_level2_modules_tree2_modules_bn2_buffers_running_var_: torch.Tensor, + L_self_modules_level2_modules_tree2_modules_bn2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_tree2_modules_bn2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_root_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_root_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_level2_modules_root_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_level2_modules_root_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level2_modules_root_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_project_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_mean_: torch.Tensor, + L_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_var_: torch.Tensor, + L_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_tree1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_: torch.Tensor, + L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_: torch.Tensor, + L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_tree1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_: torch.Tensor, + L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_: torch.Tensor, + L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_tree2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_: torch.Tensor, + L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_: torch.Tensor, + L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_tree2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_: torch.Tensor, + L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_: torch.Tensor, + L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_root_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_tree1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_: torch.Tensor, + L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_: torch.Tensor, + L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_tree1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_: torch.Tensor, + L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_: torch.Tensor, + L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_tree2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_: torch.Tensor, + L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_: torch.Tensor, + L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_tree2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_: torch.Tensor, + L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_: torch.Tensor, + L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_root_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_project_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_mean_: torch.Tensor, + L_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_var_: torch.Tensor, + L_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_tree1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_: torch.Tensor, + L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_: torch.Tensor, + L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_tree1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_: torch.Tensor, + L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_: torch.Tensor, + L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_tree2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_: torch.Tensor, + L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_: torch.Tensor, + L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_tree2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_: torch.Tensor, + L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_: torch.Tensor, + L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_root_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_tree1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_: torch.Tensor, + L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_: torch.Tensor, + L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_tree1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_: torch.Tensor, + L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_: torch.Tensor, + L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_tree2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_: torch.Tensor, + L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_: torch.Tensor, + L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_tree2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_: torch.Tensor, + L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_: torch.Tensor, + L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_root_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_project_modules_0_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_project_modules_1_buffers_running_mean_: torch.Tensor, + L_self_modules_level5_modules_project_modules_1_buffers_running_var_: torch.Tensor, + L_self_modules_level5_modules_project_modules_1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_project_modules_1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_tree1_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_tree1_modules_bn1_buffers_running_mean_: torch.Tensor, + L_self_modules_level5_modules_tree1_modules_bn1_buffers_running_var_: torch.Tensor, + L_self_modules_level5_modules_tree1_modules_bn1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_tree1_modules_bn1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_tree1_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_tree1_modules_bn2_buffers_running_mean_: torch.Tensor, + L_self_modules_level5_modules_tree1_modules_bn2_buffers_running_var_: torch.Tensor, + L_self_modules_level5_modules_tree1_modules_bn2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_tree1_modules_bn2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_tree2_modules_conv1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_tree2_modules_bn1_buffers_running_mean_: torch.Tensor, + L_self_modules_level5_modules_tree2_modules_bn1_buffers_running_var_: torch.Tensor, + L_self_modules_level5_modules_tree2_modules_bn1_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_tree2_modules_bn1_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_tree2_modules_conv2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_tree2_modules_bn2_buffers_running_mean_: torch.Tensor, + L_self_modules_level5_modules_tree2_modules_bn2_buffers_running_var_: torch.Tensor, + L_self_modules_level5_modules_tree2_modules_bn2_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_tree2_modules_bn2_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_root_modules_conv_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_root_modules_bn_buffers_running_mean_: torch.Tensor, + L_self_modules_level5_modules_root_modules_bn_buffers_running_var_: torch.Tensor, + L_self_modules_level5_modules_root_modules_bn_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_level5_modules_root_modules_bn_parameters_bias_: torch.nn.parameter.Parameter, + L_self_modules_fc_parameters_weight_: torch.nn.parameter.Parameter, + L_self_modules_fc_parameters_bias_: torch.nn.parameter.Parameter, + ): + l_self_modules_base_layer_modules_0_parameters_weight_ = ( + L_self_modules_base_layer_modules_0_parameters_weight_ + ) + l_x_ = L_x_ + l_self_modules_base_layer_modules_1_buffers_running_mean_ = ( + L_self_modules_base_layer_modules_1_buffers_running_mean_ + ) + l_self_modules_base_layer_modules_1_buffers_running_var_ = ( + L_self_modules_base_layer_modules_1_buffers_running_var_ + ) + l_self_modules_base_layer_modules_1_parameters_weight_ = ( + L_self_modules_base_layer_modules_1_parameters_weight_ + ) + l_self_modules_base_layer_modules_1_parameters_bias_ = ( + L_self_modules_base_layer_modules_1_parameters_bias_ + ) + l_self_modules_level0_modules_0_parameters_weight_ = ( + L_self_modules_level0_modules_0_parameters_weight_ + ) + l_self_modules_level0_modules_1_buffers_running_mean_ = ( + L_self_modules_level0_modules_1_buffers_running_mean_ + ) + l_self_modules_level0_modules_1_buffers_running_var_ = ( + L_self_modules_level0_modules_1_buffers_running_var_ + ) + l_self_modules_level0_modules_1_parameters_weight_ = ( + L_self_modules_level0_modules_1_parameters_weight_ + ) + l_self_modules_level0_modules_1_parameters_bias_ = ( + L_self_modules_level0_modules_1_parameters_bias_ + ) + l_self_modules_level1_modules_0_parameters_weight_ = ( + L_self_modules_level1_modules_0_parameters_weight_ + ) + l_self_modules_level1_modules_1_buffers_running_mean_ = ( + L_self_modules_level1_modules_1_buffers_running_mean_ + ) + l_self_modules_level1_modules_1_buffers_running_var_ = ( + L_self_modules_level1_modules_1_buffers_running_var_ + ) + l_self_modules_level1_modules_1_parameters_weight_ = ( + L_self_modules_level1_modules_1_parameters_weight_ + ) + l_self_modules_level1_modules_1_parameters_bias_ = ( + L_self_modules_level1_modules_1_parameters_bias_ + ) + l_self_modules_level2_modules_project_modules_0_parameters_weight_ = ( + L_self_modules_level2_modules_project_modules_0_parameters_weight_ + ) + l_self_modules_level2_modules_project_modules_1_buffers_running_mean_ = ( + L_self_modules_level2_modules_project_modules_1_buffers_running_mean_ + ) + l_self_modules_level2_modules_project_modules_1_buffers_running_var_ = ( + L_self_modules_level2_modules_project_modules_1_buffers_running_var_ + ) + l_self_modules_level2_modules_project_modules_1_parameters_weight_ = ( + L_self_modules_level2_modules_project_modules_1_parameters_weight_ + ) + l_self_modules_level2_modules_project_modules_1_parameters_bias_ = ( + L_self_modules_level2_modules_project_modules_1_parameters_bias_ + ) + l_self_modules_level2_modules_tree1_modules_conv1_parameters_weight_ = ( + L_self_modules_level2_modules_tree1_modules_conv1_parameters_weight_ + ) + l_self_modules_level2_modules_tree1_modules_bn1_buffers_running_mean_ = ( + L_self_modules_level2_modules_tree1_modules_bn1_buffers_running_mean_ + ) + l_self_modules_level2_modules_tree1_modules_bn1_buffers_running_var_ = ( + L_self_modules_level2_modules_tree1_modules_bn1_buffers_running_var_ + ) + l_self_modules_level2_modules_tree1_modules_bn1_parameters_weight_ = ( + L_self_modules_level2_modules_tree1_modules_bn1_parameters_weight_ + ) + l_self_modules_level2_modules_tree1_modules_bn1_parameters_bias_ = ( + L_self_modules_level2_modules_tree1_modules_bn1_parameters_bias_ + ) + l_self_modules_level2_modules_tree1_modules_conv2_parameters_weight_ = ( + L_self_modules_level2_modules_tree1_modules_conv2_parameters_weight_ + ) + l_self_modules_level2_modules_tree1_modules_bn2_buffers_running_mean_ = ( + L_self_modules_level2_modules_tree1_modules_bn2_buffers_running_mean_ + ) + l_self_modules_level2_modules_tree1_modules_bn2_buffers_running_var_ = ( + L_self_modules_level2_modules_tree1_modules_bn2_buffers_running_var_ + ) + l_self_modules_level2_modules_tree1_modules_bn2_parameters_weight_ = ( + L_self_modules_level2_modules_tree1_modules_bn2_parameters_weight_ + ) + l_self_modules_level2_modules_tree1_modules_bn2_parameters_bias_ = ( + L_self_modules_level2_modules_tree1_modules_bn2_parameters_bias_ + ) + l_self_modules_level2_modules_tree2_modules_conv1_parameters_weight_ = ( + L_self_modules_level2_modules_tree2_modules_conv1_parameters_weight_ + ) + l_self_modules_level2_modules_tree2_modules_bn1_buffers_running_mean_ = ( + L_self_modules_level2_modules_tree2_modules_bn1_buffers_running_mean_ + ) + l_self_modules_level2_modules_tree2_modules_bn1_buffers_running_var_ = ( + L_self_modules_level2_modules_tree2_modules_bn1_buffers_running_var_ + ) + l_self_modules_level2_modules_tree2_modules_bn1_parameters_weight_ = ( + L_self_modules_level2_modules_tree2_modules_bn1_parameters_weight_ + ) + l_self_modules_level2_modules_tree2_modules_bn1_parameters_bias_ = ( + L_self_modules_level2_modules_tree2_modules_bn1_parameters_bias_ + ) + l_self_modules_level2_modules_tree2_modules_conv2_parameters_weight_ = ( + L_self_modules_level2_modules_tree2_modules_conv2_parameters_weight_ + ) + l_self_modules_level2_modules_tree2_modules_bn2_buffers_running_mean_ = ( + L_self_modules_level2_modules_tree2_modules_bn2_buffers_running_mean_ + ) + l_self_modules_level2_modules_tree2_modules_bn2_buffers_running_var_ = ( + L_self_modules_level2_modules_tree2_modules_bn2_buffers_running_var_ + ) + l_self_modules_level2_modules_tree2_modules_bn2_parameters_weight_ = ( + L_self_modules_level2_modules_tree2_modules_bn2_parameters_weight_ + ) + l_self_modules_level2_modules_tree2_modules_bn2_parameters_bias_ = ( + L_self_modules_level2_modules_tree2_modules_bn2_parameters_bias_ + ) + l_self_modules_level2_modules_root_modules_conv_parameters_weight_ = ( + L_self_modules_level2_modules_root_modules_conv_parameters_weight_ + ) + l_self_modules_level2_modules_root_modules_bn_buffers_running_mean_ = ( + L_self_modules_level2_modules_root_modules_bn_buffers_running_mean_ + ) + l_self_modules_level2_modules_root_modules_bn_buffers_running_var_ = ( + L_self_modules_level2_modules_root_modules_bn_buffers_running_var_ + ) + l_self_modules_level2_modules_root_modules_bn_parameters_weight_ = ( + L_self_modules_level2_modules_root_modules_bn_parameters_weight_ + ) + l_self_modules_level2_modules_root_modules_bn_parameters_bias_ = ( + L_self_modules_level2_modules_root_modules_bn_parameters_bias_ + ) + l_self_modules_level3_modules_tree1_modules_project_modules_0_parameters_weight_ = L_self_modules_level3_modules_tree1_modules_project_modules_0_parameters_weight_ + l_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_mean_ = L_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_mean_ + l_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_var_ = L_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_var_ + l_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_weight_ = L_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_weight_ + l_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_bias_ = L_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_bias_ + l_self_modules_level3_modules_tree1_modules_tree1_modules_conv1_parameters_weight_ = L_self_modules_level3_modules_tree1_modules_tree1_modules_conv1_parameters_weight_ + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_ = L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_ + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_ = L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_ + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_weight_ = L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_weight_ + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_bias_ = L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_bias_ + l_self_modules_level3_modules_tree1_modules_tree1_modules_conv2_parameters_weight_ = L_self_modules_level3_modules_tree1_modules_tree1_modules_conv2_parameters_weight_ + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_ = L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_ + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_ = L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_ + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_weight_ = L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_weight_ + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_bias_ = L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_bias_ + l_self_modules_level3_modules_tree1_modules_tree2_modules_conv1_parameters_weight_ = L_self_modules_level3_modules_tree1_modules_tree2_modules_conv1_parameters_weight_ + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_ = L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_ + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_ = L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_ + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_weight_ = L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_weight_ + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_bias_ = L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_bias_ + l_self_modules_level3_modules_tree1_modules_tree2_modules_conv2_parameters_weight_ = L_self_modules_level3_modules_tree1_modules_tree2_modules_conv2_parameters_weight_ + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_ = L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_ + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_ = L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_ + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_weight_ = L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_weight_ + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_bias_ = L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_bias_ + l_self_modules_level3_modules_tree1_modules_root_modules_conv_parameters_weight_ = L_self_modules_level3_modules_tree1_modules_root_modules_conv_parameters_weight_ + l_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_mean_ = L_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_mean_ + l_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_var_ = L_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_var_ + l_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_weight_ = L_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_weight_ + l_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_bias_ = ( + L_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_bias_ + ) + l_self_modules_level3_modules_tree2_modules_tree1_modules_conv1_parameters_weight_ = L_self_modules_level3_modules_tree2_modules_tree1_modules_conv1_parameters_weight_ + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_ = L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_ + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_ = L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_ + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_weight_ = L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_weight_ + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_bias_ = L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_bias_ + l_self_modules_level3_modules_tree2_modules_tree1_modules_conv2_parameters_weight_ = L_self_modules_level3_modules_tree2_modules_tree1_modules_conv2_parameters_weight_ + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_ = L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_ + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_ = L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_ + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_weight_ = L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_weight_ + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_bias_ = L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_bias_ + l_self_modules_level3_modules_tree2_modules_tree2_modules_conv1_parameters_weight_ = L_self_modules_level3_modules_tree2_modules_tree2_modules_conv1_parameters_weight_ + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_ = L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_ + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_ = L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_ + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_weight_ = L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_weight_ + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_bias_ = L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_bias_ + l_self_modules_level3_modules_tree2_modules_tree2_modules_conv2_parameters_weight_ = L_self_modules_level3_modules_tree2_modules_tree2_modules_conv2_parameters_weight_ + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_ = L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_ + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_ = L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_ + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_weight_ = L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_weight_ + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_bias_ = L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_bias_ + l_self_modules_level3_modules_tree2_modules_root_modules_conv_parameters_weight_ = L_self_modules_level3_modules_tree2_modules_root_modules_conv_parameters_weight_ + l_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_mean_ = L_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_mean_ + l_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_var_ = L_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_var_ + l_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_weight_ = L_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_weight_ + l_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_bias_ = ( + L_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_bias_ + ) + l_self_modules_level4_modules_tree1_modules_project_modules_0_parameters_weight_ = L_self_modules_level4_modules_tree1_modules_project_modules_0_parameters_weight_ + l_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_mean_ = L_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_mean_ + l_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_var_ = L_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_var_ + l_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_weight_ = L_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_weight_ + l_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_bias_ = L_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_bias_ + l_self_modules_level4_modules_tree1_modules_tree1_modules_conv1_parameters_weight_ = L_self_modules_level4_modules_tree1_modules_tree1_modules_conv1_parameters_weight_ + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_ = L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_ + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_ = L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_ + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_weight_ = L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_weight_ + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_bias_ = L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_bias_ + l_self_modules_level4_modules_tree1_modules_tree1_modules_conv2_parameters_weight_ = L_self_modules_level4_modules_tree1_modules_tree1_modules_conv2_parameters_weight_ + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_ = L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_ + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_ = L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_ + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_weight_ = L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_weight_ + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_bias_ = L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_bias_ + l_self_modules_level4_modules_tree1_modules_tree2_modules_conv1_parameters_weight_ = L_self_modules_level4_modules_tree1_modules_tree2_modules_conv1_parameters_weight_ + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_ = L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_ + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_ = L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_ + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_weight_ = L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_weight_ + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_bias_ = L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_bias_ + l_self_modules_level4_modules_tree1_modules_tree2_modules_conv2_parameters_weight_ = L_self_modules_level4_modules_tree1_modules_tree2_modules_conv2_parameters_weight_ + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_ = L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_ + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_ = L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_ + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_weight_ = L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_weight_ + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_bias_ = L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_bias_ + l_self_modules_level4_modules_tree1_modules_root_modules_conv_parameters_weight_ = L_self_modules_level4_modules_tree1_modules_root_modules_conv_parameters_weight_ + l_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_mean_ = L_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_mean_ + l_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_var_ = L_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_var_ + l_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_weight_ = L_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_weight_ + l_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_bias_ = ( + L_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_bias_ + ) + l_self_modules_level4_modules_tree2_modules_tree1_modules_conv1_parameters_weight_ = L_self_modules_level4_modules_tree2_modules_tree1_modules_conv1_parameters_weight_ + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_ = L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_ + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_ = L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_ + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_weight_ = L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_weight_ + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_bias_ = L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_bias_ + l_self_modules_level4_modules_tree2_modules_tree1_modules_conv2_parameters_weight_ = L_self_modules_level4_modules_tree2_modules_tree1_modules_conv2_parameters_weight_ + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_ = L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_ + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_ = L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_ + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_weight_ = L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_weight_ + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_bias_ = L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_bias_ + l_self_modules_level4_modules_tree2_modules_tree2_modules_conv1_parameters_weight_ = L_self_modules_level4_modules_tree2_modules_tree2_modules_conv1_parameters_weight_ + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_ = L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_ + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_ = L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_ + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_weight_ = L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_weight_ + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_bias_ = L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_bias_ + l_self_modules_level4_modules_tree2_modules_tree2_modules_conv2_parameters_weight_ = L_self_modules_level4_modules_tree2_modules_tree2_modules_conv2_parameters_weight_ + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_ = L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_ + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_ = L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_ + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_weight_ = L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_weight_ + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_bias_ = L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_bias_ + l_self_modules_level4_modules_tree2_modules_root_modules_conv_parameters_weight_ = L_self_modules_level4_modules_tree2_modules_root_modules_conv_parameters_weight_ + l_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_mean_ = L_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_mean_ + l_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_var_ = L_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_var_ + l_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_weight_ = L_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_weight_ + l_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_bias_ = ( + L_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_bias_ + ) + l_self_modules_level5_modules_project_modules_0_parameters_weight_ = ( + L_self_modules_level5_modules_project_modules_0_parameters_weight_ + ) + l_self_modules_level5_modules_project_modules_1_buffers_running_mean_ = ( + L_self_modules_level5_modules_project_modules_1_buffers_running_mean_ + ) + l_self_modules_level5_modules_project_modules_1_buffers_running_var_ = ( + L_self_modules_level5_modules_project_modules_1_buffers_running_var_ + ) + l_self_modules_level5_modules_project_modules_1_parameters_weight_ = ( + L_self_modules_level5_modules_project_modules_1_parameters_weight_ + ) + l_self_modules_level5_modules_project_modules_1_parameters_bias_ = ( + L_self_modules_level5_modules_project_modules_1_parameters_bias_ + ) + l_self_modules_level5_modules_tree1_modules_conv1_parameters_weight_ = ( + L_self_modules_level5_modules_tree1_modules_conv1_parameters_weight_ + ) + l_self_modules_level5_modules_tree1_modules_bn1_buffers_running_mean_ = ( + L_self_modules_level5_modules_tree1_modules_bn1_buffers_running_mean_ + ) + l_self_modules_level5_modules_tree1_modules_bn1_buffers_running_var_ = ( + L_self_modules_level5_modules_tree1_modules_bn1_buffers_running_var_ + ) + l_self_modules_level5_modules_tree1_modules_bn1_parameters_weight_ = ( + L_self_modules_level5_modules_tree1_modules_bn1_parameters_weight_ + ) + l_self_modules_level5_modules_tree1_modules_bn1_parameters_bias_ = ( + L_self_modules_level5_modules_tree1_modules_bn1_parameters_bias_ + ) + l_self_modules_level5_modules_tree1_modules_conv2_parameters_weight_ = ( + L_self_modules_level5_modules_tree1_modules_conv2_parameters_weight_ + ) + l_self_modules_level5_modules_tree1_modules_bn2_buffers_running_mean_ = ( + L_self_modules_level5_modules_tree1_modules_bn2_buffers_running_mean_ + ) + l_self_modules_level5_modules_tree1_modules_bn2_buffers_running_var_ = ( + L_self_modules_level5_modules_tree1_modules_bn2_buffers_running_var_ + ) + l_self_modules_level5_modules_tree1_modules_bn2_parameters_weight_ = ( + L_self_modules_level5_modules_tree1_modules_bn2_parameters_weight_ + ) + l_self_modules_level5_modules_tree1_modules_bn2_parameters_bias_ = ( + L_self_modules_level5_modules_tree1_modules_bn2_parameters_bias_ + ) + l_self_modules_level5_modules_tree2_modules_conv1_parameters_weight_ = ( + L_self_modules_level5_modules_tree2_modules_conv1_parameters_weight_ + ) + l_self_modules_level5_modules_tree2_modules_bn1_buffers_running_mean_ = ( + L_self_modules_level5_modules_tree2_modules_bn1_buffers_running_mean_ + ) + l_self_modules_level5_modules_tree2_modules_bn1_buffers_running_var_ = ( + L_self_modules_level5_modules_tree2_modules_bn1_buffers_running_var_ + ) + l_self_modules_level5_modules_tree2_modules_bn1_parameters_weight_ = ( + L_self_modules_level5_modules_tree2_modules_bn1_parameters_weight_ + ) + l_self_modules_level5_modules_tree2_modules_bn1_parameters_bias_ = ( + L_self_modules_level5_modules_tree2_modules_bn1_parameters_bias_ + ) + l_self_modules_level5_modules_tree2_modules_conv2_parameters_weight_ = ( + L_self_modules_level5_modules_tree2_modules_conv2_parameters_weight_ + ) + l_self_modules_level5_modules_tree2_modules_bn2_buffers_running_mean_ = ( + L_self_modules_level5_modules_tree2_modules_bn2_buffers_running_mean_ + ) + l_self_modules_level5_modules_tree2_modules_bn2_buffers_running_var_ = ( + L_self_modules_level5_modules_tree2_modules_bn2_buffers_running_var_ + ) + l_self_modules_level5_modules_tree2_modules_bn2_parameters_weight_ = ( + L_self_modules_level5_modules_tree2_modules_bn2_parameters_weight_ + ) + l_self_modules_level5_modules_tree2_modules_bn2_parameters_bias_ = ( + L_self_modules_level5_modules_tree2_modules_bn2_parameters_bias_ + ) + l_self_modules_level5_modules_root_modules_conv_parameters_weight_ = ( + L_self_modules_level5_modules_root_modules_conv_parameters_weight_ + ) + l_self_modules_level5_modules_root_modules_bn_buffers_running_mean_ = ( + L_self_modules_level5_modules_root_modules_bn_buffers_running_mean_ + ) + l_self_modules_level5_modules_root_modules_bn_buffers_running_var_ = ( + L_self_modules_level5_modules_root_modules_bn_buffers_running_var_ + ) + l_self_modules_level5_modules_root_modules_bn_parameters_weight_ = ( + L_self_modules_level5_modules_root_modules_bn_parameters_weight_ + ) + l_self_modules_level5_modules_root_modules_bn_parameters_bias_ = ( + L_self_modules_level5_modules_root_modules_bn_parameters_bias_ + ) + l_self_modules_fc_parameters_weight_ = L_self_modules_fc_parameters_weight_ + l_self_modules_fc_parameters_bias_ = L_self_modules_fc_parameters_bias_ + input_1 = torch.conv2d( + l_x_, + l_self_modules_base_layer_modules_0_parameters_weight_, + None, + (1, 1), + (3, 3), + (1, 1), + 1, + ) + l_x_ = l_self_modules_base_layer_modules_0_parameters_weight_ = None + input_2 = torch.nn.functional.batch_norm( + input_1, + l_self_modules_base_layer_modules_1_buffers_running_mean_, + l_self_modules_base_layer_modules_1_buffers_running_var_, + l_self_modules_base_layer_modules_1_parameters_weight_, + l_self_modules_base_layer_modules_1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_1 = ( + l_self_modules_base_layer_modules_1_buffers_running_mean_ + ) = ( + l_self_modules_base_layer_modules_1_buffers_running_var_ + ) = ( + l_self_modules_base_layer_modules_1_parameters_weight_ + ) = l_self_modules_base_layer_modules_1_parameters_bias_ = None + input_3 = torch.nn.functional.relu(input_2, inplace=True) + input_2 = None + input_4 = torch.conv2d( + input_3, + l_self_modules_level0_modules_0_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + input_3 = l_self_modules_level0_modules_0_parameters_weight_ = None + input_5 = torch.nn.functional.batch_norm( + input_4, + l_self_modules_level0_modules_1_buffers_running_mean_, + l_self_modules_level0_modules_1_buffers_running_var_, + l_self_modules_level0_modules_1_parameters_weight_, + l_self_modules_level0_modules_1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_4 = ( + l_self_modules_level0_modules_1_buffers_running_mean_ + ) = ( + l_self_modules_level0_modules_1_buffers_running_var_ + ) = ( + l_self_modules_level0_modules_1_parameters_weight_ + ) = l_self_modules_level0_modules_1_parameters_bias_ = None + input_6 = torch.nn.functional.relu(input_5, inplace=True) + input_5 = None + input_7 = torch.conv2d( + input_6, + l_self_modules_level1_modules_0_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_6 = l_self_modules_level1_modules_0_parameters_weight_ = None + input_8 = torch.nn.functional.batch_norm( + input_7, + l_self_modules_level1_modules_1_buffers_running_mean_, + l_self_modules_level1_modules_1_buffers_running_var_, + l_self_modules_level1_modules_1_parameters_weight_, + l_self_modules_level1_modules_1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_7 = ( + l_self_modules_level1_modules_1_buffers_running_mean_ + ) = ( + l_self_modules_level1_modules_1_buffers_running_var_ + ) = ( + l_self_modules_level1_modules_1_parameters_weight_ + ) = l_self_modules_level1_modules_1_parameters_bias_ = None + input_9 = torch.nn.functional.relu(input_8, inplace=True) + input_8 = None + bottom = torch.nn.functional.max_pool2d( + input_9, 2, 2, 0, 1, ceil_mode=False, return_indices=False + ) + input_10 = torch.conv2d( + bottom, + l_self_modules_level2_modules_project_modules_0_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + bottom = ( + l_self_modules_level2_modules_project_modules_0_parameters_weight_ + ) = None + input_11 = torch.nn.functional.batch_norm( + input_10, + l_self_modules_level2_modules_project_modules_1_buffers_running_mean_, + l_self_modules_level2_modules_project_modules_1_buffers_running_var_, + l_self_modules_level2_modules_project_modules_1_parameters_weight_, + l_self_modules_level2_modules_project_modules_1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_10 = ( + l_self_modules_level2_modules_project_modules_1_buffers_running_mean_ + ) = ( + l_self_modules_level2_modules_project_modules_1_buffers_running_var_ + ) = ( + l_self_modules_level2_modules_project_modules_1_parameters_weight_ + ) = l_self_modules_level2_modules_project_modules_1_parameters_bias_ = None + out = torch.conv2d( + input_9, + l_self_modules_level2_modules_tree1_modules_conv1_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + input_9 = ( + l_self_modules_level2_modules_tree1_modules_conv1_parameters_weight_ + ) = None + out_1 = torch.nn.functional.batch_norm( + out, + l_self_modules_level2_modules_tree1_modules_bn1_buffers_running_mean_, + l_self_modules_level2_modules_tree1_modules_bn1_buffers_running_var_, + l_self_modules_level2_modules_tree1_modules_bn1_parameters_weight_, + l_self_modules_level2_modules_tree1_modules_bn1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out = ( + l_self_modules_level2_modules_tree1_modules_bn1_buffers_running_mean_ + ) = ( + l_self_modules_level2_modules_tree1_modules_bn1_buffers_running_var_ + ) = ( + l_self_modules_level2_modules_tree1_modules_bn1_parameters_weight_ + ) = l_self_modules_level2_modules_tree1_modules_bn1_parameters_bias_ = None + out_2 = torch.nn.functional.relu(out_1, inplace=True) + out_1 = None + out_3 = torch.conv2d( + out_2, + l_self_modules_level2_modules_tree1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + out_2 = ( + l_self_modules_level2_modules_tree1_modules_conv2_parameters_weight_ + ) = None + out_4 = torch.nn.functional.batch_norm( + out_3, + l_self_modules_level2_modules_tree1_modules_bn2_buffers_running_mean_, + l_self_modules_level2_modules_tree1_modules_bn2_buffers_running_var_, + l_self_modules_level2_modules_tree1_modules_bn2_parameters_weight_, + l_self_modules_level2_modules_tree1_modules_bn2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_3 = ( + l_self_modules_level2_modules_tree1_modules_bn2_buffers_running_mean_ + ) = ( + l_self_modules_level2_modules_tree1_modules_bn2_buffers_running_var_ + ) = ( + l_self_modules_level2_modules_tree1_modules_bn2_parameters_weight_ + ) = l_self_modules_level2_modules_tree1_modules_bn2_parameters_bias_ = None + out_4 += input_11 + out_5 = out_4 + out_4 = input_11 = None + out_6 = torch.nn.functional.relu(out_5, inplace=True) + out_5 = None + out_7 = torch.conv2d( + out_6, + l_self_modules_level2_modules_tree2_modules_conv1_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_level2_modules_tree2_modules_conv1_parameters_weight_ = None + out_8 = torch.nn.functional.batch_norm( + out_7, + l_self_modules_level2_modules_tree2_modules_bn1_buffers_running_mean_, + l_self_modules_level2_modules_tree2_modules_bn1_buffers_running_var_, + l_self_modules_level2_modules_tree2_modules_bn1_parameters_weight_, + l_self_modules_level2_modules_tree2_modules_bn1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_7 = ( + l_self_modules_level2_modules_tree2_modules_bn1_buffers_running_mean_ + ) = ( + l_self_modules_level2_modules_tree2_modules_bn1_buffers_running_var_ + ) = ( + l_self_modules_level2_modules_tree2_modules_bn1_parameters_weight_ + ) = l_self_modules_level2_modules_tree2_modules_bn1_parameters_bias_ = None + out_9 = torch.nn.functional.relu(out_8, inplace=True) + out_8 = None + out_10 = torch.conv2d( + out_9, + l_self_modules_level2_modules_tree2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + out_9 = ( + l_self_modules_level2_modules_tree2_modules_conv2_parameters_weight_ + ) = None + out_11 = torch.nn.functional.batch_norm( + out_10, + l_self_modules_level2_modules_tree2_modules_bn2_buffers_running_mean_, + l_self_modules_level2_modules_tree2_modules_bn2_buffers_running_var_, + l_self_modules_level2_modules_tree2_modules_bn2_parameters_weight_, + l_self_modules_level2_modules_tree2_modules_bn2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_10 = ( + l_self_modules_level2_modules_tree2_modules_bn2_buffers_running_mean_ + ) = ( + l_self_modules_level2_modules_tree2_modules_bn2_buffers_running_var_ + ) = ( + l_self_modules_level2_modules_tree2_modules_bn2_parameters_weight_ + ) = l_self_modules_level2_modules_tree2_modules_bn2_parameters_bias_ = None + out_11 += out_6 + out_12 = out_11 + out_11 = None + out_13 = torch.nn.functional.relu(out_12, inplace=True) + out_12 = None + cat = torch.cat([out_13, out_6], 1) + out_13 = out_6 = None + x = torch.conv2d( + cat, + l_self_modules_level2_modules_root_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat = l_self_modules_level2_modules_root_modules_conv_parameters_weight_ = None + x_1 = torch.nn.functional.batch_norm( + x, + l_self_modules_level2_modules_root_modules_bn_buffers_running_mean_, + l_self_modules_level2_modules_root_modules_bn_buffers_running_var_, + l_self_modules_level2_modules_root_modules_bn_parameters_weight_, + l_self_modules_level2_modules_root_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x = ( + l_self_modules_level2_modules_root_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_level2_modules_root_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_level2_modules_root_modules_bn_parameters_weight_ + ) = l_self_modules_level2_modules_root_modules_bn_parameters_bias_ = None + x_2 = torch.nn.functional.relu(x_1, inplace=True) + x_1 = None + bottom_1 = torch.nn.functional.max_pool2d( + x_2, 2, 2, 0, 1, ceil_mode=False, return_indices=False + ) + bottom_2 = torch.nn.functional.max_pool2d( + x_2, 2, 2, 0, 1, ceil_mode=False, return_indices=False + ) + input_12 = torch.conv2d( + bottom_2, + l_self_modules_level3_modules_tree1_modules_project_modules_0_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + bottom_2 = l_self_modules_level3_modules_tree1_modules_project_modules_0_parameters_weight_ = (None) + input_13 = torch.nn.functional.batch_norm( + input_12, + l_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_mean_, + l_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_var_, + l_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_weight_, + l_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_12 = l_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_mean_ = l_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_var_ = l_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_weight_ = l_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_bias_ = (None) + out_14 = torch.conv2d( + x_2, + l_self_modules_level3_modules_tree1_modules_tree1_modules_conv1_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_2 = l_self_modules_level3_modules_tree1_modules_tree1_modules_conv1_parameters_weight_ = (None) + out_15 = torch.nn.functional.batch_norm( + out_14, + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_, + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_, + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_weight_, + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_14 = l_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_ = l_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_ = l_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_weight_ = l_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_bias_ = (None) + out_16 = torch.nn.functional.relu(out_15, inplace=True) + out_15 = None + out_17 = torch.conv2d( + out_16, + l_self_modules_level3_modules_tree1_modules_tree1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + out_16 = l_self_modules_level3_modules_tree1_modules_tree1_modules_conv2_parameters_weight_ = (None) + out_18 = torch.nn.functional.batch_norm( + out_17, + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_, + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_, + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_weight_, + l_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_17 = l_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_ = l_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_ = l_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_weight_ = l_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_bias_ = (None) + out_18 += input_13 + out_19 = out_18 + out_18 = input_13 = None + out_20 = torch.nn.functional.relu(out_19, inplace=True) + out_19 = None + out_21 = torch.conv2d( + out_20, + l_self_modules_level3_modules_tree1_modules_tree2_modules_conv1_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_level3_modules_tree1_modules_tree2_modules_conv1_parameters_weight_ = ( + None + ) + out_22 = torch.nn.functional.batch_norm( + out_21, + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_, + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_, + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_weight_, + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_21 = l_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_ = l_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_ = l_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_weight_ = l_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_bias_ = (None) + out_23 = torch.nn.functional.relu(out_22, inplace=True) + out_22 = None + out_24 = torch.conv2d( + out_23, + l_self_modules_level3_modules_tree1_modules_tree2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + out_23 = l_self_modules_level3_modules_tree1_modules_tree2_modules_conv2_parameters_weight_ = (None) + out_25 = torch.nn.functional.batch_norm( + out_24, + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_, + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_, + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_weight_, + l_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_24 = l_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_ = l_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_ = l_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_weight_ = l_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_bias_ = (None) + out_25 += out_20 + out_26 = out_25 + out_25 = None + out_27 = torch.nn.functional.relu(out_26, inplace=True) + out_26 = None + cat_1 = torch.cat([out_27, out_20], 1) + out_27 = out_20 = None + x_3 = torch.conv2d( + cat_1, + l_self_modules_level3_modules_tree1_modules_root_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_1 = l_self_modules_level3_modules_tree1_modules_root_modules_conv_parameters_weight_ = (None) + x_4 = torch.nn.functional.batch_norm( + x_3, + l_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_mean_, + l_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_var_, + l_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_weight_, + l_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_3 = l_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_mean_ = l_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_var_ = l_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_weight_ = ( + l_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_bias_ + ) = None + x_5 = torch.nn.functional.relu(x_4, inplace=True) + x_4 = None + out_28 = torch.conv2d( + x_5, + l_self_modules_level3_modules_tree2_modules_tree1_modules_conv1_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_level3_modules_tree2_modules_tree1_modules_conv1_parameters_weight_ = ( + None + ) + out_29 = torch.nn.functional.batch_norm( + out_28, + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_, + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_, + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_weight_, + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_28 = l_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_ = l_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_ = l_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_weight_ = l_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_bias_ = (None) + out_30 = torch.nn.functional.relu(out_29, inplace=True) + out_29 = None + out_31 = torch.conv2d( + out_30, + l_self_modules_level3_modules_tree2_modules_tree1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + out_30 = l_self_modules_level3_modules_tree2_modules_tree1_modules_conv2_parameters_weight_ = (None) + out_32 = torch.nn.functional.batch_norm( + out_31, + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_, + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_, + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_weight_, + l_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_31 = l_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_ = l_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_ = l_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_weight_ = l_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_bias_ = (None) + out_32 += x_5 + out_33 = out_32 + out_32 = None + out_34 = torch.nn.functional.relu(out_33, inplace=True) + out_33 = None + out_35 = torch.conv2d( + out_34, + l_self_modules_level3_modules_tree2_modules_tree2_modules_conv1_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_level3_modules_tree2_modules_tree2_modules_conv1_parameters_weight_ = ( + None + ) + out_36 = torch.nn.functional.batch_norm( + out_35, + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_, + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_, + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_weight_, + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_35 = l_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_ = l_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_ = l_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_weight_ = l_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_bias_ = (None) + out_37 = torch.nn.functional.relu(out_36, inplace=True) + out_36 = None + out_38 = torch.conv2d( + out_37, + l_self_modules_level3_modules_tree2_modules_tree2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + out_37 = l_self_modules_level3_modules_tree2_modules_tree2_modules_conv2_parameters_weight_ = (None) + out_39 = torch.nn.functional.batch_norm( + out_38, + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_, + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_, + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_weight_, + l_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_38 = l_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_ = l_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_ = l_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_weight_ = l_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_bias_ = (None) + out_39 += out_34 + out_40 = out_39 + out_39 = None + out_41 = torch.nn.functional.relu(out_40, inplace=True) + out_40 = None + cat_2 = torch.cat([out_41, out_34, bottom_1, x_5], 1) + out_41 = out_34 = bottom_1 = x_5 = None + x_6 = torch.conv2d( + cat_2, + l_self_modules_level3_modules_tree2_modules_root_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_2 = l_self_modules_level3_modules_tree2_modules_root_modules_conv_parameters_weight_ = (None) + x_7 = torch.nn.functional.batch_norm( + x_6, + l_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_mean_, + l_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_var_, + l_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_weight_, + l_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_6 = l_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_mean_ = l_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_var_ = l_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_weight_ = ( + l_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_bias_ + ) = None + x_8 = torch.nn.functional.relu(x_7, inplace=True) + x_7 = None + bottom_3 = torch.nn.functional.max_pool2d( + x_8, 2, 2, 0, 1, ceil_mode=False, return_indices=False + ) + bottom_4 = torch.nn.functional.max_pool2d( + x_8, 2, 2, 0, 1, ceil_mode=False, return_indices=False + ) + input_14 = torch.conv2d( + bottom_4, + l_self_modules_level4_modules_tree1_modules_project_modules_0_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + bottom_4 = l_self_modules_level4_modules_tree1_modules_project_modules_0_parameters_weight_ = (None) + input_15 = torch.nn.functional.batch_norm( + input_14, + l_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_mean_, + l_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_var_, + l_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_weight_, + l_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_14 = l_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_mean_ = l_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_var_ = l_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_weight_ = l_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_bias_ = (None) + out_42 = torch.conv2d( + x_8, + l_self_modules_level4_modules_tree1_modules_tree1_modules_conv1_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_8 = l_self_modules_level4_modules_tree1_modules_tree1_modules_conv1_parameters_weight_ = (None) + out_43 = torch.nn.functional.batch_norm( + out_42, + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_, + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_, + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_weight_, + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_42 = l_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_ = l_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_ = l_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_weight_ = l_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_bias_ = (None) + out_44 = torch.nn.functional.relu(out_43, inplace=True) + out_43 = None + out_45 = torch.conv2d( + out_44, + l_self_modules_level4_modules_tree1_modules_tree1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + out_44 = l_self_modules_level4_modules_tree1_modules_tree1_modules_conv2_parameters_weight_ = (None) + out_46 = torch.nn.functional.batch_norm( + out_45, + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_, + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_, + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_weight_, + l_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_45 = l_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_ = l_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_ = l_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_weight_ = l_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_bias_ = (None) + out_46 += input_15 + out_47 = out_46 + out_46 = input_15 = None + out_48 = torch.nn.functional.relu(out_47, inplace=True) + out_47 = None + out_49 = torch.conv2d( + out_48, + l_self_modules_level4_modules_tree1_modules_tree2_modules_conv1_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_level4_modules_tree1_modules_tree2_modules_conv1_parameters_weight_ = ( + None + ) + out_50 = torch.nn.functional.batch_norm( + out_49, + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_, + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_, + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_weight_, + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_49 = l_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_ = l_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_ = l_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_weight_ = l_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_bias_ = (None) + out_51 = torch.nn.functional.relu(out_50, inplace=True) + out_50 = None + out_52 = torch.conv2d( + out_51, + l_self_modules_level4_modules_tree1_modules_tree2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + out_51 = l_self_modules_level4_modules_tree1_modules_tree2_modules_conv2_parameters_weight_ = (None) + out_53 = torch.nn.functional.batch_norm( + out_52, + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_, + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_, + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_weight_, + l_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_52 = l_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_ = l_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_ = l_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_weight_ = l_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_bias_ = (None) + out_53 += out_48 + out_54 = out_53 + out_53 = None + out_55 = torch.nn.functional.relu(out_54, inplace=True) + out_54 = None + cat_3 = torch.cat([out_55, out_48], 1) + out_55 = out_48 = None + x_9 = torch.conv2d( + cat_3, + l_self_modules_level4_modules_tree1_modules_root_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_3 = l_self_modules_level4_modules_tree1_modules_root_modules_conv_parameters_weight_ = (None) + x_10 = torch.nn.functional.batch_norm( + x_9, + l_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_mean_, + l_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_var_, + l_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_weight_, + l_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_9 = l_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_mean_ = l_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_var_ = l_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_weight_ = ( + l_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_bias_ + ) = None + x_11 = torch.nn.functional.relu(x_10, inplace=True) + x_10 = None + out_56 = torch.conv2d( + x_11, + l_self_modules_level4_modules_tree2_modules_tree1_modules_conv1_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_level4_modules_tree2_modules_tree1_modules_conv1_parameters_weight_ = ( + None + ) + out_57 = torch.nn.functional.batch_norm( + out_56, + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_, + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_, + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_weight_, + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_56 = l_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_ = l_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_ = l_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_weight_ = l_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_bias_ = (None) + out_58 = torch.nn.functional.relu(out_57, inplace=True) + out_57 = None + out_59 = torch.conv2d( + out_58, + l_self_modules_level4_modules_tree2_modules_tree1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + out_58 = l_self_modules_level4_modules_tree2_modules_tree1_modules_conv2_parameters_weight_ = (None) + out_60 = torch.nn.functional.batch_norm( + out_59, + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_, + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_, + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_weight_, + l_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_59 = l_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_ = l_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_ = l_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_weight_ = l_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_bias_ = (None) + out_60 += x_11 + out_61 = out_60 + out_60 = None + out_62 = torch.nn.functional.relu(out_61, inplace=True) + out_61 = None + out_63 = torch.conv2d( + out_62, + l_self_modules_level4_modules_tree2_modules_tree2_modules_conv1_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_level4_modules_tree2_modules_tree2_modules_conv1_parameters_weight_ = ( + None + ) + out_64 = torch.nn.functional.batch_norm( + out_63, + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_, + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_, + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_weight_, + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_63 = l_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_ = l_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_ = l_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_weight_ = l_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_bias_ = (None) + out_65 = torch.nn.functional.relu(out_64, inplace=True) + out_64 = None + out_66 = torch.conv2d( + out_65, + l_self_modules_level4_modules_tree2_modules_tree2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + out_65 = l_self_modules_level4_modules_tree2_modules_tree2_modules_conv2_parameters_weight_ = (None) + out_67 = torch.nn.functional.batch_norm( + out_66, + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_, + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_, + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_weight_, + l_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_66 = l_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_ = l_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_ = l_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_weight_ = l_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_bias_ = (None) + out_67 += out_62 + out_68 = out_67 + out_67 = None + out_69 = torch.nn.functional.relu(out_68, inplace=True) + out_68 = None + cat_4 = torch.cat([out_69, out_62, bottom_3, x_11], 1) + out_69 = out_62 = bottom_3 = x_11 = None + x_12 = torch.conv2d( + cat_4, + l_self_modules_level4_modules_tree2_modules_root_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_4 = l_self_modules_level4_modules_tree2_modules_root_modules_conv_parameters_weight_ = (None) + x_13 = torch.nn.functional.batch_norm( + x_12, + l_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_mean_, + l_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_var_, + l_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_weight_, + l_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_12 = l_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_mean_ = l_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_var_ = l_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_weight_ = ( + l_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_bias_ + ) = None + x_14 = torch.nn.functional.relu(x_13, inplace=True) + x_13 = None + bottom_5 = torch.nn.functional.max_pool2d( + x_14, 2, 2, 0, 1, ceil_mode=False, return_indices=False + ) + input_16 = torch.conv2d( + bottom_5, + l_self_modules_level5_modules_project_modules_0_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + l_self_modules_level5_modules_project_modules_0_parameters_weight_ = None + input_17 = torch.nn.functional.batch_norm( + input_16, + l_self_modules_level5_modules_project_modules_1_buffers_running_mean_, + l_self_modules_level5_modules_project_modules_1_buffers_running_var_, + l_self_modules_level5_modules_project_modules_1_parameters_weight_, + l_self_modules_level5_modules_project_modules_1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + input_16 = ( + l_self_modules_level5_modules_project_modules_1_buffers_running_mean_ + ) = ( + l_self_modules_level5_modules_project_modules_1_buffers_running_var_ + ) = ( + l_self_modules_level5_modules_project_modules_1_parameters_weight_ + ) = l_self_modules_level5_modules_project_modules_1_parameters_bias_ = None + out_70 = torch.conv2d( + x_14, + l_self_modules_level5_modules_tree1_modules_conv1_parameters_weight_, + None, + (2, 2), + (1, 1), + (1, 1), + 1, + ) + x_14 = ( + l_self_modules_level5_modules_tree1_modules_conv1_parameters_weight_ + ) = None + out_71 = torch.nn.functional.batch_norm( + out_70, + l_self_modules_level5_modules_tree1_modules_bn1_buffers_running_mean_, + l_self_modules_level5_modules_tree1_modules_bn1_buffers_running_var_, + l_self_modules_level5_modules_tree1_modules_bn1_parameters_weight_, + l_self_modules_level5_modules_tree1_modules_bn1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_70 = ( + l_self_modules_level5_modules_tree1_modules_bn1_buffers_running_mean_ + ) = ( + l_self_modules_level5_modules_tree1_modules_bn1_buffers_running_var_ + ) = ( + l_self_modules_level5_modules_tree1_modules_bn1_parameters_weight_ + ) = l_self_modules_level5_modules_tree1_modules_bn1_parameters_bias_ = None + out_72 = torch.nn.functional.relu(out_71, inplace=True) + out_71 = None + out_73 = torch.conv2d( + out_72, + l_self_modules_level5_modules_tree1_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + out_72 = ( + l_self_modules_level5_modules_tree1_modules_conv2_parameters_weight_ + ) = None + out_74 = torch.nn.functional.batch_norm( + out_73, + l_self_modules_level5_modules_tree1_modules_bn2_buffers_running_mean_, + l_self_modules_level5_modules_tree1_modules_bn2_buffers_running_var_, + l_self_modules_level5_modules_tree1_modules_bn2_parameters_weight_, + l_self_modules_level5_modules_tree1_modules_bn2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_73 = ( + l_self_modules_level5_modules_tree1_modules_bn2_buffers_running_mean_ + ) = ( + l_self_modules_level5_modules_tree1_modules_bn2_buffers_running_var_ + ) = ( + l_self_modules_level5_modules_tree1_modules_bn2_parameters_weight_ + ) = l_self_modules_level5_modules_tree1_modules_bn2_parameters_bias_ = None + out_74 += input_17 + out_75 = out_74 + out_74 = input_17 = None + out_76 = torch.nn.functional.relu(out_75, inplace=True) + out_75 = None + out_77 = torch.conv2d( + out_76, + l_self_modules_level5_modules_tree2_modules_conv1_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + l_self_modules_level5_modules_tree2_modules_conv1_parameters_weight_ = None + out_78 = torch.nn.functional.batch_norm( + out_77, + l_self_modules_level5_modules_tree2_modules_bn1_buffers_running_mean_, + l_self_modules_level5_modules_tree2_modules_bn1_buffers_running_var_, + l_self_modules_level5_modules_tree2_modules_bn1_parameters_weight_, + l_self_modules_level5_modules_tree2_modules_bn1_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_77 = ( + l_self_modules_level5_modules_tree2_modules_bn1_buffers_running_mean_ + ) = ( + l_self_modules_level5_modules_tree2_modules_bn1_buffers_running_var_ + ) = ( + l_self_modules_level5_modules_tree2_modules_bn1_parameters_weight_ + ) = l_self_modules_level5_modules_tree2_modules_bn1_parameters_bias_ = None + out_79 = torch.nn.functional.relu(out_78, inplace=True) + out_78 = None + out_80 = torch.conv2d( + out_79, + l_self_modules_level5_modules_tree2_modules_conv2_parameters_weight_, + None, + (1, 1), + (1, 1), + (1, 1), + 1, + ) + out_79 = ( + l_self_modules_level5_modules_tree2_modules_conv2_parameters_weight_ + ) = None + out_81 = torch.nn.functional.batch_norm( + out_80, + l_self_modules_level5_modules_tree2_modules_bn2_buffers_running_mean_, + l_self_modules_level5_modules_tree2_modules_bn2_buffers_running_var_, + l_self_modules_level5_modules_tree2_modules_bn2_parameters_weight_, + l_self_modules_level5_modules_tree2_modules_bn2_parameters_bias_, + False, + 0.1, + 1e-05, + ) + out_80 = ( + l_self_modules_level5_modules_tree2_modules_bn2_buffers_running_mean_ + ) = ( + l_self_modules_level5_modules_tree2_modules_bn2_buffers_running_var_ + ) = ( + l_self_modules_level5_modules_tree2_modules_bn2_parameters_weight_ + ) = l_self_modules_level5_modules_tree2_modules_bn2_parameters_bias_ = None + out_81 += out_76 + out_82 = out_81 + out_81 = None + out_83 = torch.nn.functional.relu(out_82, inplace=True) + out_82 = None + cat_5 = torch.cat([out_83, out_76, bottom_5], 1) + out_83 = out_76 = bottom_5 = None + x_15 = torch.conv2d( + cat_5, + l_self_modules_level5_modules_root_modules_conv_parameters_weight_, + None, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + cat_5 = ( + l_self_modules_level5_modules_root_modules_conv_parameters_weight_ + ) = None + x_16 = torch.nn.functional.batch_norm( + x_15, + l_self_modules_level5_modules_root_modules_bn_buffers_running_mean_, + l_self_modules_level5_modules_root_modules_bn_buffers_running_var_, + l_self_modules_level5_modules_root_modules_bn_parameters_weight_, + l_self_modules_level5_modules_root_modules_bn_parameters_bias_, + False, + 0.1, + 1e-05, + ) + x_15 = ( + l_self_modules_level5_modules_root_modules_bn_buffers_running_mean_ + ) = ( + l_self_modules_level5_modules_root_modules_bn_buffers_running_var_ + ) = ( + l_self_modules_level5_modules_root_modules_bn_parameters_weight_ + ) = l_self_modules_level5_modules_root_modules_bn_parameters_bias_ = None + x_17 = torch.nn.functional.relu(x_16, inplace=True) + x_16 = None + x_18 = torch.nn.functional.adaptive_avg_pool2d(x_17, 1) + x_17 = None + x_19 = torch.nn.functional.dropout(x_18, 0.0, False, False) + x_18 = None + x_20 = torch.conv2d( + x_19, + l_self_modules_fc_parameters_weight_, + l_self_modules_fc_parameters_bias_, + (1, 1), + (0, 0), + (1, 1), + 1, + ) + x_19 = ( + l_self_modules_fc_parameters_weight_ + ) = l_self_modules_fc_parameters_bias_ = None + x_21 = x_20.flatten(1, -1) + x_20 = None + return (x_21,) diff --git a/samples/timm/dla34.in1k/weight_meta.py b/samples/timm/dla34.in1k/weight_meta.py new file mode 100644 index 0000000000..5c0e8c3304 --- /dev/null +++ b/samples/timm/dla34.in1k/weight_meta.py @@ -0,0 +1,1940 @@ +class Program_weight_tensor_meta_L_self_modules_base_layer_modules_0_parameters_weight_: + name = "L_self_modules_base_layer_modules_0_parameters_weight_" + shape = [16, 3, 7, 7] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.052 + data = None + + +class Program_weight_tensor_meta_s1: + name = "s1" + shape = [] + dtype = "torch.int64" + device = "cpu" + mean = None + std = None + data = [4] + + +class Program_weight_tensor_meta_L_x_: + name = "L_x_" + shape = [1, 3, 224, 224] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.230 + std = 1.289 + data = None + + +class Program_weight_tensor_meta_L_self_modules_base_layer_modules_1_buffers_running_mean_: + name = "L_self_modules_base_layer_modules_1_buffers_running_mean_" + shape = [16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_base_layer_modules_1_buffers_running_var_: + name = "L_self_modules_base_layer_modules_1_buffers_running_var_" + shape = [16] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_base_layer_modules_1_parameters_weight_: + name = "L_self_modules_base_layer_modules_1_parameters_weight_" + shape = [16] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_base_layer_modules_1_parameters_bias_: + name = "L_self_modules_base_layer_modules_1_parameters_bias_" + shape = [16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level0_modules_0_parameters_weight_: + name = "L_self_modules_level0_modules_0_parameters_weight_" + shape = [16, 16, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.117 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level0_modules_1_buffers_running_mean_: + name = "L_self_modules_level0_modules_1_buffers_running_mean_" + shape = [16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level0_modules_1_buffers_running_var_: + name = "L_self_modules_level0_modules_1_buffers_running_var_" + shape = [16] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level0_modules_1_parameters_weight_: + name = "L_self_modules_level0_modules_1_parameters_weight_" + shape = [16] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level0_modules_1_parameters_bias_: + name = "L_self_modules_level0_modules_1_parameters_bias_" + shape = [16] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level1_modules_0_parameters_weight_: + name = "L_self_modules_level1_modules_0_parameters_weight_" + shape = [32, 16, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.083 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level1_modules_1_buffers_running_mean_: + name = "L_self_modules_level1_modules_1_buffers_running_mean_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level1_modules_1_buffers_running_var_: + name = "L_self_modules_level1_modules_1_buffers_running_var_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level1_modules_1_parameters_weight_: + name = "L_self_modules_level1_modules_1_parameters_weight_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level1_modules_1_parameters_bias_: + name = "L_self_modules_level1_modules_1_parameters_bias_" + shape = [32] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_project_modules_0_parameters_weight_: + name = "L_self_modules_level2_modules_project_modules_0_parameters_weight_" + shape = [64, 32, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.001 + std = 0.179 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_project_modules_1_buffers_running_mean_: + name = "L_self_modules_level2_modules_project_modules_1_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_project_modules_1_buffers_running_var_: + name = "L_self_modules_level2_modules_project_modules_1_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_project_modules_1_parameters_weight_: + name = "L_self_modules_level2_modules_project_modules_1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_project_modules_1_parameters_bias_: + name = "L_self_modules_level2_modules_project_modules_1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree1_modules_conv1_parameters_weight_: + name = "L_self_modules_level2_modules_tree1_modules_conv1_parameters_weight_" + shape = [64, 32, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.058 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree1_modules_bn1_buffers_running_mean_: + name = "L_self_modules_level2_modules_tree1_modules_bn1_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree1_modules_bn1_buffers_running_var_: + name = "L_self_modules_level2_modules_tree1_modules_bn1_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree1_modules_bn1_parameters_weight_: + name = "L_self_modules_level2_modules_tree1_modules_bn1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree1_modules_bn1_parameters_bias_: + name = "L_self_modules_level2_modules_tree1_modules_bn1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree1_modules_conv2_parameters_weight_: + name = "L_self_modules_level2_modules_tree1_modules_conv2_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.001 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree1_modules_bn2_buffers_running_mean_: + name = "L_self_modules_level2_modules_tree1_modules_bn2_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree1_modules_bn2_buffers_running_var_: + name = "L_self_modules_level2_modules_tree1_modules_bn2_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree1_modules_bn2_parameters_weight_: + name = "L_self_modules_level2_modules_tree1_modules_bn2_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree1_modules_bn2_parameters_bias_: + name = "L_self_modules_level2_modules_tree1_modules_bn2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree2_modules_conv1_parameters_weight_: + name = "L_self_modules_level2_modules_tree2_modules_conv1_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree2_modules_bn1_buffers_running_mean_: + name = "L_self_modules_level2_modules_tree2_modules_bn1_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree2_modules_bn1_buffers_running_var_: + name = "L_self_modules_level2_modules_tree2_modules_bn1_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree2_modules_bn1_parameters_weight_: + name = "L_self_modules_level2_modules_tree2_modules_bn1_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree2_modules_bn1_parameters_bias_: + name = "L_self_modules_level2_modules_tree2_modules_bn1_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree2_modules_conv2_parameters_weight_: + name = "L_self_modules_level2_modules_tree2_modules_conv2_parameters_weight_" + shape = [64, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.059 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree2_modules_bn2_buffers_running_mean_: + name = "L_self_modules_level2_modules_tree2_modules_bn2_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree2_modules_bn2_buffers_running_var_: + name = "L_self_modules_level2_modules_tree2_modules_bn2_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree2_modules_bn2_parameters_weight_: + name = "L_self_modules_level2_modules_tree2_modules_bn2_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_tree2_modules_bn2_parameters_bias_: + name = "L_self_modules_level2_modules_tree2_modules_bn2_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_root_modules_conv_parameters_weight_: + name = "L_self_modules_level2_modules_root_modules_conv_parameters_weight_" + shape = [64, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.177 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_root_modules_bn_buffers_running_mean_: + name = "L_self_modules_level2_modules_root_modules_bn_buffers_running_mean_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_root_modules_bn_buffers_running_var_: + name = "L_self_modules_level2_modules_root_modules_bn_buffers_running_var_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_root_modules_bn_parameters_weight_: + name = "L_self_modules_level2_modules_root_modules_bn_parameters_weight_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level2_modules_root_modules_bn_parameters_bias_: + name = "L_self_modules_level2_modules_root_modules_bn_parameters_bias_" + shape = [64] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_project_modules_0_parameters_weight_: + name = "L_self_modules_level3_modules_tree1_modules_project_modules_0_parameters_weight_" + shape = [128, 64, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.003 + std = 0.124 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_mean_: + name = "L_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_var_: + name = "L_self_modules_level3_modules_tree1_modules_project_modules_1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_weight_: + name = "L_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_bias_: + name = ( + "L_self_modules_level3_modules_tree1_modules_project_modules_1_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree1_modules_conv1_parameters_weight_: + name = "L_self_modules_level3_modules_tree1_modules_tree1_modules_conv1_parameters_weight_" + shape = [128, 64, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_: + name = "L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_: + name = "L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_weight_: + name = "L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_bias_: + name = ( + "L_self_modules_level3_modules_tree1_modules_tree1_modules_bn1_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree1_modules_conv2_parameters_weight_: + name = "L_self_modules_level3_modules_tree1_modules_tree1_modules_conv2_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_: + name = "L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_: + name = "L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_weight_: + name = "L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_bias_: + name = ( + "L_self_modules_level3_modules_tree1_modules_tree1_modules_bn2_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree2_modules_conv1_parameters_weight_: + name = "L_self_modules_level3_modules_tree1_modules_tree2_modules_conv1_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_: + name = "L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_: + name = "L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_weight_: + name = "L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_bias_: + name = ( + "L_self_modules_level3_modules_tree1_modules_tree2_modules_bn1_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree2_modules_conv2_parameters_weight_: + name = "L_self_modules_level3_modules_tree1_modules_tree2_modules_conv2_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_: + name = "L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_: + name = "L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_weight_: + name = "L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_bias_: + name = ( + "L_self_modules_level3_modules_tree1_modules_tree2_modules_bn2_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_root_modules_conv_parameters_weight_: + name = "L_self_modules_level3_modules_tree1_modules_root_modules_conv_parameters_weight_" + shape = [128, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.126 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_mean_: + name = "L_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_var_: + name = "L_self_modules_level3_modules_tree1_modules_root_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_weight_: + name = ( + "L_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_weight_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_bias_: + name = ( + "L_self_modules_level3_modules_tree1_modules_root_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree1_modules_conv1_parameters_weight_: + name = "L_self_modules_level3_modules_tree2_modules_tree1_modules_conv1_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_: + name = "L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_: + name = "L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_weight_: + name = "L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_bias_: + name = ( + "L_self_modules_level3_modules_tree2_modules_tree1_modules_bn1_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree1_modules_conv2_parameters_weight_: + name = "L_self_modules_level3_modules_tree2_modules_tree1_modules_conv2_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_: + name = "L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_: + name = "L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_weight_: + name = "L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_bias_: + name = ( + "L_self_modules_level3_modules_tree2_modules_tree1_modules_bn2_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree2_modules_conv1_parameters_weight_: + name = "L_self_modules_level3_modules_tree2_modules_tree2_modules_conv1_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_: + name = "L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_: + name = "L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_weight_: + name = "L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_bias_: + name = ( + "L_self_modules_level3_modules_tree2_modules_tree2_modules_bn1_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree2_modules_conv2_parameters_weight_: + name = "L_self_modules_level3_modules_tree2_modules_tree2_modules_conv2_parameters_weight_" + shape = [128, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.042 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_: + name = "L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_: + name = "L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_weight_: + name = "L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_weight_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_bias_: + name = ( + "L_self_modules_level3_modules_tree2_modules_tree2_modules_bn2_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_root_modules_conv_parameters_weight_: + name = "L_self_modules_level3_modules_tree2_modules_root_modules_conv_parameters_weight_" + shape = [128, 448, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.125 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_mean_: + name = "L_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_mean_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_var_: + name = "L_self_modules_level3_modules_tree2_modules_root_modules_bn_buffers_running_var_" + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_weight_: + name = ( + "L_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_weight_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_bias_: + name = ( + "L_self_modules_level3_modules_tree2_modules_root_modules_bn_parameters_bias_" + ) + shape = [128] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_project_modules_0_parameters_weight_: + name = "L_self_modules_level4_modules_tree1_modules_project_modules_0_parameters_weight_" + shape = [256, 128, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_mean_: + name = "L_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_var_: + name = "L_self_modules_level4_modules_tree1_modules_project_modules_1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_weight_: + name = "L_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_bias_: + name = ( + "L_self_modules_level4_modules_tree1_modules_project_modules_1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree1_modules_conv1_parameters_weight_: + name = "L_self_modules_level4_modules_tree1_modules_tree1_modules_conv1_parameters_weight_" + shape = [256, 128, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_: + name = "L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_: + name = "L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_weight_: + name = "L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_bias_: + name = ( + "L_self_modules_level4_modules_tree1_modules_tree1_modules_bn1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree1_modules_conv2_parameters_weight_: + name = "L_self_modules_level4_modules_tree1_modules_tree1_modules_conv2_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_: + name = "L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_: + name = "L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_weight_: + name = "L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_bias_: + name = ( + "L_self_modules_level4_modules_tree1_modules_tree1_modules_bn2_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree2_modules_conv1_parameters_weight_: + name = "L_self_modules_level4_modules_tree1_modules_tree2_modules_conv1_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_: + name = "L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_: + name = "L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_weight_: + name = "L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_bias_: + name = ( + "L_self_modules_level4_modules_tree1_modules_tree2_modules_bn1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree2_modules_conv2_parameters_weight_: + name = "L_self_modules_level4_modules_tree1_modules_tree2_modules_conv2_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_: + name = "L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_: + name = "L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_weight_: + name = "L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_bias_: + name = ( + "L_self_modules_level4_modules_tree1_modules_tree2_modules_bn2_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_root_modules_conv_parameters_weight_: + name = "L_self_modules_level4_modules_tree1_modules_root_modules_conv_parameters_weight_" + shape = [256, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_mean_: + name = "L_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_var_: + name = "L_self_modules_level4_modules_tree1_modules_root_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_weight_: + name = ( + "L_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_weight_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_bias_: + name = ( + "L_self_modules_level4_modules_tree1_modules_root_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree1_modules_conv1_parameters_weight_: + name = "L_self_modules_level4_modules_tree2_modules_tree1_modules_conv1_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_: + name = "L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_: + name = "L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_weight_: + name = "L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_bias_: + name = ( + "L_self_modules_level4_modules_tree2_modules_tree1_modules_bn1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree1_modules_conv2_parameters_weight_: + name = "L_self_modules_level4_modules_tree2_modules_tree1_modules_conv2_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_: + name = "L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_: + name = "L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_weight_: + name = "L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_bias_: + name = ( + "L_self_modules_level4_modules_tree2_modules_tree1_modules_bn2_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree2_modules_conv1_parameters_weight_: + name = "L_self_modules_level4_modules_tree2_modules_tree2_modules_conv1_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.030 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_: + name = "L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_: + name = "L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_weight_: + name = "L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_bias_: + name = ( + "L_self_modules_level4_modules_tree2_modules_tree2_modules_bn1_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree2_modules_conv2_parameters_weight_: + name = "L_self_modules_level4_modules_tree2_modules_tree2_modules_conv2_parameters_weight_" + shape = [256, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.029 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_: + name = "L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_: + name = "L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_weight_: + name = "L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_weight_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_bias_: + name = ( + "L_self_modules_level4_modules_tree2_modules_tree2_modules_bn2_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_root_modules_conv_parameters_weight_: + name = "L_self_modules_level4_modules_tree2_modules_root_modules_conv_parameters_weight_" + shape = [256, 896, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.088 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_mean_: + name = "L_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_mean_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_var_: + name = "L_self_modules_level4_modules_tree2_modules_root_modules_bn_buffers_running_var_" + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_weight_: + name = ( + "L_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_weight_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_bias_: + name = ( + "L_self_modules_level4_modules_tree2_modules_root_modules_bn_parameters_bias_" + ) + shape = [256] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_project_modules_0_parameters_weight_: + name = "L_self_modules_level5_modules_project_modules_0_parameters_weight_" + shape = [512, 256, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.062 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_project_modules_1_buffers_running_mean_: + name = "L_self_modules_level5_modules_project_modules_1_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_project_modules_1_buffers_running_var_: + name = "L_self_modules_level5_modules_project_modules_1_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_project_modules_1_parameters_weight_: + name = "L_self_modules_level5_modules_project_modules_1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_project_modules_1_parameters_bias_: + name = "L_self_modules_level5_modules_project_modules_1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree1_modules_conv1_parameters_weight_: + name = "L_self_modules_level5_modules_tree1_modules_conv1_parameters_weight_" + shape = [512, 256, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree1_modules_bn1_buffers_running_mean_: + name = "L_self_modules_level5_modules_tree1_modules_bn1_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree1_modules_bn1_buffers_running_var_: + name = "L_self_modules_level5_modules_tree1_modules_bn1_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree1_modules_bn1_parameters_weight_: + name = "L_self_modules_level5_modules_tree1_modules_bn1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree1_modules_bn1_parameters_bias_: + name = "L_self_modules_level5_modules_tree1_modules_bn1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree1_modules_conv2_parameters_weight_: + name = "L_self_modules_level5_modules_tree1_modules_conv2_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree1_modules_bn2_buffers_running_mean_: + name = "L_self_modules_level5_modules_tree1_modules_bn2_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree1_modules_bn2_buffers_running_var_: + name = "L_self_modules_level5_modules_tree1_modules_bn2_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree1_modules_bn2_parameters_weight_: + name = "L_self_modules_level5_modules_tree1_modules_bn2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree1_modules_bn2_parameters_bias_: + name = "L_self_modules_level5_modules_tree1_modules_bn2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree2_modules_conv1_parameters_weight_: + name = "L_self_modules_level5_modules_tree2_modules_conv1_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = -0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree2_modules_bn1_buffers_running_mean_: + name = "L_self_modules_level5_modules_tree2_modules_bn1_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree2_modules_bn1_buffers_running_var_: + name = "L_self_modules_level5_modules_tree2_modules_bn1_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree2_modules_bn1_parameters_weight_: + name = "L_self_modules_level5_modules_tree2_modules_bn1_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree2_modules_bn1_parameters_bias_: + name = "L_self_modules_level5_modules_tree2_modules_bn1_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree2_modules_conv2_parameters_weight_: + name = "L_self_modules_level5_modules_tree2_modules_conv2_parameters_weight_" + shape = [512, 512, 3, 3] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.021 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree2_modules_bn2_buffers_running_mean_: + name = "L_self_modules_level5_modules_tree2_modules_bn2_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree2_modules_bn2_buffers_running_var_: + name = "L_self_modules_level5_modules_tree2_modules_bn2_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree2_modules_bn2_parameters_weight_: + name = "L_self_modules_level5_modules_tree2_modules_bn2_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_tree2_modules_bn2_parameters_bias_: + name = "L_self_modules_level5_modules_tree2_modules_bn2_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_root_modules_conv_parameters_weight_: + name = "L_self_modules_level5_modules_root_modules_conv_parameters_weight_" + shape = [512, 1280, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.063 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_root_modules_bn_buffers_running_mean_: + name = "L_self_modules_level5_modules_root_modules_bn_buffers_running_mean_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_root_modules_bn_buffers_running_var_: + name = "L_self_modules_level5_modules_root_modules_bn_buffers_running_var_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_root_modules_bn_parameters_weight_: + name = "L_self_modules_level5_modules_root_modules_bn_parameters_weight_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 1.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_level5_modules_root_modules_bn_parameters_bias_: + name = "L_self_modules_level5_modules_root_modules_bn_parameters_bias_" + shape = [512] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.000 + data = None + + +class Program_weight_tensor_meta_L_self_modules_fc_parameters_weight_: + name = "L_self_modules_fc_parameters_weight_" + shape = [1000, 512, 1, 1] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.045 + data = None + + +class Program_weight_tensor_meta_L_self_modules_fc_parameters_bias_: + name = "L_self_modules_fc_parameters_bias_" + shape = [1000] + dtype = "torch.float32" + device = "cuda:0" + mean = 0.000 + std = 0.025 + data = None